National Constitutions and EU Integration 1509906762, 9781509906765

Table of contents :
Preface
Contents
The Authors
1. Belgium
I. The Belgian Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership
IV. Constitutional Limits to EU Integration
V. The Implementation of EU Law in Belgium: Rules and Practices
VI. Conclusion: The Relation between EU Law and Belgian Law
References
2. Bulgaria
I. Introduction
II. Main Characteristics of the National Constitutional System
III. Constitutional Culture
IV. Constitutional Foundations of EU Membership
V. Constitutional Limits to EU Integration
VI. Constitutional Rules and Practice on Implementing EU Law
VII. Resulting Relationship between EU Law and National Law
References
3. Czech Republic
I. Main Characteristics of the Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership
IV. Constitutional Limits to EU Integration
V. The Constitution and the Implementation of EU Law
VI. Conclusion: Ongoing Dynamic Relationship between EU Law and the Czech Constitutional Order
References
4. Denmark
I. Introduction
II. Main Characteristics of the National Constitutional System
III. Constitutional Practice and Culture: The Parliament as the Ultimate Symbol of Democracy
IV. Constitutional Foundations of EU Membership
V. Constitutional Limits to EU Integration
VI. Constitutional Rules and Practice on Implementing EU Law
VII. Conclusion
References
5. Germany
I. Germany's Special Situation
II. Characteristics of the Constitutional System in Germany
III. Constitutional Culture in Germany
IV. The Basic Law's Openness vis-à-vis European Integration
V. European Integration and the Federal Constitutional Court
References
6. Estonia
I. Introduction: Shifts in Estonian Constitutionalism in the Light of a 'Disconnect' between National and EU Legal Discourses
II. Main Characteristics of the National Constitutional System
III. Constitutional Culture
IV. Constitutional Foundations of EU Membership and Closely Related Instruments
V. Constitutional Limits to EU Integration: Exceptionless Primacy of EU Law
VI. Constitutional Rules and the Practice of Implementing EU Law
VII. Resulting Relationship between EU Law and National Law: Perspectives Regarding Substantive Constitutional Law
VIII. Concluding Remarks: A Case for a Joined-Up Discussion on the Impact of EU law on the Member States' Constitutional Systems as Part of the Debate on the Future of Europe
References
7. Ireland
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and/or Practice Concerning the Implementation of EU Law
VI. The Resulting Relationship between European Law and National Law
References
8. Greece
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and/or Practice on Implementing EU Law
VI. Resulting Relationship between EU Law and National law
References
9. Spain
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership and Closely Related Instruments
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and Practice on Implementing EU Law
VI. Resulting Relationship between EU Law and National Law
References
10. France
I. Main Characteristics of the National Constitutional System and Constitutional Culture
II. Constitutional Foundations of EU Membership and Closely Related Instruments
III. Constitutional Limits to EU Integration
IV. Concluding Remarks
References
11. Croatia
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU-Membership
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and/or Practice on Implementing EU Law
VI. Conclusion – The Relationship between the Croatian Constitution and the EU Legal Order
References
12. Italy
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership and Closely Related Instruments
IV. Constitutional Limits to European Integration53
V. Constitutional Rules and Practice of Implementing EU Law
VI. Conclusion: Resulting Relationship between EU Law and National Law
References
13. Cyprus
I. Introduction
II. Main Characteristics of the RoC’s Constitutional System
III. Constitutional Culture
IV. Constitutional Foundations of EU Membership
V. Constitutional Limits to EU Integration
VI. Constitutional Rules and/or Practice on Implementing EU Law
VII. Resulting Relationship between EU Law and National Law
VIII. Conclusion
References
14. Latvia
I. Introduction
II. Main Characteristics of the National Constitutional System
III. Constitutional Culture
IV. Constitutional Foundations of EU Membership
V. Constitutional Limits to EU Integration
VI. Constitutional Rules and Practices on Implementing EU Law
VII. Relationship between EU Law and National Law
VIII. Conclusions
References
15. Lithuania
I. Main Characteristics of the Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership and Closely Related Instruments
IV. Constitutional Limits to EU Integration
V. Constitutional and Sub-Constitutional Rules on Implementing EU Law
VI. Resulting Relationship between EU Law and National Law
References
16. Luxembourg
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership and Closely Related Instruments
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and/or Practice on Implementing EU Law
VI. Relationship between EU Law and National Law
VII. Miscellaneous: The Current Constitutional Amendment Procedure
References
17. Hungary
I. Main Characteristics of the National Constitutional System, Including the System of the Judiciary
II. Constitutional Culture
III. Constitutional Foundations of EU Membership and Closely Related Instruments
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and Practice on Implementing EU Law
VI. Resulting Relationship between EU Law and National Law
References
18. Malta
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. The Constitutional Foundations of EU-Membership
IV. Constitutional Limits to EU-integration
V. Constitutional Rules and/or Practice on Implementing EU Law
VI. Resulting Relationship between EU Law and National Law
References
19. The Netherlands
I. Main Characteristics of the Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership and Closely Related Instruments
IV. Constitutional Limits to European Integration?
V. Constitutional Rules and/or Practice on Implementing EU Law
VI. Resulting Relationship between EU Law and National Law
VII. Conclusion
References
20. Austria
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership and Closely Related Instruments
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and Practice on Implementing EU Law
VI. Concluding Remarks
References
21. Poland
I. Main Characteristics of the National Constitutional System
II. Constitutional Foundations of EU Membership and Closely Related Instruments
III. Constitutional Limits to EU Integration
IV. Constitutional Rules and/or Practice on Implementing EU Law
V. The Resulting Relationship between EU Law and National Law
References
22. Portugal
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Integration
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and Practice on Implementing EU Law
VI. Relationship between EU Law and National Law
VII. Concluding Remarks
References
23. Romania
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and Practice on Implementing EU Law
VI. Resulting Relationship between EU Law and National Law
References
24. Slovenia
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership and Closely Related Instruments
IV. Constitutional Limits to EU-Integration
V. Constitutional Rules and/or Practice on Implementing EU Law
VI. Resulting Relationship between EU Law and National Law
References
25. Slovakia
I. Main Characteristics of the Slovak Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU Membership and Closely Related Instruments
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and Practice on Implementing EU Law
VI. Resulting Relationship of EU Law and National Law
References
26. Finland
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundations of EU-Membership
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and/or Practice in Implementing EU Law
VI. Resulting Relationship between EU Law and National Law
VII. Concluding Remarks: From Constitutional Outsider to Insider
References
27. Sweden
I. Main Characteristics of the National Constitutional System
II. Constitutional Culture
III. Constitutional Foundation of EU Membership and Closely Related Instruments
IV. Constitutional Limits to EU Integration
V. Constitutional Rules and/or Practice on Implementing EU Law
VI. Resulting Relationship between EU Law and National Law
References
28. United Kingdom
I. UK Constitution: Principal Characteristics
II. UK Constitution: Underlying Culture
III. EU Membership: UK Constitutional Foundations
IV. Constitutional Limits on EU Membership
V. Implementing EU Law: UK Constitutional Rules/Practice
VI. UK and EU Law: Resulting Relationship
References
29. Squaring the Circle – How National Constitutions and EU Law Interact
I. Introductory Remarks
II. National Constitutional Systems and Cultures
III. Constitutional Foundations of EU Membership
IV. Constitutional Obstacles to Further EU integration
V. The EU Law Perspective – Primacy Revisited
VI. Doctrinal Solutions to the Potential Clash of Constitutions?
VII. A Modest Conclusion: Cooperation is a Common Constitutional Principle
References
Index

Citation preview

NATIONAL CONSTITUTIONS AND EU INTEGRATION Do individual constitutions, and the legal cultures underlying them, pose an obstacle to future EU integration? This ambitious collection brings together reports from all the European Member States, systematically setting out their individual constitutional guarantees. In doing so, it tracks possible roadblocks to the future evolution of European integration. Written by recognised authorities in each Member State, it offers an authoritative and rigorous overview of the European Union’s constitutional landscape. Its single-structure approach allows for comparison while maintaining consistency. It will become the standard reference work for academics, students and practitioners in the field of European Union law and integration.

ii

National Constitutions and EU Integration Edited by

Stefan Griller Lina Papadopoulou and

Roman Puff

HART PUBLISHING Bloomsbury Publishing Plc Kemp House, Chawley Park, Cumnor Hill, Oxford, OX2 9PH, UK 1385 Broadway, New York, NY 10018, USA 29 Earlsfort Terrace, Dublin 2, Ireland HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2022 Copyright © The editors and contributors severally 2022 The editors and contributors 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/, 1998–2022. A catalogue record for this book is available from the British Library. A catalogue record for this book is available from the Library of Congress. ISBN: HB: 978-1-50990-676-5 ePDF: 978-1-50990-674-1 ePub: 978-1-50990-675-8 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 With the ‘EU Treaty reform’ saga having come to an at least preliminary conclusion by the end of 2009, the European Parliament’s Policy Department for ‘Citizens’ Rights and Constitutional Affairs’ issued an invitation to tender for a study on ‘National constitutional law and European integration’, aimed at exploring possible future paths for further EU integration beyond the Lisbon Treaty.1 The tender of a team of academics from the Vienna University of Economics and Business – some of whom are found in this volume – was finally accepted. It resulted in a study covering the Czech Republic, Germany, France, Italy, Austria, Poland, Finland and Sweden.2 While working on this project, some members of the research team started to feel that, despite its achievement, the study’s deficiencies called for a follow-up in two main aspects. First and foremost, covering all EU Member States might provide great added value to the results of the project. Second, the multifaceted and numerous developments in many EU Member States nowadays makes it almost compelling to build on the expertise – including political and cultural background – of participants rooted in each Member State to present a complete picture. It is not only the language barrier, but also the intricate topics themselves and their delicate systemic dimension, that are best dealt with by specialists from the respective countries. In this situation, Stefan Griller and his crew decided to build a team of academics and legal practitioners from all EU Member States and to develop a structure that fits all of their constitutional orders. The aim was to provide for an organisational framework that ensures that the borders between the European constitutional traditions as well as between academic disciplines – mainly constitutional lawyers, of course, but also political scientists and historians – can be overcome by debate. Thereby, the study undertaken for the European Parliament could be further developed and improved. What followed from this decision – which was either heroic or megalomaniac, according to your preferences – turned out to become a ‘National Constitutions and EU Integration’ saga, the results of which are finally bestowed upon the public with this book. Although probably one may detect a pessimistic undertone in this sentence, working on the project was amazingly intellectually inspiring and resulted in a close cooperation of experts comprising academics, judges, and other practitioners covering the entire Union. Four workshops – in Vienna in 2012 and 2013, in Maastricht in 2014, and in Thessaloniki in 2015 – in which the relevant issues from all 28 (at the time) EU Member States underwent unusually thorough discussions, forged a unique, informal, yet stable network of specialists on constitutional issues of European integration that is of continuing value. Working on the project created a peculiar fascination, at least for the editors of this volume, if not for most participants. Discussing protracted constitutional issues from the angles of 28 diverse nations, from various professional backgrounds including academics from various disciplines and traditions, judges, or public servants, became an example of ‘united in diversity’. 1 Invitation to tender no IP/C/AFCO/IC/2009-73. 2 European Parliament, DG for Internal Policies (ed), National Constitutional Law and European Integration (Brussels, European Parliament, 2011).

vi  Preface Perhaps the most striking insight was the broad spectrum of motives for EU integration offered in the Member States’ constitutions – namely peace, security, democracy, and cooperation. Surprisingly, for a union that started out as an ‘Economic Community’, economic effects did not figure explicitly at the forefront. What is noteworthy, too, is the predominant willingness of most Member States to engage further in integration, notwithstanding its practical, day-to-day difficulties. The national reports follow a common structure to allow for comparison. It features a brief outline of the main characteristics of the national constitutional system including a sketch of the prevailing constitutional culture that is followed by more detailed analyses of the constitutional foundations of EU membership and the possible constitutional limits to (further) EU integration. Some reports then also dwell on the constitutional rules for and the legal practice of implementing EU law at the national level. All reports conclude with a final summary of the relationship between EU law and national law. Almost all national reports explicitly follow this structure. The few that differ slightly in form do it at least in their substance. Equally, all teams of authors consist of experts rooted in the respective Member States and their legal traditions, cultures and languages – either by their origin or their working environment. It is a requirement of fairness to reveal also those points where we failed. First and foremost, against the backdrop of continuing constitutional developments in practically each and every EU Member State, we can only offer a constitutional snapshot. Constant change is part of living legal orders. Second, also overarching developments like Brexit came at a price – now that this book is available to the public, the former Union of 28 has become one of 27 and a very peculiar, important constitutional tradition is missing from its ranks – although, we are proud to say, not from this book! Ambitious endeavours leave issues unsettled. In our case, further deepening would certainly be desirable on substance of what might be, in the field of constitutional openness of the Member States’ legal orders vis-à-vis EU law, the ‘constitutional traditions common to the Member States’, that ‘shall constitute general principles of the Union’s law’.3 In times where even the common values of the Member States (Article 2 TEU) are getting more and more controversial, this is a task for future research. Finally, we are very thankful to many friends and colleagues who helped in creating and finishing this project. In organisational terms, the ‘European Community Studies Association Austria – ECSA Austria’ and the Jean Monnet Programme of the European Commission as well as the University of Salzburg shall be mentioned in the first place, not only, but also not least for financial support. We also have to pay thanks to Monica Claes and her team from the University of Maastricht in the Netherlands, for hosting the third project workshop in 2014, and the team of the Aristotle University in Thessaloniki, Greece, for doing the same for the workshop in 2015. Sinead Moloney and her colleagues at Hart Publishers in Oxford have earned our sincere thanks not only for their invaluable professional help in producing this volume, but also for their unwavering patience with our numerous shortcomings as editors. And finally, special thanks go to all authors of the country reports and their collaborators, mentioned or not, who helped them setting up their accounts – it was their willingness to co-operate, their engagement in the project, and their contributions that made the project ‘National Constitutions and EU Integration’ develop into a unique forum of discussion over its lifetime! Stefan Griller, Lina Papadopoulou and Roman Puff Salzburg and Thessaloniki, September 2021

3 Art

6(3) TEU.

CONTENTS Preface���������������������������������������������������������������������������������������������������������������������������������������������������� v The Authors������������������������������������������������������������������������������������������������������������������������������������������ ix 1. Belgium������������������������������������������������������������������������������������������������������������������������������������������� 1 Willem Verrijdt 2. Bulgaria���������������������������������������������������������������������������������������������������������������������������������������� 33 Martin Belov 3. Czech Republic����������������������������������������������������������������������������������������������������������������������������� 59 Luboš Tichý and Tomáš Dumbrovský 4. Denmark�������������������������������������������������������������������������������������������������������������������������������������� 89 Marlene Wind 5. Germany������������������������������������������������������������������������������������������������������������������������������������� 123 Matthias Jestaedt 6. Estonia���������������������������������������������������������������������������������������������������������������������������������������� 171 Hent Kalmo and Anneli Albi 7. Ireland���������������������������������������������������������������������������������������������������������������������������������������� 211 Gavin Barrett 8. Greece����������������������������������������������������������������������������������������������������������������������������������������� 247 Lina Papadopoulou 9. Spain������������������������������������������������������������������������������������������������������������������������������������������� 283 Aida Torres Pérez and Maribel González Pascual 10. France����������������������������������������������������������������������������������������������������������������������������������������� 305 Rainer Palmstorfer, Laurent Sermet and Erich Vranes 11. Croatia���������������������������������������������������������������������������������������������������������������������������������������� 333 Tamara Ćapeta 12. Italy��������������������������������������������������������������������������������������������������������������������������������������������� 357 Marta Cartabia and Barbara Guastaferro 13. Cyprus���������������������������������������������������������������������������������������������������������������������������������������� 385 Nikolas Kyriakou and Nikos Skoutaris

viii  Contents 14. Latvia������������������������������������������������������������������������������������������������������������������������������������������ 401 Martins Mits 15. Lithuania������������������������������������������������������������������������������������������������������������������������������������ 427 Egidijus Kūris 16. Luxembourg������������������������������������������������������������������������������������������������������������������������������� 459 Jörg Gerkrath 17. Hungary�������������������������������������������������������������������������������������������������������������������������������������� 483 Renáta Uitz 18. Malta������������������������������������������������������������������������������������������������������������������������������������������ 507 Peter G Xuereb and Mireille M Caruana 19. The Netherlands������������������������������������������������������������������������������������������������������������������������� 533 Monica Claes 20. Austria���������������������������������������������������������������������������������������������������������������������������������������� 553 Stefan Griller 21. Poland����������������������������������������������������������������������������������������������������������������������������������������� 585 Mirosław Wyrzykowski and Michał Ziółkowski 22. Portugal�������������������������������������������������������������������������������������������������������������������������������������� 627 Ana Maria Guerra Martins 23. Romania������������������������������������������������������������������������������������������������������������������������������������� 655 Mihaela Vrabie 24. Slovenia�������������������������������������������������������������������������������������������������������������������������������������� 679 Matej Avbelj 25. Slovakia�������������������������������������������������������������������������������������������������������������������������������������� 699 Zuzana Vikarská and Kamil Baraník 26. Finland��������������������������������������������������������������������������������������������������������������������������������������� 727 Tuomas Ojanen 27. Sweden���������������������������������������������������������������������������������������������������������������������������������������� 749 Joakim Nergelius 28. United Kingdom������������������������������������������������������������������������������������������������������������������������� 777 Paul Craig 29. Squaring the Circle – How National Constitutions and EU Law Interact������������������������������ 795 Stefan Griller and Lina Papadopoulou Index�������������������������������������������������������������������������������������������������������������������������������������������������� 843

THE AUTHORS Anneli Albi is Professor of European and Comparative Constitutional Law at the University of Kent, Canterbury, UK. Matej Avbelj is Professor of European Law at the Graduate School of Government and European Studies at Kranj, Slovenia. Kamil Baraník is Associate Professor of Constitutional Law at the University of Matej Bel Faculty of Law, Slovakia and Legal Advisor at the Slovak Constitutional Court. Gavin Barrett is Professor of Law and Head of Teaching and Learning at the Sutherland School of Law, University College Dublin, Ireland. Martin Belov is Professor in Constitutional and Comparative Constitutional Law at the University St Kliment Ohridski of Sofia, Bulgaria, Faculty of Law and Vice Dean. Tamara Ćapeta is Professor of EU law, Faculty of Law in Zagreb, Croatia and since October 2021 Advocate General at the Court of Justice of the European Union, Luxembourg. Marta Cartabia, is Constitutional Law Professor at Bocconi University, Milan, was President of the Corte Costituzianale in Rome and since February 2021 is the Italian Minister of Justice. Mireille Caruana is Lecturer in EU and Information Technology Law at the Faculty of Laws, University of Malta. Monica Claes is Professor of European and Comparative Constitutional Law at the Faculty of Law of Maastricht University, the Netherlands. Paul Craig is Emeritus Professor of English Law and Fellow of St John’s College at the University of Oxford, United Kingdom. Tomáš Dumbrovský is Senior Scientist at the Faculty of Law, Charles University, Prague, Czech Republic and Visiting Professor at the Doha Institute for Graduate Studies, Qatar. Jörg Gerkrath is Professor for European and Public law at the University of Luxembourg. Maribel González Pascual is Adviser to the Minister of Territorial Policy and Public Administration of the Spanish Government, Madrid, Spain. Stefan Griller at the time of writing was Professor for European Law and Director of the Salzburg Centre of EU Studies – SCEUS at the University of Salzburg, Austria. Barbara Guastaferro is Associate Professor of Constitutional Law at the University Federico II of Naples, Italy.

x  The Authors Ana Maria Guerra Martins is Professor at the Law School of the University of Lisbon, and since April 2020 serves as Judge at the European Court of Human Rights, Strasbourg, France. Matthias Jestaedt is Professor for Constitutional Law and Legal Theory at the Law Faculty of the University of Freiburg/Br, Germany. Hent Kalmo is Researcher of Legal and Political Theory at the Johan Skytte Institute for Political Studies of the University of Tartu, Estonia. Egidijus Kūris is the Lithuanian Judge at the ECtHR, Strasbourg, France, former President of the Constitutional Court of Lithuania and Professor at the Law Faculty of Vilnius University. Nikolas Kyriakou is Senior Agent – Lawyer in commercial matters at Europol in The Hague, the Netherlands. Martins Mits is the Latvian Judge at the European Court of Human Rights, Strasbourg, France. Joakim Nergelius is Professor for Constitutional Law at Örebro University, Sweden. Tuomas Ojanen is Professor of Constitutional Law at the University of Helsinki, Finland. Rainer Palmstorfer is Professor for European Law at the University of Linz, Austria. Lina Papadopoulou is Associate Professor of Constitutional Law at the Law Faculty of the Aristotle University, Thessaloniki, Greece. Roman Puff is historian and Senior Scientist at the University of Salzburg, Austria, and Executive Director of the Salzburg Centre of European Union Studies – SCEUS. Laurent Sermet is Professor of European Law at the Institut des Sciences Politiques – Sciences Po, Aix-en-Provence, France. Nikos Skoutaris is Associate Professor in EU Law at the University of East Anglia School of Law, Norwich, United Kingdom. Luboš Tichý is Professor of European, International Private and Civil Law and Head of the Centre for Comparative Law of Charles University, Prague, Czech Republic. Aida Torres Pérez is Associate Professor of Constitutional Law and ICREA Research Professor at Pompeu Fabra University, Barcelona, Spain. Renáta Uitz is Professor of Comparative Constitutional Law at the Central European University, now in Vienna, Austria. Willem Verrijdt is a judge in the Belgian Constitutional Court and a clinic teacher at the Catholic University Leuven. Zuzana Vikarska is Assistant Professor in Constitutional Law at the Masaryk University and Judicial Law Clerk at the Czech Constitutional Court, both in Brno, Czech Republic. Mihaela Vrabie is Associated Lecturer in European Union Law at the Faculty of Law, University of Bucharest and Lawyer, member of the Bucharest Bar, Romania. Erich Vranes is Professor of European, Public, Public International and International Economic Law at the Vienna University of Economics and Business, Austria.

The Authors   xi Marlene Wind is Professor and the Director of the Centre for European Politics at the Department of Political Science of the University of Copenhagen, Denmark. Mirosław Wyrzykowski was Professor of Constitutional Law at the University of Warsaw, Poland and completed a term as Judge at the Polish Constitutional Tribunal in 2010. Peter G Xuereb is Professor of European Law at the University of Malta and since 2018 serves as Judge at the European Court of Justice in Luxembourg. Michał Ziółkowski is Assistant Professor in Constitutional Law at the Kozminski University in Warsaw, Poland, and Max Weber Fellow at the EUI in Florence, Italy (2019–21).

xii

1 Belgium WILLEM VERRIJDT

I.  The Belgian Constitutional System A.  The Emergence of Belgium and its Ongoing Process of Federalisation After being ruled by Burgundy (1433–77), Spain (1477–1713), Austria (1713–95), and France (1795–1815), the current Belgian territory was assigned to the United Kingdom of the Netherlands at the Vienna Congress. The Dutch King Willem I’s Decree Government1 and the restrictions he imposed on the freedoms of religion, education, press and language, led to a ‘Monster Coalition’ of Catholics and Liberals, which started the Belgian revolution and brought about Belgium’s independence in 1830. Proclaimed on 7 February 1831, the Belgian Constitution was a very modern one, influencing many later constitutions, and currently standing as one of the oldest still in force.2 According to its Article 195, amending the Constitution is a complex three-stage procedure. The Preconstituent Assembly, ie a simple majority in both the Chamber of Representatives and the Senate, decides which constitutional provisions are open for amendment. That decision automatically leads to the dissolution of both Houses of Parliament and the organisation of elections within 40 days. The newly elected Houses are the Constituent Assembly, which may only amend, by a twothirds majority, the provisions opened for amendment by the Preconstituent Assembly. Both the Preconstituent and the Constituent phases require the King’s consent. This rigid procedure explains why, until 1970, the Constitution was a very stable document, which had only been amended in 1893 and in 1921 in order to democratise the election process. In 1831, the Belgian state structure was rather straightforward. Belgium was a unitary, albeit decentralised, state, with one legislature, composed of a Chamber of Representatives and a Senate, one executive, led by the King, and one judiciary, with a single supreme court, the Cour de cassation. However, after six rounds of ‘state reform’ (1970, 1980, 1988, 1993, 2001 and 2014) a complex sui generis federal state, composed of three Communities and three Regions, was established. This ongoing federalisation process started because of Flemish demands for linguistic 1 He claimed the residuary powers in order to adopt the entire socio-economic policy through royal decrees, he prohibited judicial review of these royal decrees and he decided autocratically upon conflicts of competence. 2 André Alen, David Haljan and Willem Verrijdt, Constitutional Law in Belgium (Alphen aan den Rijn, Wolters Kluwer, 2020) 28 ff. The Belgian Constitution has been referred to as ‘the first lasting document of European constitutionalism’ (ECtHR 16.06 2015 App no 64569/09 [Delfi AS v Estonia] ECLI:CE:ECHR:2015:0616JUD006456909).

2  Willem Verrijdt and cultural autonomy and Walloon demands for economic autonomy. The process was further fuelled by student protests in Leuven in 1968. Belgian federalism is territorial. The principle of territoriality (Article 4 of the Constitution) originated in the Flemish demand to be governed by local authorities in their own language.3 This required dividing the country into four language zones: three unilingual ones (the Dutch language zone in the north, the French language zone in the south, and the German language zone in the east) and the bilingual Brussels-Capital language zone (Dutch and French). Although the principle of territoriality is one of the most fundamental rules in Belgian constitutional law, it is mitigated for 27 municipalities near the linguistic borders. While these municipalities are located in unilingual zones, a significant minority (in some cases even a majority) belonging to another language group has the right to be addressed by the local Government in their own language. When the Communities and the Regions were created, they were attached to this territorial blueprint. The first reform of state (1970) gave birth to the three Communities: the Flemish Community consists of the Dutch language zone and the Dutch-speaking institutions4 in the Brussels-Capital language zone, the French Community consists of the French language zone and the French-speaking institutions in the Brussels-Capital language zone, and the Germanspeaking Community coincides with the German language zone. Currently, the main Community competences are education, culture, health policy, aid to persons, and the use of languages. These attributed competences, however, are subject to significant exceptions, such as social security, public health, etc, which remain federal competences.5 The second reform of state (1980) saw the creation of the three Regions. The Flemish Region coincides with the Dutch language zone, the Brussels-Capital Region coincides with the bilingual language zone Brussels-Capital, and the Walloon Region comprises both the French and the German language zones.6 The regional competences, laid down in a Special Majority Act,7 currently include spatial planning, the protection of the environment, housing, agriculture, economy, energy, municipalities and provinces, employment and infrastructure projects. Again, significant federal exceptions limit these attributed regional competences.8 The third reform of state (1988–89) mainly concerned the complex functioning of the Brussels-Capital Region and the financing of the Communities and the Regions. The federal structure was completed by the fourth reform of state (1993–94). After this reform, a renumbering of the Constitution was necessary. The fifth reform of state (2001), which was the only one adopted without amending the Constitution, mainly concerned the municipalities and provinces.

3 During the 19th century, many civil servants, even in Flanders, were French-speaking, and invoked their own linguistic freedom to answer Flemish citizens in French. 4 Only the institutions in Brussels are concerned, as there is no sub-nationality within the Belgian state. This implies that the inhabitants of Brussels may not be divided into ‘Flemish’ and ‘French-speaking’ sub-nationals. 5 Arts 127–30 of the Constitution, further specified by Arts 4 and 5 of the Special Majority Act of 8 August 1980 on the Reform of the Institutions and Arts 4 and 5 of the Act of 31 December 1983 on the Reform of the Institutions of the German-speaking Community. 6 Almost 58% of the population live in the Flemish Region, a little over 32% live in the Walloon Region and 10% live in the Brussels-Capital Region, the vast majority of whom are French-speaking. The German-speaking Community consists of nine municipalities which became part of Belgium after the First World War and only account for some 0.7% of the population. 7 Special Majority Acts can only be adopted for matters explicitly listed in the Constitution. They require a two-third majority in both Houses of Parliament, including a simple majority in both language groups within these bodies (Art 4, in fine, of the Constitution). 8 Art 6 of the Special Majority Act of 8 August 1980 on the Reform of the Institutions.

Belgium  3 The sixth reform of state (2012–14)9 has been the largest so far from the point of view of competence transfers and financing.10 It also encompassed the resolution of several politically sensitive issues,11 the adoption of some new techniques of cooperative federalism, a reform of the Senate and the legislative procedure, new techniques for the exercise of competences and the financing of the Brussels-Capital Region, the enhancement of constitutive autonomy of the federated entities, and minor reforms of the monarchy and the Constitutional Court.

B.  Main Characteristics of Belgian Constitutional Law Even after these reforms, the Belgian Constitution still clearly reflects its origins. It was drafted by a group of young liberal lawyers and was, subsequently, strongly influenced by the Catholics, thus showing both liberal and religious features. The authors’ background also explains why many of its provisions should be regarded as direct reactions against King Willem I’s Government: it institutionalises mistrust in the Executive, it contains a vast catalogue of human rights, among which are those Willem I suppressed the most, and it protects the judiciary’s independence, including its competence to review executive acts against higher norms. The separation of powers is guaranteed by several constitutional provisions, which also establish a set of checks and balances between these powers. The King is involved in the legislative procedure. His consent, as a branch of the legislature, is necessary for adopting legislation (‘approval’) and his initiative, as the head of the Executive, is required for an act’s execution (‘promulgation’). The King Himself cannot be held democratically accountable and therefore, a Minister’s signature is needed for all his actions. Only the Government as a whole is accountable to the Parliament. The King’s competence to dissolve the Federal Chamber of Representatives is strictly limited. In the 1831 Constitution, the elected Legislature outweighed both other powers: it was vested with the residuary powers, ie the powers not expressly assigned to another organ, it was granted the exclusive competence to provide authoritative interpretations of statutory law binding the judiciary, and judicial review of legislation was, at that time, non-existent. Meanwhile, the executive has become the true motor of state policy. This is because the Council of Ministers discusses all significant policy issues and prepares most of the draft legislation, which is then adopted by a majority in Parliament, ie by the political parties represented in the Government. The federal legislature preserved its bicameral nature, but since the sixth reform of state, the Senate’s involvement in the legislative process is very limited ratione materiae. All Communities and Regions possess their own parliaments, though the parliaments of the Flemish Community and of the Flemish Region have merged. In the Brussels-Capital Region, several legislative powers in community matters have been assigned to the Commission communautaire française and the Commission communautaire commune.12 All parliaments are elected for a five-year term through a system of mandatory universal suffrage and proportional representation. 9 See Jurgen Goossens and Pieter Cannoot, ‘Belgian Federalism after the Sixth State Reform’, 7 Perspectives on Federalism (2015) no 2, 29; Patrick Peeters and David Haljan, ‘Belgium’s Sixth State Reform: The State of the Nation(s)’, 22 European Public Law (2016) 411. 10 The new competences of the communities and the regions, all coming with several exceptions, are situated in the fields of employment, health policy, child benefit, traffic, expropriation, the environment and energy. 11 Eg the scission of the bilingual electoral and judicial district Brussels-Hal-Vilvorde, and the judicial protection of the inhabitants of the 27 municipalities with special language arrangements. 12 The CCF is composed of the 72 French-speaking members of the Brussels Regional Parliament and the CCC is composed of all members of the Brussels Regional Parliament.

4  Willem Verrijdt Federal Acts of Parliament (‘laws’) have the same legal value as the Acts of Parliament of the Communities and the Regions (‘decrees’) and the Acts of Parliament of the Brussels-Capital Parliament (‘ordinances’). The distribution of legislative competences between the federal level and the federated entities is largely based on mutual exclusivity. Belgium is a constitutional monarchy, with a hereditary King or Queen as its head of state. The King’s competences are limited to those expressly attributed to him by the Constitution or an Act of Parliament (Article 105 of the Constitution). One of these attributed competences allows him to take all necessary measures for the execution of legislation (Article 108 of the Constitution). Although its active role in the formation of the Government has diminished due to the increasing power of political parties, he still acts as a mediator behind the scenes in times of political crisis. The King is a federal institution. At the federated level, the executive role is fulfilled by the federated governments. These governments, as collegial bodies, approve and promulgate decrees or ordinances, and take the measures required for the execution of these legislative acts. They only possess attributed competences, which are to be exercised through consensus. The judiciary has remained a federal institution. It interprets and applies both federal and federated legislation and administrative acts. Parties usually have a right to higher appeal and a right to a cassation appeal, the Cour de cassation having jurisdiction to quash a judgment if it violates any written or unwritten norm. Article 159 of the Constitution obliges all judges to disapply administrative acts if they violate higher norms. By contrast, the Constitution does not grant them any jurisdiction to review Acts of Parliament against higher norms. Nevertheless, the Cour de cassation has decided that all judges may examine whether an Act of Parliament applicable in a case they are adjudicating violates self-executing international or supranational law. If they find a violation, they are to disapply that Act of Parliament in the case at hand, but they cannot annul it.13 This competence does not extend to reviewing Acts of Parliament against the Constitution, which is an exclusive competence of the Constitutional Court. Administrative jurisdiction has existed since 1946, when the Conseil d’Etat was established. Apart from its Administrative Litigation Department, which may annul administrative decisions violating higher norms, the Conseil d’Etat also has a Legislation Department, which renders nonbinding legal opinions on some draft legislation and draft executive decisions. The Conseil d’Etat’s Administrative Litigation Department has long been the only administrative jurisdiction. But since the turn of the century, the federal and federated legislators increasingly create specialised administrative jurisdictions with limited competences, such as building permits, asylum decisions, competition, education, etc. The decisions of most of these administrative jurisdictions are subject to a cassation review by the Conseil d’Etat. The Constitutional Court14 was established in 1984 and rendered its first judgment in 1985. Only then did the inviolability of Acts of Parliament vis-à-vis the Constitution come to an end.15 Until 1989, the Constitutional Court could only examine whether legislation was enacted in compliance with the distribution of legislative competences between the federal level and the federated entities. In 1989, the principle of equality and the freedom of education were added to its reference norms. Using the principle of equality in Articles 10 and 11 of the Constitution, the Court subsequently established its own competence to review whether legislation violates, in a discriminatory manner, other provisions of the Constitution, general principles of law, 13 Cass., 27.05.1971, Pas. [= Pasicrisie Belge, the collection of judgments of the Cour de Cassation], I, 1971, 836 (Fromagerie Franco-Suisse Le Ski) ECLI:BE:CASS:1971:ARR.19710527.16. 14 Art 142 of the Constitution and the Special Majority Act of 6 January 1989 on the Constitutional Court. See Géraldine Rosoux, Contentieux constitutionnel (Brussels, Larcier, 2021). 15 Alen/Haljan/Verrijdt, Constitutional Law (fn 2) 130 ff.

Belgium  5 and even EU and international law (see fn 140). In 2003, all human rights in the Constitution (Articles 8–32, 170, 172 and 191) became direct reference norms. In 2014, the Court’s competences were extended with addition of the principle of federal loyalty (‘Bundestreue’) as a basis for review, with the power to exercise preliminary review of the admissibility of regional non-binding referenda, and with a competence to review sanctions imposed by a parliamentary committee on MPs for exceeding the campaign expenses limit. The Court is composed of 12 judges, half of whom are Dutch-speaking and half of whom are French-speaking. Also, half of the judges are former MPs, whereas the other half consists of academics, former magistrates in the Cour de cassation or the Conseil d’Etat and former law clerks in the Constitutional Court. Both genders must be represented by at least four judges each. Also, the Court’s composition proportionally reflects the political parties’ strength in the Chamber of Representatives. The judges are appointed by the King after being nominated by the Chamber of Representatives or by the Senate, and they remain in office until the age of 70. Each language group chooses its own president. They alternate as the Court’s acting president, the pivotal date being 1 September of each year. If a deliberation ends in a 6-6 standoff, the Court’s acting president has a casting vote, but is not obliged to use it, and can also choose to continue the deliberation. All legislators, governments and individuals (including companies and associations) can address the Court with an action for annulment against new legislation within six months of its publication in the Moniteur belge. If the Court finds a violation, the unconstitutional norm is, as a general rule, annulled ex tunc and erga omnes. Furthermore, all ordinary and administrative judges are competent – and, at a party’s request, often even obliged – to refer a pending case to the Constitutional Court for a preliminary ruling on the constitutionality of an applicable legislative norm. If the Court finds a violation, the unconstitutional provision is not annulled but must be disapplied by the judges.

II.  Constitutional Culture The 1831 Constitution sets out the framework and the principal functions of the state’s organs and lists the principles according to which they must operate. It also enshrines the basic tenets of judicial protection of the citizens from all branches of government, including an enumeration of fundamental rights. Because of its aforementioned rigid amendment procedure, several techniques have been developed to adapt the Constitution to present-day conditions. These techniques can be textual, such as the ‘implicit amendment’, ie amending a constitutional provision which was opened for amendment, and thereby implicitly modifying a constitutional provision which was not opened for amendment.16 ‘De-constitutionalisation’ is another textual adaptive technique: the content of many constitutional provisions is to be further developed by Special Majority Acts and ordinary legislation. Meanwhile, these complementing norms encompass important aspects of the Belgian state structure, such as the legislative competences of the Communities and the Regions as well as their financing, the Constitutional Court’s competences and functioning, and the composition and functioning of Member State organs. These Special Majority Acts and legislative acts may 16 A notable example is Art 129, § 1, of the Constitution, inserted in 1970, which allows the Communities to regulate the use of languages in the relations between employers and employees. This provision implicitly amended Art 30 of the Constitution, which guarantees the freedom of language, adding that the use of languages may only be regulated for executive and judicial matters.

6  Willem Verrijdt not run counter to the Constitution, but they can be amended without reverting to the complex constitutional amendment procedure. The most important adaptive techniques, however, are non-textual, such as the ­evolutionary17 and teleological18 interpretation and the emergence of unwritten constitutional principles and constitutional custom19 which complement the Constitution, allowing for its application to developments not foreseen by its drafters in 1831. Modern international texts, such as the ECHR (hereinafter: the Convention), are sometimes used as a source for implicitly updating the interpretation of Belgian constitutional provisions. By contrast, comparison with other countries’ constitutions is not used as an interpretation technique. Whereas old constitutional provisions are formulated vaguely enough to allow such an evolutionary interpretation, more recent constitutional provisions, often codifying political agreements in sensitive matters, are more detailed and, therefore, do not offer the same degree of flexibility.20 All judicial and administrative judges consider upholding the rule of law an essential part of their job, and they have therefore always made ample use of their aforementioned competences to review administrative and legislative norms against higher norms. By contrast, the constitutionality of legislation has long been a mere academic interest or, at best, just a topic of political debate. When constitutionality review of formal legislation was finally established in 1984, it soon flourished. In recent years, new legislation is challenged almost systematically before the Constitutional Court, often by associations defending the interests of specific categories of persons, and often receiving significant media coverage.21 Parliamentary committees only play a minor role in upholding the Constitution. For example, the Parliamentary Committee on the Evaluation of Legislation, established in 2007, which is competent to suggest legislative amendments of legislation declared unconstitutional by the Constitutional Court, last convened in April 2014.

III.  Constitutional Foundations of EU Membership Until 1970, Belgium’s accessions to the European Communities (EC), the Council of Europe, the United Nations, and other international organisations lacked any constitutional basis, and 17 Eg the aforementioned Art 105 of the Constitution, which was at first interpreted as strictly limiting the King’s powers, but was later used in order to grant him, by a formal Act of Parliament, the competence to regulate major parts of socio-economic policy through Royal Decisions. 18 Ie interpreting constitutional provisions in light of the general purposes of the Constitution’s drafters, even if these purposes or principles are not explicitly mentioned in the Constitution (eg the rule of law, democracy and the separation of powers). Likewise, the words used in the Constitution may be interpreted in light of technological evolutions: hence, the ‘printing press’ in its Art 25 also includes written opinions on the internet (ECLI:BE:CASS:2012:ARR.20120306.5). 19 The functioning of the Council of Ministers, the role of the Prime Minister, the formation of the federal Government and the opening and closure of the parliamentary year are governed primarily by customary law. 20 Alen/Haljan/Verrijdt, Constitutional Law (fn 2) 52. 21 Eg Belgian Constitutional Court (hereafter BECC) No. 145/2012, ECLI:BE:GHCC:2012:ARR.20121206.3 (the burqa ban in public places is constitutional); BECC No 144/2013, ECLI:BE:GHCC:2013:ARR.20131107.1 (annulment of a Flemish decree reserving plots of land to people with a historical connection with the region); BECC No 153/2013, ECLI:BE:GHCC:2015:ARR.20151029.13 (euthanasia for minors under certain circumstances is constitutional); BECC No 150/2017, ECLI:BE:GHCC:2017:ARR.20171221.6 (annulment of a legislative reform downsizing the competence of the jury in criminal cases); BECC No 16/2018, ECLI:BE:GHCC:2018:ARR.20180207.12 (withdrawal of Belgian nationality in case of terrorist activity is constitutional if the subject also has another nationality); BECC No 99/2019, ECLI:BE:GHCC:2019:ARR.20190619.14 (the new Act improving transgender rights violates the principle of equality because it doesn’t go far enough); BECC No 34/2020, ECLI:BE:GHCC:2020:ARR.20200305.1 (the Act prolonging the lifespan of nuclear power plants is unconstitutional because of a lack of environmental impact assessment); BECC No 81/2020, ECLI:BE:GHCC:2020:ARR.20200604.18 (a Decree forbidding wearing religious signs in public schools

Belgium  7 were even considered to be in violation of the principle of national sovereignty laid down in Article 33 of the Constitution. Although the Conseil d’Etat’s Legislation Department often noted this problem,22 such treaties were always approved and ratified without first passing the necessary constitutional amendments. This problem was solved by introducing the current Article 34 of the Constitution.

A.  The Principle of National Sovereignty The drafters of the Belgian Constitution laid down Sieyès’ theory of national sovereignty23 in Article 33 of the Constitution, which reads: ‘All powers emanate from the Nation. These powers are exercised in the manner laid down by the Constitution.’24 The ‘Nation’ is considered to be an abstract collective encompassing all citizens of the past, the present and the future. Such an abstract collective cannot gather to decide upon policy issues. Therefore, it needs an elected representation, in which the MPs do not represent their voters, but the nation itself. This theory limits the citizens’ task to electing a new Parliament every now and then, after which the MPs decide autonomously upon all public matters. National sovereignty is both exclusive and inalienable, so that it cannot be assigned to any other authority.25 Article 33 is one of the Constitution’s most fundamental provisions, anchoring the national sovereignty, underpinning the unwritten principle of separation of powers and the representative system, and, more generally, expressing the idea of constitutionalism itself. But it also enshrines a living principle with multiple legal consequences, such as the prohibition of referenda. This prohibition covers even non-binding ones, because direct democracy is considered to be contrary to the representative system. Thus, every referendum requires a specific constitutional basis.26 Furthermore, Article 33 forbids delegation, because all powers must be exercised by the organs to which they were assigned by the Constitution.27 Several legal and factual developments have, however, mitigated the rigidity of the principle of national sovereignty, both in theory and in practice. National sovereignty’s exclusivity has been nuanced ‘upwards’, because of the emergence of supranational organisations, and ‘downwards’, by the federalisation process. Furthermore, the possibility of holding local and regional non-binding is constitutional); BECC No 2/2021, ECLI:BE:GHCC:2021:ARR.20210114.6 (electronic finger prints on identity cards do not violate the right to privacy); BECC No 57/2021, ECLI:BE:GHCC:2021:ARR.20210422.4 (annulment of the Act allowing systemic and undifferentiated data retention); BECC No 76/2021, ECLI:BE:GHCC:2021:ARR.20210520.11 (annulment of the COVID-19 measure restricting the right to be heard in Court for persons of unsound mind in cases involving the deprivation of their liberty). Many recent judgments concern socio-economic measures in light of budgetary constraints. Several other recent judgments concern anti-terrorism measures. In such cases, the Court grants the legislator an ample margin of appreciation. 22 Eg CE, Legislation Department, opinion of 10 July 1952, Parl Doc, Senate, 1952–1953, No 279. 23 Recent legal doctrine goes against many decades of applying ‘national sovereignty’, by suggesting that the drafters of the Belgian Constitution did not make a definitive choice between this theory of Siéyès and Rousseau’s theory of popular sovereignty, and that the latter theory actually fits the Belgian Constitution better (Brecht Deseure, Raf Geenens and Stefan Sottiaux, Sovereignty, Civic Participation, and Constitutional Law. The People versus the Nation in Belgium (London, Routledge, 2021)). As this is still a minority position, the following lines only describe the majority position of national sovereignty. 24 Only the Dutch, French and German versions of the Constitution are authentic. The Chamber of Representatives’ website, however, offers an up to date non-authentic English translation of the Constitution, www.dekamer.be/kvvcr/ pdf_sections/publications/constitution/GrondwetUK.pdf. 25 Alen/Haljan/Verrijdt, Constitutional Law (fn 2) 34 ff. 26 CE, Legislation Department, Opinion of 15 May 1985, Parl Doc, Chamber, 1983–1984, no 783/2; ibid, Opinion of 23 November 2004, Parl Doc, Chamber, 2004–2005, No 51-0281/004. 27 ECLI:BE:CASS:1891:ARR.18910206.3.

8  Willem Verrijdt referenda has been inserted into the Constitution.28 Likewise, delegation of power has been increasingly accepted. The Cour de cassation allows the competent authority to delegate accessory aspects of its power to other authorities, after having decided upon the essential aspects, and under its supervision.29 The Constitutional Court even accepts the delegation of essential aspects of matters reserved to the legislator by the Constitution, provided that certain conditions are met.30 Also, functional decentralisation, ie the attribution of regulatory powers to organs not belonging to the executive sensu stricto, yet under its administrative supervision, is on the rise. If functional decentralisation follows from EU obligations, even limitations to the administrative supervision and parliamentary control can be accepted (see fn 56). Although these developments nuance the principle of national sovereignty, they have not neutralised it: European integration and federalisation do not change the fact that KompetenzKompetenz still lies at the national federal level. The principled ban on referenda also remains intact for the federal level and for the matters in which local and regional referenda are not allowed by the Constitution. Moreover, indirect democracy remains the general rule at all levels. Likewise, the principled ban on delegations is still the rule.

B.  National Sovereignty and Accession to International Organisations These developments of the national sovereignty principle provided the legal context in which Belgium’s membership in several international organisations had to be accommodated. In order to provide for a post factum constitutional basis for Belgium’s membership in the EC and in several other international organisations having the competence to take decisions binding Belgium, a new Article 34 was inserted into the Constitution in 1970.31 It reads: ‘The exercising of specific powers can be assigned by a treaty or by a law to institutions of public international law.’ This provision can be read either as an exception to or as an application of Article 33 of the Constitution. As an exception, it makes the exercise of powers by institutions not mentioned in the Constitution possible. As an application, it allows the legislator, which is a power mentioned in the Constitution, to assign the exercise of a part of its powers to other institutions. Whichever reading is chosen, Articles 33 and 34 of the Constitution are part of one whole, and the latter should be interpreted restrictively. Since 1970, the Conseil d’Etat’s Legislation Department used Article 34 of the Constitution to justify further conferral of powers concerning, among others, the EC’s own resources,32 the Benelux Agreement on cross-border cooperation,33 the Dutch Language Union,34 the International Criminal Court,35 the European Constitution,36 the Lisbon Treaty,37 the Fiscal Compact,38 and the ESM Treaty.39 28 Arts 41 and 162 of the Constitution (local referenda) have been amended in 1999; Art 39bis (regional referenda) has been inserted in 2014. 29 ECLI:BE:CASS:1920:ARR.19200504.8. 30 These conditions are: the existence of exceptional circumstances justifying the delegation, its explicit and unambiguous framing, and a legislative validation within a short notice (BECC No 88/2004, ECLI:BE:GHCC:2004:ARR.20040519.2). 31 In the Constitution’s old numbering, Art 25 held the principle of national sovereignty, whereas the current Art 34 was laid down in Art 25bis. 32 CE, Legislation Department, opinion of 26 October 1970, Parl Doc, Chamber of Representatives, 1970–1971, No 711/3. 33 CE, Legislation Department, opinion of 16 June 1987, Parl Doc, Senate, 1988–1989, No 651/1. 34 CE, Legislation Department, opinion of 24 January 1995, Parl Doc, Flemish Parliament, 1994–1995, No 728/1. 35 CE, Legislation Department, opinion of 21 April 1999, Parl Doc, Senate, 1999–2000, No 2-329/1. 36 CE, Legislation Department, opinion of 15 February 2005, Parl Doc, Senate, 2004–2005, No 3-1091/1. 37 CE, Legislation Department, opinion of 29 January 2008, Parl Doc, Senate, 2007–2008, No 4-568/1. 38 CE, Legislation Department, opinion of 18 September 2012, Parl Doc, Flemish Parliament, 2012–2013, No 1815/1. 39 CE, Legislation Department, opinion of 3 April 2012, Parl Doc, Senate, 2011–2012 No 5-1598/1.

Belgium  9 In these legal opinions, the Conseil d’Etat developed five criteria which must be met for such a conferral to respect Article 34 of the Constitution: (1) the word ‘exercising’ in the text of Article 34 means that the Nation’s representatives cannot give away (parts of) the sovereignty as such: the conferral can only concern the t­ emporary exercise of powers. Hence, the Parliament can take these powers back at any time, even though this will likely imply Belgium’s leaving the international organisation. Also, if an international organisation ceases to exist, the powers assigned to it automatically revert to the Nation; (2) the conferral should only concern specific powers, which must be limited in scope and clearly defined; (3) every conferral requires the legislator’s consent, eg by approving the treaty establishing an international organisation. This approval must be clear and unambiguous, specifying which supranational organs may take decisions, in which matters and in which form; (4) the conferral must be to the benefit of an institution of public international law, ie not to the benefit of another state or of municipalities; (5) these transferrals may only deviate from the constitutional provisions concerning the exercise of powers: if a treaty or a legislative provision goes further than the mere conferral of powers, eg by deviating from other constitutional provisions, such as human rights, the treaty may only be approved and ratified after amending the constitutional provisions concerned. If a treaty provision allows the treaty organs to make further amendments to the treaty without the parties’ consent, two additional conditions apply. First, the legislator must explicitly consent to the conferral of such a treaty amending power to an international organ. Second, the limits to this power must be clearly defined before such consent is given. Unfortunately, the legislator has often ignored these legal opinions, approving some of these treaties without passing the necessary constitutional amendments. Also, the consecutive conferrals of powers to the EU have led to a body of supranational competences which can no longer be called ‘limited in scope’.40 The Constitutional Court’s case law on Article 34 of the Constitution is limited. It has not yet had the opportunity to examine the merits of a case challenging the compliance of an Act approving a treaty with the constitutional rules on the conferral of powers to institutions of public international law. On one occasion, it has used Article 34 of the Constitution to justify the implementation of an EU directive which was argued to violate the Constitution (see fn 56). In another judgment, it mentioned Article 34 of the Constitution when accepting that an EU regulation is a sufficient legal basis for a criminal indictment, because it is directly applicable in the Belgian legal order.41 More recently, it has based its national identity doctrine on Article 34 of the Constitution (see fn 79).42

C.  European Integration Within and Outside the EU Legal Order Article 34 of the Constitution is framed as a passe-partout for all international organisations, both within and outside the EU legal order. In the absence of a more specific provision, it is the 40 As the Conseil d’Etat’s Legislation Department itself stated in its opinion of 29 January 2008, Parl Doc, Senate, 2007–2008, No 4-568/1, 339. 41 BECC No 37/2010, ECLI:BE:GHCC:2010:ARR.20100422.2. 42 BECC No 62/2016, ECLI:BE:GHCC:2016:ARR.20160428.11.

10  Willem Verrijdt constitutional foundation of Belgium’s EU membership, justifying both the initial accession (even though this is an anachronism), later treaty amendments, and all decisions taken by EU organs. Although the criteria elaborated by the Conseil d’Etat’s Legislation Department for further conferral of powers to the EU or to other international organisations are large in number (see fn 39), none of them constitutes a substantive limit to further EU integration. Most of them are of a procedural nature, with the exception of the requirement that a conferral should concern specific powers. But the latter requirement is hardly justiciable and it has never obstructed the cumulative conferral of broadly defined powers to the EU. Only the Constitutional Court’s recently developed national identity doctrine (see fn 79) has the potential to develop into a substantive limit in the future. Concerning EU integration outside the EU legal order, the Conseil d’Etat’s Legislation Department made almost no remarks about the ESM Treaty. This can be explained, however, by the very sharp deadline set by the drafters of the approval Act, which only allowed it to examine which legislators had to be involved in the approval process. Noticing that Articles 5.6, 11.6, 19 and 44 of the ESM Treaty allow the ESM organs to make further amendments to this treaty and its annexes, it found that the limits to this conferral of power was clearly defined. It did add, however, that the legislator should be more explicit about its consent to this technique.43 The legislator implemented this suggestion, adding an Article 3 to the 20 June 2012 Act approving the ESM Treaty, stating that ‘the decisions of the Board of Governors based on the Articles 5.6, 11.6, 19 and 44 of the ESM Treaty shall have full effect’.

D.  The Complex Hierarchy Between International Treaties, EU Law and the Constitution The foregoing considerations raise the question about the hierarchical relation between the Constitution, international law, and primary and secondary EU law. In this regard, a distinction must be made between, on the one hand, the hierarchical relation between the Constitution and treaty law, including primary EU law, and, on the other hand, the hierarchical relation between the Constitution and secondary EU law.44

1.  The Relation between the Constitution and International Law According to most authors, the TEU, the TFEU and the Charter of Fundamental Rights ­(hereinafter: the Charter) should be considered as treaties of plain international law and should therefore follow the logic of the relation between the Constitution and international law.45 Concerning that relation, the Cour de cassation states that international law ranks higher than the Constitution, whereas the Constitutional Court and the Conseil d’Etat see that hierarchy the opposite way. According to the Cour de cassation, all international law ranks higher than all domestic law. In 1971, it ruled that all judges are to disapply legislation which violates a self-executing 43 CE, Legislation Department, opinion of 3 April 2012, Parl Doc, Senate, 2011–2012 No 5-1598/1, 42. 44 André Alen and Willem Verrijdt, ‘La relation entre la Constitution belge et le droit international et européen’ in I Riassetto et al (eds), Liber amicorum Rusen Ergeç (Luxembourg, Pasicrisie Luxembourgeoise, 2017) 31. 45 Eg Jérémie Van Meerbeeck and Michel Mahieu, ‘Traité international et Constitution nationale’, 61 Revue critique de jurisprudence belge (2007) 42, 45 ff; Emmanuel Slautsky, ‘De la hiérarchie entre Constitution et droit international’, 32 Administration Publique (2009) 227, 236; Willem Verrijdt, ‘Het Grondwettelijk Hof en het Unierecht: over ­constitutioneel pluralisme en rechterlijke dialoog’ in I Samoy et al (eds), Invloed van het Europese recht op het Belgische privaatrecht (Mortsel, Intersentia, 2012) 41, 89.

Belgium  11 international treaty which has been ratified in accordance with the constitutional procedures.46 In 2004, it applied this principle to constitutional provisions, without taking into account their special nature.47 The justifications given in the 1971 case were that the precedence of international law ‘follows from the nature itself of international law’. In a later judgment concerning primary EU law, the Cour de cassation stated that its precedence over all national norms is an unwritten general principle of law.48 According to the Constitutional Court, however, the Constitution trumps international treaties. It derives this hierarchy from Article 167 of the Constitution, which stipulates that treaties ‘take effect only after they have received the approval of the [competent legislator]’. Such an approval act is subject to constitutional review like any other Act of Parliament. Since such an approval only consists of one provision, stipulating that the treaty shall have full effect, the Court’s review must include the content of the treaty itself. As an additional argument, the Constitutional Court states that the legislator cannot be allowed, through ratifying a treaty, to circumvent existing constitutional limitations of his legislative powers.49 The aforementioned opinions of the Conseil d’Etat (see fn 39) show that it agrees with the primacy of the Constitution over international law, as it systematically advises that an unconstitutional treaty may not be ratified before the Constitution has been amended. The special majority legislator has implicitly endorsed the view of the Constitutional Court and the Conseil d’Etat. It did so by shortening the period for introducing an action for the annulment of legislative provisions approving a treaty to 60 days after their publication in the Moniteur belge,50 whereas the normal period is six months. Specifying a period would be pointless if the special majority legislator believed that legislative provisions approving international treaties cannot be subject to constitutional review. Likewise, as far as preliminary references are concerned, the special majority legislator only disallowed the judicial and administrative judges to refer a case to the Constitutional Court for a preliminary ruling if the legislative provision at hand approves a ‘constitutive’ EU treaty or the Convention or its additional protocols.51 This means that, a contrario, preliminary references about legislative provisions approving other treaties are not forbidden, implying that they may be subject to constitutional scrutiny. The majority of the legal doctrine52 takes the side of the Constitutional Court and the Conseil d’Etat. Other authors advocate a chronological hierarchy between the Constitution and treaties, disallowing the Parliament to either approve an unconstitutional treaty or pass a constitutional amendment violating a treaty. Recently, it is often stressed that, although the Constitution should be seen as the logical highest norm, because all powers are derived from it, international good faith and the pacta sunt servanda principle oblige all state actors, including all judges, to avoid conflicts between the Constitution and international law and to accommodate the requirements

46 Cour de Cassation, Le Ski (fn 13); ECLI:BE:CASS:1971:ARR.19710527.16; ECLI:BE:CASS:1989:ARR.19890510.8; ECLI:BE:CASS:2002:ARR.20021217.12. 47 ECLI:BE:CASS:2004:ARR.20041109.13; ECLI:BE:CASS:2004:ARR.20041116.16; ECLI:BE:CASS:2004.ARR.20041116.17. These cases concerned the freedom of opinion and the principle nullum crimen, nulla poena, sine lege. 48 ECLI:BE:CASS:2008:ARR.20080404.2. 49 BECC No 26/1991, ECLI:BE:GHCC:1991:ARR.199110161; BECC No 12/1994, ECLI:BE:GHCC:1994:ARR.19940203.10; BECC No 20/2004, ECLI:BE:GHCC:2004:ARR.20040204.4; BECC No 120/2011, ECLI:BE:GHCC:2011:ARR.20110630.5; BECC No 32/2013, ECLI:BE:GHCC:2013:ARR.20130307.5; BECC No 156/2016, ECLI:BE:GHCC:2016:ARR.20161208.2; BECC No 57/2020, ECLI:BE:GHCC:2020:ARR.20200507.9. 50 Art 3, § 2 of the Special Majority Act of 6 January 1989 on the Constitutional Court. 51 Art 26, § 1bis of the Special Majority Act of 6 January 1989 on the Constitutional Court. 52 For a summary of this doctrinal debate, see Yves Lejeune, ‘Les rapports normatifs entre la Constitution belge et le droit international ou européen dans l’ordre interne à l’aune des relations entre ordres juridiques primaires’, 47 Revue belge de droit international (2012) 372, 396; Verrijdt, ‘Grondwettelijk Hof ’ (fn 45) 81–98; Alen/Verrijdt, ‘Relation’ (fn 44) 32–39.

12  Willem Verrijdt of international law into the Belgian legal order. It is also argued that a hierarchical view on the relation between the national and the international ‘constitutions’ in a multi-layered legal order is outdated and should be replaced by more promising present-day views, such as constitutional pluralism and judicial dialogue, which should allow the effective and coherent functioning of all of these parts. In reality, conflicts between the Constitution and international law are very unlikely to occur. The Constitutional Court takes a deferential approach, systematically stressing that treaties are not mere expressions of national sovereignty, but also produce effects outside the Belgian legal order.53 So far, it has not declared any treaty provision unconstitutional.

2.  The Hierarchy between the Constitution and Secondary EU Law Article 34 of the Constitution plays a pivotal role for the hierarchy between the Constitution and secondary EU law.54 The Cour de cassation is the only one of the three highest courts to disregard Articles 33 and 34 of the Constitution in this context, as it bases the primacy of directly applicable secondary EU law over the Constitution on EU law itself.55 The Constitutional Court, by contrast, applies Article 34 of the Constitution in this context. The Court did so in a case where EU directives required the legislator to confer vast powers upon the independent federal energy regulatory office (CREG). The legislative provisions transposing these directives were challenged before the Court for the CREG’s lack of accountability towards the Minister and towards the Parliament, in violation of Articles 33 and 37 of the Constitution. The Constitutional Court, however, ruled that, ‘insofar as necessary’, the deviation from these constitutional rules was justified because of Article 34 of the Constitution.56 This judgment cannot be read as accepting the absolute primacy of secondary EU law over the Constitution. Before mentioning Article 34 of the Constitution, the Constitutional Court stated that parliamentary control over the CREG’s decisions existed and that judicial review was in no way reduced. Article 34 of the Constitution was thus used as an argument in the proportionality test. The Court rather reads Article 34 of the Constitution as granting the legislator a very large, albeit not unlimited, margin of appreciation when transposing EU law obligations. Accordingly, this judgment did not stand in the way of the development of a Belgian version of the Honeywell jurisprudence. Indeed, six years later, in its judgment on the ESM Treaty, the Constitutional Court did develop such jurisprudence (see fn 79). In that judgment, it further clarified its position, making three implicit statements about the hierarchy between secondary EU law and the Constitution: (i) this hierarchy is governed by Article 34 of the Constitution; (ii) in principle, this constitutional provision ranks secondary EU law higher than the Constitution; (iii) but it does not allow the legislative bodies of international organs to breach the principle of conferral, nor to disregard the Belgian national identity or the basic values of constitutional protection.57 53 See fn 50. 54 According to some authors, Art 34 also plays a role regarding the hierarchy between the Constitution and primary EU law (eg Lejeune, ‘Rapports normatifs’ [fn 52] 386–91), but this position is hard to reconcile with the Conseil d’Etat’s condition that a conferral of powers to an institution of public international law may not deviate from other constitutional provisions than the ones concerning the exercise of powers (fn 39). 55 ECLI:BE:CASS:2003:ARR.20030602.13, mentioning the Internationale Handelsgesellschaft judgment (CJEU 17.12.1970 11/70 (Internationale Handelsgesellschaft mbH v Einfuhr- und Vorratsstelle für Getreide und Futtermittel) ECLI:EU:C:1970:114). 56 BECC No 130/2010, ECLI:BE:GHCC:2010:ARR.20101118.2. See Michel Pâques, ‘Décentralisaton, régulation et contrôle démocratique. L’arrêt 130/2010 en question’ in D D’Hooghe et al (eds), Liber amicorum Marc Boes (Bruges, Die Keure, 2011) 411. 57 BECC 62/2016 (fn 43). See Philippe Gérard and Willem Verrijdt, ‘Belgian Constitutional Court Adopts National Identity Discourse’, 13 European Constitutional Law Review (2017) 182, 193–97.

Belgium  13 An earlier judgment concerning the relation between secondary EU law and the Convention had already – without mentioning Article 34 of the Constitution – shown that the Constitutional Court would not automatically accept that human rights are subordinate to secondary EU law. The Court requested a preliminary ruling from the European Court of Justice (ECJ) on whether the Money Laundering Directive violates Article 6 of the Convention insofar as it extends the duty to report suspect transactions to lawyers. Although the ECJ did not find a violation of Article 6 Convention,58 the Constitutional Court required the legislative provisions transposing the Money Laundering Directive to be interpreted in conformity with another Convention provision – Article 8.59 If the Constitutional Court had simply applied the logic of the primacy of secondary EU law over human rights, it either would have followed the ECJ’s judgment by finding no violation whatsoever or would have referred the same case to the ECJ again for a preliminary ruling on the directive’s compliance with the right to respect for private life. The Conseil d’Etat’s Legislation Department also uses Article 34 of the Constitution as foundation for the hierarchy between secondary EU law and the Constitution. It states that, once the exercise of a power has been conferred to an international organisation, this organisation can exercise it without being bound by the Belgian Constitution.60 According to the Conseil d’Etat, this even holds true for the ECJ when interpreting primary EU law. These interpretations become part of EU law itself, which, as far as Belgium is concerned, is justified because of Article 34 of the Constitution.61

IV.  Constitutional Limits to EU Integration Belgium is one of the six founding Member States of the EEC and has always been one of the strongest advocates of further integration. Several of the country’s most significant political figures have been involved in clearing the ground for subsequent steps in the integration process. Belgium has clear economic interests in further integration, given its size and location, its export and transit orientation, its hosting most EU institutions and housing many EU civil servants, its internationally oriented education, etc. It can, therefore, be expected that the Belgian Constitution and its interpretation will not be the ones posing the highest hurdles for further integration. Apart from the recently developed national identity limit, the most important limits concerning Belgium are indeed not of a substantive, but rather of a procedural and practical nature.

A.  The Conclusion and Amendment of Treaties in a Federal State A significant procedural hurdle to further EU integration is posed by the ‘mixed treaty’ procedure. Whereas in most federal states the treaty-making power resides at the federal level, the Belgian 58 ECJ 26.06.2007 C-305/05 (Ordre des barreaux francophones and germanophone and Others) ECLI:EU:C:2007:383. 59 BECC No 10/2008, ECLI:BE:GHCC:2008:ARR.20080123.1. According to the Constitutional Court, the only interpretation which respects Art 8 of the Convention requires that the information the lawyer receives during a confidential discussion with his client, whether or not in the course of legal representation, remains protected by the attorney-client privilege. 60 CE, Legislative Department, Opinion of 4 November 2005, Parl Doc, Chamber of Representatives, 2005–2006, No 51-2189/1, 113–16. 61 CE, Goosse, 5 October 1996, No 62.921; CE, Orfinger, 5 October 1996, No 62.922. The CJEU’s law shaping power was also accepted by the Constitutional Court (BECC No 90/2012, ECLI:BE:GHCC:2012:ARR.20120712.3 and BECC No 118/2012, ECLI:BE:GHCC:2012:ARR.20121010.4).

14  Willem Verrijdt Communities and Regions have been granted a genuine competence in this regard.62 Although Article 167 § 1 of the Constitution appoints the King as the head of Belgium’s foreign relations, the basic rule concerning the legislative approval of treaties is ‘in foro interno, in foro externo’: the substantive scope of a treaty defines the legislator competent for approving it (Article 167 § 3 of the Constitution). Treaties concerning a regional competence will, therefore, be negotiated, approved, and ratified by the Executives and the Legislatures of the Regions. The federal level is not involved, with the exception of the King’s competence to suspend treaty negotiations which threaten the coherence of Belgium’s foreign relations. Many treaties, however, are linked to more than one substantive competence and may concern both federal and federated matters. Article 167 § 4 of the Constitution delegates the elaboration of a procedure for negotiating and approving such ‘mixed treaties’ to the special majority legislator, which has in turn sub-delegated63 this task to the federal and federated legislators, by simply stating that they should regulate it in an ‘agreement of cooperation’.64 Although the Conseil d’Etat heavily criticised this sub-delegation, stressing that a matter for which the Constitution requires a special majority cannot be regulated in ordinary legislation,65 the conclusion of mixed treaties is currently regulated by the Agreement of cooperation of 8 March 1994 on the specific rules for the conclusion of mixed treaties.66 According to the agreement of cooperation, mixed treaty negotiations cannot start before being discussed in the Inter-Ministerial Conference on Foreign Policy, which decides upon the composition of the Belgian delegation. The signature at the international level is given by both the federal Minister of Foreign Affairs and a Minister appointed by each federated Government involved. All governments involved then submit the treaty for approval to their respective legislative bodies. Mixed treaties can only be ratified by the King and produce effects in the Belgian legal order after being approved by all competent legislative bodies. If even a single approval is lacking, all judges must refuse to apply the treaty.67 A treaty establishing further EU integration will often be a complex one, encompassing or influencing several substantive competences, and hence requiring approval through the mixed treaties procedure.68 After the sixth reform of state, amendments to the European treaties may necessitate the approval of as much as eight parliamentary assemblies: the Chamber of Representatives,69 the Flemish Parliament, the Walloon Parliament, the Brussels-Capital Parliament, the Parliament of the French Community, the Parliament of the German-speaking Community, the Commission communautaire francophone and the Commission communautaire commune. In theory, a simple majority in the Parliament of, say, the German-speaking Community, which has 25 members and represents some 76,000 citizens, can obstruct the entry into force of a treaty constituting a new step in European integration, despite the fact that all other European parliaments, including all other Belgian parliaments, have given their consent. 62 André Alen and Koen Muylle, Handboek van het Belgisch Staatsrecht (Mechelen, Kluwer, 2011) 745–61. 63 See Art 92bis, § 4ter of the Special Majority Act of 8 August 1980 on Institutional Reform, as amended on 5 May 1993. 64 An ‘agreement of cooperation’, as defined in Art 92bis, § 1, of the Special Majority Act of 8 August 1980 on Institutional Reform, is an agreement, negotiated between the federal Government and the governments of the communities and/or the regions, and subsequently approved by all legislators concerned if they involve aspects to be regulated by a legislative provision. Such agreements may concern the joint establishment or management of institutions, the joint exercise of proper competences, and the development of joint initiatives. 65 CE, Legislation Department, opinion of 16 September 1992, Parl Doc, Senate, 1991–1992, No 457/2. 66 Moniteur belge 06.03.1996. 67 ECLI:BE:CASS:2001:ARR.20010312.8. 68 The Conseil d’Etat’s Legislation Department examines for each treaty which legislators should be involved in the approval process (e.g., for the ESM Treaty, Doc Parl., Senate, 2011–2012, No 5-1598/1, 39). 69 The Senate is no longer involved in the approval of mixed treaties, as it has become an emanation of the federated parliaments which take part in the approval process directly.

Belgium  15 This possibility of a federated entity’s Parliament or Government blocking or slowing down the conclusion or amendment of a treaty is not merely theoretical. In 1998, the mixed treaty procedure delayed Belgium’s ratification of the Amsterdam Treaty. In the Commission communautaire commune, decisions must be taken by a majority in both language groups. Thus, nine out of the 17 Dutch-speaking members70 in this parliamentary body sufficed to bar the ratification by Belgium of a new EU treaty. When the Amsterdam Treaty was submitted for approval, the Dutch-speaking members71 threatened to use their de facto veto right, not because of any discontent with the treaty, but as a means of political pressure in order to receive concessions regarding some purely internal Flemish demands in the Brussels-Capital Region. As a result, the Commission communautaire commune only approved the Amsterdam Treaty on 14 January 1999, and Belgium only ratified it on 19 February 1999, the second to last Member State. It cannot be excluded that such political power play will occur again in the approval process of future treaty amendments. The fifth reform of state has, however, reduced that risk. In particular, draft legislation rejected by one language group in the Commission communautaire commune may now be resubmitted after 30 days, and in that case, one-third of the votes in each language group suffices for adopting the draft, provided that a simple majority is reached in total. In October 2016, the mixed treaty procedure delayed the approval of the Comprehensive Economic and Trade Agreement between Canada, the EU and its Member States. The governments of the Walloon Region, the French Community and the Brussels-Capital Region had opposed that treaty’s arbitration mechanism during the treaty negotiations. As the final draft included this mechanism anyway, these federated governments refused to authorise the Belgian Minister of Foreign Affairs to sign it. As a consequence, the ceremony during which the treaty was supposed to be signed was cancelled. One week later, the three opposing governments conceded, because the Federal Government promised that it would request the opinion of the ECJ on whether CETA’s arbitration mechanism is compatible with EU law. The treaty could not enter into force until the ECJ had given its opinion, nor if it concluded that it was incompatible with EU law.72

B.  Constitutionality Review of Treaty Amendments As the Constitution is usually not amended before approving and ratifying treaties contrary to it (see fn 40), it is not unlikely that a new EU Treaty is challenged before the Constitutional Court.73 In the past, such cases have always been inadmissible ratione temporis or because of a lack of standing, and even if they were admissible, they fell outside the Court’s review competences or lacked substance. Unlike some other Member States, Belgium currently has no strong lobby which systematically challenges the constitutionality of new constitutive EU Treaties before the Constitutional Court.

70 The CCC, competent for the exercise of certain bilingual community competences in the bilingual language zone Brussels-Capital, consists of all 89 members of the Brussels-Capital Parliament, i.e. 72 French-speaking and 17 Dutchspeaking MPs. 71 At that time, the CCC consisted of 75 MPs, 10 of whom were Dutch-speaking. Five of these Dutch-speaking members delayed the Commission’s approval for several months. 72 In its opinion 1/17 of 30 April 2019, the ECJ decided that this dispute settlement mechanism did not violate the autonomy of EU law (ECLI:EU:C:2019:341). 73 Eg, concerning the Lisbon Treaty BECC No 58/2009, ECLI:BE:GHCC:2009:ARR.20090319.7; No 125/2009, ECLI:BE:GHCC:2009:ARR.20090716.11, and, concerning the Fiscal Compact BECC No 62/2016, ECLI:BE:GHCC:2016:ARR.20160428.11.

16  Willem Verrijdt Procedurally, the window of opportunity for such cases is limited by the shortened delay for filing actions for annulment against acts approving treaties (see fn 50). The ratio legis behind this shortened delay of 60 days is ‘assuring the certainty and the stability of Belgium’s international relations’.74 The King or the federated Government should await the expiry of this period – or the Constitutional Court’s judgment, if an action for annulment is filed – before ratifying the treaty. If not, Belgium or the federated entity might be bound at the international level by a treaty which would subsequently be deprived of all its effects within the Belgian legal order by a Constitutional Court judgment. Unfortunately, the period’s expiry is usually not awaited. Hence, the shortened period is not a solid guarantee against disruptions of Belgium’s international relations.75 By contrast, the prohibition of preliminary references about acts approving constitutive EU treaties (see fn 51) contributes to the stability of Belgium’s international relations. The deadline for launching actions for annulment against all prior treaties establishing and broadening the EU’s substantive competences has expired. Thus, the conferral of powers to the EU under them can no longer be reviewed. A post factum impediment to further European integration steps as a result of Belgian constitutional review, thus, remains possible. Nevertheless, the Constitutional Court’s outspoken caution (see fn 53) in reviewing acts approving treaties is an important reassurance that it will not easily declare a treaty unconstitutional if such a finding would disturb Belgium’s international relations. This caution is likely to apply a fortiori to constitutive EU treaties, because the Constitutional Court would probably have regard to Belgium’s prior engagements in the field of European integration. It would presumably take into account that whether or not to take part in the further integration process is a political choice. Whereas in some EU Member States, the constitutional courts are the most likely actors to slow down further European integration, eg by annulling the adoption of parts of new EU treaties, and pose the most realistic threat to the unity of EU law, eg by annulling the legislation implementing or transposing secondary EU law, the Belgian Constitutional Court’s openness towards the EU (see fn 140) rather reinforces than threatens the full effect of EU law in Belgium.

C.  Substantive Limits to EU Integration The possibility of constitutional review of new EU treaties raises the question which constitutional provisions or principles have the potential to pose substantive limits to further European integration. The Constitutional Court has not yet developed limits such as the requirement that Belgium must remain an independent state or that certain core competences must remain in the hands of the Belgian legislators. Neither has it required the EU to feature certain constitutional characteristics, such as a minimum standard of human rights protection, democracy or the rule of law. Therefore, the most likely substantive limits to further EU integration are, on the one hand, Articles 33 and 34 of the Constitution, and, on the other hand, the human rights in Articles 8–32 of the Constitution. Articles 33 and 34 of the Constitution are not direct reference provisions for the Constitutional Court. However, they can be used as indirect reference provisions, combined with the principle of equality. Petitioners advocating that national sovereignty is violated by the conferral of new competences to the EU might argue that some of the criteria developed by the Conseil d’Etat (see fn 39) are not fulfilled. For example, it could be argued that the powers conferred to the

74 Parl

Doc, Senate, 1981–1982, No 246/1. and Muylle, Handboek (fn 62) 53.

75 Alen

Belgium  17 EU are virtually unlimited or are too vaguely defined. Such petitioners might also argue that constitutional provisions other than the ones concerning the exercise of powers are violated. On the basis of the Constitutional Court’s judgment concerning the Fiscal Compact (see fn 79), they might also invoke a breach of the Belgian national identity or of the basic values of constitutional protection. However, such arguments are unlikely to succeed due to the Constitutional Court’s outspoken caution (see fn 53) concerning constitutionality review of treaties. As far as human rights review is concerned, the Constitutional Court will probably apply the same caution. Moreover, when conducting its proportionality test, the Court may consider that the EU itself is an international organisation based on human rights and that the human rights enshrined in the Belgian Constitution largely coincide with those enshrined in the Convention and the Charter. Nevertheless, the CJEU’s Melloni jurisprudence and its opinion on the EU’s accession to the Convention76 might be a matter of concern in this regard.77

D.  Scrutiny of Secondary EU Legislation Against the Yardstick of Constitutional Requirements The aforementioned CREG judgment shows that the Constitutional Court is willing to allow the transposition of secondary EU law by the Belgian legislators, even if it raises questions of constitutionality (see fn 56). Nevertheless, it has recently developed a national identity doctrine, and, more implicitly, an ultra vires doctrine, based on Article 34 of the Constitution, reviving the possibility of examining secondary EU law against the yardstick of constitutional requirements. Article 34 of the Constitution neither defines nor limits the substantive scope of the competences which might be conferred on the EU. The Constitution does not feature an Ewigkeitsklausel and it does not distinguish between ‘ordinary’ and ‘higher’ constitutional provisions. The legal doctrine denies the existence of supra-constitutional provisions.78 Nevertheless, the Constitutional Court has recently used Article 34 of the Constitution to set three limits (two explicit and one implicit) to the conferral of new powers to the EU and to the rules of secondary EU law based on them. While rejecting an action for annulment against the act approving the Fiscal Compact because the petitioners failed to show how this treaty could adversely affect their interests, it added the following obiter dictum consideration: When approving a treaty which [assigns new powers to EU institutions], the legislator must respect Article 34 of the Constitution. By virtue of that provision, the exercising of specific powers can be assigned by a treaty or by a law to institutions of public international law. While these institutions may subsequently decide autonomously about how they exercise these competences, Article 34 of the Constitution cannot be interpreted as granting an unlimited permit to the legislator, when approving that treaty, or to the said institutions, when exercising the powers assigned to them. Article 34 of the Constitution does not allow a discriminating derogation to the national identity inherent in [Belgium’s] fundamental structures, political and constitutional, or to the basic values of the protection offered by the Constitution to all legal subjects.79

The ultra vires doctrine is implicitly present in Article 34 of the Constitution. Being an exception to and an application of Article 33 of the Constitution, and being limited to assigning the exercise 76 CJEU 26.02.2013 C-399/11 (Melloni) ECLI:EU:C:2013:107; CJEU 18.12.2014 Avis 2/13 (Accession of the Union to the ECHR) ECLI:EU:C:2014:2454. 77 Gérard/Verrijdt, ‘Identity Discourse’ (fn 57) 192 and 203 f. 78 Alen/Haljan/Verrijdt, Constitutional Law (fn 2) 54. 79 BECC 62/2016 (fn 42) B.8.7. See Gérard/Verrijdt, ‘Identity Discourse’ (fn 57) 182–205.

18  Willem Verrijdt of ‘specific’ competences, Article 34 of the Constitution can only justify the implementation and transposition of norms of secondary EU law adopted in respect of the EU’s substantive competences. However, it cannot justify the implementation and transposition of norms of secondary EU law falling outside the powers assigned to the EU. The Constitutional Court has not yet examined whether a norm of secondary EU law involved in a case under its scrutiny violated the principle of conferral. While only implicitly mentioning the principle of conferral, judgment No 62/2016 explicitly acknowledges the principle of national identity,80 in the words used in Article 4.2 TEU, and the basic values of constitutional rights protection, as limiting the exercise of powers by EU organs. It is not a coincidence that these three principles are also used by the Bundesverfassungsgericht. The Court did, however, confine itself to mentioning these principles, without specifying whether they are justiciable as a shield against primary or secondary EU law, and, if so, which judge is competent to decide in such an action. It did not delve into some key questions such as the legal consequences of an identity review, the procedure to be followed, and the margin to be granted to the EU. Given the Constitutional Court’s strong existing dialogue with the CJEU and its general stance regarding the full effect of EU law (see fn 140), one can expect that it will not apply these principles without due respect to the obligation of loyal cooperation with the EU or without requesting a preliminary ruling from the CJEU. The Court’s silence on these key items presumably shows that it merely intended to send a symbolic signal towards the CJEU, rather than announcing that it will actively start conducting national identity review any time soon. Nevertheless, the Court has made clear that it does not accept the CJEU’s mantra of absolute primacy, full effect, and uniform interpretation of EU law. The most difficult question in this regard is which parts of the Belgian Constitution can be considered to be part of Belgium’s national identity.81 Such aspects should be rather particular for one Member State, though not necessarily unique.82 They are likely to be linked to a country’s political and cultural history. As far as Belgium is concerned, it can be argued that such aspects relate, on the one hand, to the reasons why Belgium became an independent state, and, on the other hand, to the reasons why Belgium, as a bipolar state uniting two separate polities, has survived. The first set of aspects may be less significant than the second one. This is due to the long time that has passed since 1831 and the evolutionary interpretations of many of the constitutional provisions which were framed as a reaction to the Dutch King Willem I’s reign. Nevertheless, this historical background still explains why the Belgian Constitution puts so much emphasis on the legality principle, requiring the intervention of a democratically elected legislator in several matters, such as limitations to human rights. It also explains why several human rights, which were systematically ignored by King Willem I, still benefit from a wider protection by the Belgian Constitution than their counterparts in the Convention and the Charter (freedom of opinion and the press, freedom of assembly, freedom of religion, freedom of education and freedom of

80 See on the concepts ‘national identity’ and ‘constitutional identity’ in Elke Cloots, National identity in European Law (Oxford, OUP, 2015) 165–70. 81 See Elke Cloots, ‘Europese integratie en de eerbiediging van de nationale identiteit van de lidstaten’, in E Vandenbossche and S Van Drooghenbroeck (eds), Europese voorschriften en Staatshervorming/Contraintes européennes et réforme de l’Etat (Bruges, Die Keure, 2013) 19; Hugues Dumont, ‘L’intégration européenne et le respect de l’identité nationale des états (notamment fédéraux), in E Vandenbossche and S Van Drooghenbroeck (eds), Europese voorschriften … (2013) 55; Verrijdt, ‘Grondwettelijk Hof ’ (fn 45) 96 f. 82 Dumont, ‘Respect de l’identité’ (fn 80) 59 puts it as follows: ‘ce qui fait qu’un État est lui-même et non un autre, ce qui permet de le reconnaître et de le distinguer des autres’ (‘what makes a state to be itself and not another, what allows to recognise and distinguish it from others’; translation by the editor).

Belgium  19 language). If these specific human rights were to be considered to be part of Belgium’s constitutional identity, and to be opposable to primary or secondary EU law, this would create tension with the CJEU’s current interpretation of Article 53 of the Charter. The second set of aspects, which might be part of Belgium’s constitutional identity, reveals another particularity of the Belgian polity, ie its aptness for compromises. Belgian history shows several examples of bipolar oppositions, such as the ideological opposition between Catholics and Liberals, culminating in the ‘School Struggles’, which were resolved by the ‘School Pact’ of 1958,83 and the linguistic opposition between Flemish and Walloons, starting as soon as the 1840s, cumulating in violent student protests in 1968 and eventually leading to Belgium’s transformation into a federal state sui generis. Arguably, these fundamental compromises, which were reached after difficult negotiations and every single aspect of which is an equally important element of the pacifying equilibrium, are part of Belgium’s ‘fundamental structures, political and constitutional’,84 because they have resolved deeply rooted crises that could have meant the end of the Belgian state. If one or two elements of such a delicate compromise would later perish under EU scrutiny, the whole equilibrium would be lost. Therefore, it should at the very least be accepted that the basic choices which were made during Belgium’s federalisation process should be considered as part of its national identity. In this regard, Cloots considers that part of the national identity should be (1) the choice whether or not to grant regional authorities a degree of political autonomy, (2) the choice to become a federal state, (3) the circumscription of the federated entities, (4) the definition of their legislative powers, as well as their legislation’s applicability ratione personae and ratione materiae, and (5) the organs competent for exercising these powers and for safeguarding the distribution of competences.85 Dumont adds to that list the specific choices made regarding the operation of participative and cooperative federalism.86 In that regard, the highly politically sensitive regulation of knowledge and use of languages in Belgium comes to mind. That regulation is not only a particular feature of Belgian federalism, but also a matter in which compromise politics was stretched to its limits. In the Las case, the ECJ accepted that protection of the use of Dutch within the Dutch language zone is indeed justified under Article 4.2 TEU, but in its proportionality assessment, it stated that this goal could in casu have been pursued by means less intrusive into the free movement of persons than the obligation to exclusively use Dutch in all relations between employers and employees on Flemish soil.87

E.  Scrutiny of Secondary EU Legislation against the Yardstick of EU Law It is very unlikely that Belgian judges, including the Constitutional Court, would jeopardise the unity of EU law by conducting their own review of secondary EU law against primary EU law. Being well aware of the Foto-Frost jurisprudence, they will, in case of potential conflicts, always 83 The essential aspects of this School Pact, regarding the active and passive freedom of education and its financing, were anchored in Art 24 of the Constitution in 1989, in order to allow for the Constitutional Court’s review of Community Decrees’ compliance with this important compromise. 84 BECC 62/2016 (fn 42) B.8.7. 85 Cloots, ‘Europese integratie’ (fn 80) 25 f. 86 Dumont, ‘Respect de l’identité’ (fn 80) 66. 87 CJEU 16.04.2013 C-202/11 (Las) ECLI:EU:C:2013:239. Cloots argues that balancing is not the best technique to be used in the context of national identity, but that a categorisation technique offers better prospects (Cloots, ‘National identity’ [fn 80] 299–316).

20  Willem Verrijdt refer such cases to the CJEU, as the Constitutional Court has already done on thirteen occasions.88 A review of secondary EU law against the yardstick of the Convention, however, is not to be excluded. Such conclusion can be drawn from the aforementioned money laundering case (see fn 59). However, as that case shows, even such a review will only be conducted after a preliminary dialogue with the CJEU.

F.  Constitutional Amendments as a Consequence of EU Law As was stated above, despite the persistence of the Conseil d’Etat’s Legislation Department in stressing that the Constitution should be amended before approving treaties contrary to its provisions, other than the ones concerning the exercise of powers, these necessary amendments are often not passed (see fn 39). For example, the Rome Statute for the International Criminal Court was approved in 2000, despite the Conseil d’Etat’s Legislation Department pointing out that Article 27 of the Statute violated the constitutional provisions on the immunities of the King, Ministers and MPs. The necessary constitutional amendments still have not been adopted. Concerning the Lisbon Treaty, the Conseil d’Etat’s Legislation Department suggested the prior amendment of the principle nullum crimen, nulla poena sine lege in Article 12 of the Constitution (stating that ‘lege’ refers to a Belgian legislator). It considered that amendment necessary in the context of the creation of a European Public Prosecutor’s Office. However, such an amendment is yet to be adopted.89 In such cases, the constitutionality of the legislative provisions approving the treaty concerned can be challenged before the Constitutional Court, albeit within short deadlines (see fn 50). Whether Article 34 of the Constitution constitutes blanket permission for the transposition of unconstitutional directives or instead upholds the obligation of a prior amendment of the Constitution is a more difficult question. In 2005, the Conseil d’Etat’s Legislation Department subscribed to the first point of view, advising that the legislator was allowed to transpose a directive forbidding certain types of advertising for medication, although this directive violated Article 19 of the Constitution, which forbids preventive measures limiting the freedom of opinion. It argued that, once the exercise of power has been assigned to an international organisation, the latter is allowed to take autonomous decisions without being bound by the Belgian Constitution. Therefore, the Belgian Constitution cannot be used in order to refuse the transposition or implementation of an obligation laid down in secondary EU law.90 In its Lisbon opinion, The Conseil d’Etat’s Legislation Department was less firm on this point. It stated that Article 34 of the Constitution is open to two possible interpretations. One interpretation requires that the Constitution must be amended before unconstitutional directives may 88 The preliminary references already answered by the CJEU are ECJ 03.05.2007 C-303/05 (Advocaten voor de Wereld) ECLI:EU:C:2007:261; ECJ C-305/05 (fn 58); CJEU 01.03.2011 C-236/09 (Association Belge des Consommateurs Test-Achats and Others) ECLI:EU:C:2011:100; CJEU 21.10.2010 C-306/09 (B.) ECLI:EU:C:2010:626; CJEU 07.11.2013 C-473/12 (IPI) ECLI:EU:C:2013:715; CJEU 11.06.2015 C-1/14 (Base Company and Mobistar) ECLI:EU:C:2015:378; CJEU 28.07.2016 C-543/14 (Ordre des barreaux francophones and germanophone and Others) ECLI:EU:C:2016:605; CJEU 21.12.2016 C-76/15 (Vervloet and Others) ECLI:EU:C:2016:975; CJEU 06.06.2019 C-264/18 (P.M. and Others) ECLI:EU:C:2019:472; CJEU 17.12.2020 C-336/19 (Centraal Israëlitisch Consistorie van België and Others) ECLI:EU:C:2020:1031. See also the yet unanswered validity questions in BECC No 135/2019, ECLI:BE:GHCC:2019:ARR.20191017.1; BECC No 167/2020, ECLI:BE:GHCC:2020:ARR.20201217.6; BECC No 50/2021, ECLI:BE:GHCC:2021:ARR.20210325.2. The Constitutional Court’s website contains an overview of its entire dialogue with the CJEU. 89 Meanwhile, the Constitutional Court has accepted that an EU regulation suffices as a legal basis because of its direct applicability (BECC No 37/2010 (fn 41)), but this would probably not be accepted for secondary EU law lacking direct applicability. 90 CE, Legislation Department, opinion of 4 November 2005, Parl Doc Chamber of Representatives, 2005–2006, No 51-2189/1, 113–16.

Belgium  21 be transposed, whereas the other allows their transposition without amending the Constitution. Concerning the protection of constitutional rights, the Conseil d’Etat tried to trivialise its indecisiveness by suggesting that, under a certain interpretation of Article 53 of the Charter, national judges are allowed to set aside secondary EU law or its transpositions if they violate constitutional rights offering a higher human rights standard than the Charter.91 After the CJEU’s choice against that interpretation in its Melloni judgment, the discussion about the interpretation of Article 34 of the Constitution regains its full relevance. Following the Constitutional Court’s aforementioned jurisprudence on unconstitutional directives, most discrepancies will probably be interpreted away in a proportionality test. However, the theoretical possibility of declaring a directive unconstitutional persists. This is especially so since the emergence of the ‘national identity’ and ‘basic values of constitutional rights protection’ doctrines (see fn 79). Therefore, it is desirable that constitutional provisions that are incompatible with a directive which is to be transposed, are systematically amended.92 If the latter opinion is followed, Belgium’s rigid procedure for amending the Constitution may pose problems for the timely transposition of directives. It also makes a swift ratification of EU treaties impossible if a prior amendment of the Constitution is called for. As declarations for the revision of the Constitution entail legislative elections, these declarations are usually bundled and adopted at the end of the Parliament’s five-year term. Such a delay is likely to fall well beyond the deadline set for the directive’s transposition and may well be politically unacceptable in the context of ratifying new EU Treaties. The Conseil d’Etat’s Legislation Department, therefore, suggested amending Article 195 of the Constitution. Such amendment would provide for a more lenient procedure for constitutional revision in ‘European’ cases.93 This revision of the amendment procedure has, however, not yet taken place. So far, only one constitutional amendment has been a direct consequence of an EU obligation. In 1998, a third and fourth sentence were inserted into Article 8 of the Constitution. They were inserted in order to ensure the Constitution’s compliance with provisions of primary and ­secondary94 EU law on the voting rights of EU citizens. These European obligations were incompatible with Article 8 of the Constitution, which reserved the active and passive voting rights to Belgian nationals. Given that amendment of this provision took too long, Belgium was condemned in an infringement procedure.95 The necessary constitutional amendment was adopted a couple of months later, extending the active and passive voting rights to EU citizens, albeit only ‘in accordance with Belgium’s international and supranational obligations’, ie for the elections of the European Parliament and the municipal elections, and not for the elections of the Belgian parliaments.

V.  The Implementation of EU Law in Belgium: Rules and Practices A.  The Belgian Vote in the Council of Ministers The Minister representing Belgium in the Council may be a Minister either in the Federal Government or in a federated Government. Nevertheless, the vote that is cast is on behalf of 91 CE, Legislation Department, opinion of 29 January 2008, Parl Doc Senate, 2007–2008, No 4-568/1, 346. 92 The only other option, according to the Conseil d’Etat, is leaving the European Union (ibid, 343). 93 Ibid. 94 Council Directive 94/80/EC of 19 December 1994 laying down detailed arrangements for the exercise of the right to vote and to stand as a candidate in municipal elections by citizens of the Union residing in a Member State of which they are not nationals, [1994] OJ L 368/38. 95 ECJ 09.07.1997 C-323/97 (Commission v Belgium) ECLI:EU:C:1998:347.

22  Willem Verrijdt Belgium as a whole.96 This Minister does not receive any binding parliamentary instructions. However, the federal logic requires that the position to be taken in the Council is first discussed between the federal and federated governments concerned.97 To that end, coordination meetings with federal and federated representatives are organised by the Ministry of Foreign Affairs. If no agreement is reached there, the matter is discussed in the Inter-Ministerial Conference on Foreign Policy. If the latter does not reach an agreement either, Belgium will abstain in the Council. If these coordination meetings do lead to a consensus on Belgium’s position, the Minister representing Belgium has to take that position during the Council meeting. If, however, the Belgian position needs to be changed during the Council meeting in order to remain meaningful, the Minister may take a provisional position according to his analysis of the general interest, and the definitive Belgian position will be communicated to the president of the Council within three days, after being discussed in the aforementioned national organs.

B.  No Significant Parliamentary Involvement in the Drafting of EU Law When it comes to the parliamentary involvement in the drafting of secondary EU law, the Belgian federal structure entails some procedural difficulties, especially after the sixth reform of state. Belgium has, in general terms, declared that In accordance with its constitutional law, not only the Chamber of Representatives and Senate of the Federal Parliament but also the parliamentary assemblies of the Communities and the Regions act, in terms of the competences exercised by the Union, as components of the national parliamentary system or chambers of the national Parliament.98

This unilateral declaration, however, does not bind the EU institutions, especially when the European instruments for safeguarding the principle of subsidiarity regulate matters differently. According to Article 7 of protocol no 2 on the application of the principles of subsidiarity and proportionality,99 the two votes of a Member State with a bicameral parliamentary system are divided between these two chambers. This leaves no room for the parliaments of the Belgian federated entities to directly give reasoned opinions on the subsidiarity of draft EU legislation concerning their competences.100 Also, although Belgium still has a bicameral parliamentary system, the Senate can no longer play a significant role in this regard. After the sixth reform of state, the Belgian Senate lost almost all of its substantive competences and currently has no competences relevant in the context of EU law. Furthermore, it meets only once a month. Conversely, the significant extension of regional competences implies that many legislative initiatives at the EU level fall under the substantive competences of the parliaments of the Regions. Therefore, an agreement of cooperation between the federal and federated parliaments was needed in order to 96 Art 81 § 6 of the Special Majority Act of 8 August 1980 on Reform of the Institutions. 97 Art 92bis § 4bis of the Special Majority Act of 8 August 1980 on Reform of the Institutions, which was implemented by the Agreement of co-operation of 8 March 1994 concerning the representation of the Kingdom of Belgium in the Council of Ministers of the European Union. More specific rules apply for councils of ministers concerning Agriculture and Fishery. See Frank Ingelaere, ‘De Belgische deelstaten en de Europese Unie’ in F Judo and G Geudens (eds), Internationale betrekkingen en federalisme (Ghent, Larcier, 2006) 147; Luc Van Looy, ‘Het Vlaams Parlement als “nationaal parlement” in de Europese Unie (Ceci n’est pas une fiction)’, 8 Tijdschrift voor Wetgeving (2006) 28, 35 f. 98 TFEU, Declaration (No 51) by the Kingdom of Belgium on national Parliaments, [2016] OJ C 202/355. 99 TFEU, Protocol (No 2) on the application of the principles of subsidiarity and proportionality, [2008] OJ C 115/206–9. 100 The Flemish Parliament has nevertheless delivered one reasoned opinion, but it was sent by courtesy of the Senate, and in the IPEX database it is listed as being delivered by the Belgian Senate.

Belgium  23 allow the regional parliaments to ‘use’ the two Belgian votes by virtue of both federal Houses of Parliament.101 On 29 March 2017, the presidents of all Belgian parliamentary assemblies finally concluded this agreement of cooperation.102 Article 5 of the agreement allocates the two Belgian votes based on the number of reasoned opinions and the parliaments involved. If only one Parliament delivers a reasoned opinion, Belgium will cast only one vote if the subject-matter also belongs to the competences of other Belgian parliaments. Two votes will be cast if the subject-matter is an exclusive competence of the Parliament delivering the reasoned opinion. If, in a subject-matter belonging to the federal competences, both the Chamber of Representatives and the Senate deliver a reasoned opinion, Belgium will cast two votes.103 Where a subject-matter belongs to the federal and the federated competences, two votes will only be cast if at least one reasoned opinion is delivered by either the Chamber of Representatives or the Senate, and at least one reasoned opinion is delivered by a parliament of a federated entity. Where a subject-matter belongs to the federated competences, two votes will only be cast if a reasoned opinion is delivered by at least two parliaments of federated entities, provided that they belong to different language zones. Even leaving these procedural difficulties aside, the Belgian parliaments have made some use of the Early Warning System (EWS) in the beginning, but seem to have stopped using it since 2014. Between 2010 and 2014, the Belgian parliaments discussed draft EU legislation on 28 occasions. Twenty-one of these discussions took place in the Chamber of Representatives, three in the Walloon Parliament, two in the Flemish Parliament, one in the Senate, and one in the BrusselsCapital Parliament.104 These discussions led to only six reasoned opinions: four in the Chamber of Representatives,105 one in the Senate,106 and one in the Flemish Parliament.107 The Walloon Parliament and the Brussels-Capital Parliament each found one violation of the principle of subsidiarity, but these decisions were taken after the expiry of the eight-week period provided by the EWS.108 One of the reasoned opinions given by the Chamber of Representatives concerned the draft Regulation ‘on the exercise of the right to take collective action within the context of the freedom of establishment and the freedom to provide services’, which was shown a ‘yellow card’ and which was eventually withdrawn by the Commission.109 In 13 out of the 15 other occasions in which the Chamber of Representatives discussed draft EU legislation, it preferred the political dialogue under Protocol No 1.110 Between 2015 and mid-2021, the Chamber of Representatives discussed draft EU legislation on three occasions, opting for the political dialogue once111 and finding no breach of the principles of subsidiarity and proportionality twice.112 The other Belgian 101 Wouter Pas, ‘The Belgian “National Parliament” from the Perspective of the EU Constitutional Treaty’ in P Kiiver (ed), National and Regional Parliaments in the European Constitutional Order (Groningen, Europa Law Publishing, 2006) 57, 70–75; Patricia Popelier and Werner Vandenbruwaene, ‘De subsidiariteitstoets door de parlementen in België: politieke dialoog, lethargie en een gemiste kans’, 17 Tijdschrift voor Wetgeving (2015) 17, 24 f. 102 Agreement of cooperation of 29 March 2017 between the Federal State, the Regions, the Communities, the CCF and the CCC concerning the execution of the competences assigned to the national parliaments by the TEU, the TFEU and the Protocols attached thereto, Moniteur belge 17 July 2018. 103 This will hardly ever happen in practice, as the Senate is only very rarely involved in the legislative process since the sixth reform of state, most often in matters concerning the Belgian state structure. 104 Popelier and Vandenbruwaene, ‘Subsidiariteitstoets’ (fn 100) 25. 105 Parl Doc, Chamber of Representatives, 2011–2012, No 53-2068/1; ibid, 2011–2012, no 53-2145/1; ibid, 2011–2012, No 53-2221/1; ibid, 2012–2013, No 53-2910/1. See http://europe.dekamer.be. 106 Parl Doc, Senate, 2011–2012, No 5-1382/1. 107 Parl Doc, Flemish Parliament, 2012–2013, No 2038/2. 108 Parl Doc, Brussels-Capital Parliament, 2011–2012, A-282/2; Parl Doc, Walloon Parliament, 2011–2012, No 631/1. 109 Parl Doc, Chamber of Representatives, 2011–2012, No 53-2221/1. 110 TFEU, Protocol (No 1) on the Role of National Parliaments in the EU, [2016] OJ C 202/203. 111 Parl Doc, Chamber of Representatives, 2014–2015, No 54-1243/1. 112 Parl Doc, Chamber of Representatives, 2014–2015, no 54 1172/1; ibid, 2016–2017, No 54-2290/1.

24  Willem Verrijdt parliaments have not delivered any reasoned opinions during that period. The last reasoned opinions delivered by a Belgian Parliament dates back to 2013. There are several reasons for the limited contribution of Belgium’s parliaments, and especially the federated parliaments, to the EWS. First of all, EU integration is ‘rather well perceived in Belgium, implying that people do not necessarily want to object to any draft EU decision’.113 Secondly, smaller parliaments are not always well equipped to perform this demanding task: whereas the Chamber of Representatives has put in place a mechanism analysing the subsidiarity of all draft EU legislation, the smaller parliaments do not have the means to cope with the vast amount of EU documents within the short delays stipulated by Protocol No 2.114 For example, the one reasoned opinion delivered by the Flemish Parliament was an initiative of five MPs and was not amended during the plenary session. Furthermore, EWS’s effectiveness is limited, as it cannot lead to a ‘red card’ being pulled by the national parliaments. Therefore, next to the political dialogue under Protocol No 1, parliamentary assemblies may well prefer influencing the political process through their Ministers in the Council over using the EWS.115 In the drafting of primary EU law, the involvement of the federal Parliament is mandatory by virtue of Article 168 of the Constitution, which reads: From the beginning, the Houses are informed of negotiations concerning any revision of the treaties establishing the European Community [sic] and the treaties and acts which have modified or complemented them. They receive the draft treaty before its signature.

This involvement is limited to the provision of information, and does not necessarily lead to binding instructions for the negotiators. In practice, the chronology set by Article 167 of the Constitution, which empowers the Executive to negotiate and sign new treaties before submitting them to the competent parliaments for their approval, is not altered by this obligation.

C.  Joint Effort for the Implementation of EU Law 1.  Legislative Implementation The legislature is the key player in the transposition of directives and, more generally, in taking measures for the implementation of EU law. This is due to the residuary powers being located with the legislature and the importance of the legality principle in the Belgian Constitution. Belgium has never been a champion in the field of transposing EU legislation. This is mainly due to its complex state structure and to the complexity of the existing legislation. Several late and incorrect transpositions were, however, caused by administrative resistance and political lobbying. Two recent infringements sanctioned by the ECJ concerned the Urban Wastewater Directive116 and the Unfair Commercial Practices Directive117 of 11 May 2005. In the first case, Belgium was first condemned by the ECJ in an infringement procedure in 2004, because none of the three Regions had complied with the obligations following from the 113 Leuven Centre for Global Governance, The Subsidiarity Early Warning System of the Lisbon Treaty – the role of regional parliaments with legislative powers and other subnational authorities (2013) 39. 114 Popelier and Vandenbruwaene, ‘Subsidiariteitstoets’ (fn 101) 26 f; Leuven Centre for Global Governance, Subsidiarity (fn 113) 39. 115 Popelier and Vandenbruwaene, ‘Subsidiariteitstoets’ (fn 101) 28 f; Leuven Centre for Global Governance, Subsidiarity (fn 113) 39. 116 Council Directive 91/271/EEC of 21 May 1991 concerning urban waste-water treatment, [1991] OJ L 135/40. 117 Directive 2005/29/EC of the European Parliament and of the Council of 11 May 2005 concerning unfair business-toconsumer commercial practices in the internal market, [2005] OJ L 149/22.

Belgium  25 directive. Belgium was condemned again in 2013, leading to a million euro fine, because the Walloon Region still did not comply.118 The Federal Government provisionally paid this fine awaiting a political agreement on its division between the Regions.119 In the second case, the federal legislator had transposed the directive in time, but because of lobbying, it had excluded the liberal professions from the transposing Act’s scope. The Constitutional Court has, in two judgments on preliminary references, declared this exception to be in violation of the principle of equality, juncto the Unfair Commercial Practices Directive, and it has subsequently annulled it and retroactively extended the Act’s scope to liberal professions.120 Nevertheless, the CJEU still condemned Belgium, because the annulment post-dated the expiry of the deadline set by the Commission and the judgments on preliminary references were found not to constitute a sufficient legal basis for a directive’s correct transposition.121 In order to improve Belgium’s record concerning the implementation of EU law, the Senate ordered one of its commissions to file a report on the transposition of EU law and to make further recommendations to both the parliaments and the governments. This report, published in May 2015,122 included a number of recommendations on (1) the closure of Agreements of cooperation, (2) better sharing of information, (3) the appointment of ministers for European Affairs in all governments and of Europromotors in all federal and federated institutions, (4) more significant involvement in the drafting process of EU legislation, (5) the preparation of its transposition even before its publication in the Official Journal, (6) an early start of advisory procedures such as the determination of the competent legislator by the Conseil d’Etat’s Legislation Department, and (7) better determination of priorities. After this report, Belgium’s transposition deficit decreased from 2.2 per cent in 2016 to 0.8 per cent in 2019. This is, however, still higher than the EU average of 0.6 per cent. Given the fact that no less than 29 infringement procedures against Belgium are currently pending, predominantly in the fields of direct taxation and transport, Belgian still has room for improvement. Compliance with the ECJ judgments condemning Belgium in such procedures, takes 29 months on average.123 Both figures are close to the EU average. If Belgium is condemned by the CJEU because of a failure by a Community or a Region to transpose EU legislation correctly, the substitution mechanism of Article 169 of the Constitution may be applied.124 According to this procedure, the Federal Parliament or the Federal Government may temporarily take the place of a parliamentary assembly or a Government of a Community or a Region,125 if the latter’s actions or omissions infringe Belgium’s international or supranational obligations. This substitution mechanism can only be applied a posteriori, ie after Belgium has been condemned by the CJEU, the ECtHR, or another international tribunal, and only if Belgium had involved the federated entity concerned in the procedure leading to that judgment. Furthermore, the Council of Ministers must first have duly notified the federated entity of its intention to use the substitution mechanism, granting it a three-month period to comply with the judgment itself. After the expiry of that deadline, the Federal Parliament or the Federal 118 ECJ 08.07.2004 C-27/03 (Commission v Belgium) ECLI:EU:C:2004:418; CJEU 17.10.2013 C-533/11 (Commission v Belgium) ECLI:EU:C:2013:659. 119 The ‘Agreement on Government’ of the current federal Government, published on 9 October 2014, announces the creation of a tribunal competent for determining each entity’s share in future EU fines. 120 BECC No 55/2011, ECLI:BE:GHCC:2011:ARR.20110406.7; BECC No 192/2011, ECLI:BE:GHCC:2011:ARR.20111215.6; BECC No 99/2013, ECLI:BE:GHCC:2013:ARR.20130709.6. 121 CJEU 10.07.2014 C-421/12 (Commission v Belgium) ECLI:EU:C:2014:2064. 122 Parl Doc, Senate, 2014–2015, No 6-131/2, 290 p. 123 Single Market Scoreboard Edition 12/2019, http://ec.europa.eu/internal_market/scoreboard/. 124 This procedure is further elaborated by Art 16 § 3 of the Special Majority Act of 8 August 1980 on the Reform of the Institutions. 125 With the exception of the CCC.

26  Willem Verrijdt Government are allowed to take the necessary measures in order to implement the judgment instead of the federated entity concerned, even if this means disregarding the substantive distribution of competences. The measures taken at the federal level cease to have effect as soon as federated measures that sufficiently implement the judgment, enter into force.

2.  Executive Implementation Through Delegation No constitutional or statutory rule currently grants the Executive general competence to take all necessary measures in order to implement EU law. Nevertheless, a statute often delegates the implementation of specific international or EU law obligations, such as the transposition of directives, to the executive.126 Such delegations build on Article 105 of the Constitution, which allows the legislator to attribute specific competences to the King, who is then temporarily allowed to change existing legislation if this is necessary in order to comply with this task conferred upon him. Although delegations are generally considered to be difficult to reconcile with Article 33 of the Constitution (see fn 28), the Conseil d’Etat’s Legislation Department sees no harm in this procedure, because the international or European obligations to be implemented strictly limit the executive’s discretion.127 Nevertheless, the powers attributed to the executive must be welldefined: a general delegation to transpose all future EU legislation on telecommunication, while only some non-binding resolutions have been issued at the European level, cannot be justified under Article 105 of the Constitution.128

3.  Judicial Implementation Since the Cour de cassation’s Franco-Suisse Le Ski judgment of 27 May 1971,129 all judges are to refuse to apply legislative provisions violating international or EU law, provided that these international norms meet two criteria. First, the treaty must have been adopted according to the correct procedures, as laid down – for international treaties130 – in Article 167 of the Constitution: no treaty shall produce any effect if it has not been approved by the competent legislator.131 This also means that the treaty concerned must have entered into force in the international legal order.132 For secondary EU law, parliamentary approval is not necessary, because of the logic of Article 34 of the Constitution (see fn 31).133 The second criterion holds that the treaty or the norm of secondary EU law concerned is self-executing, ie directly grants rights to individuals, without

126 Paul Lemmens, ‘Opdrachten aan de Koning voor de omzetting van internationale handelingen in de interne wetgeving’, in J Krings (ed), Liber amicorum G. Baeteman (Deurne, Story-Scientia, 1997) 565. 127 CE, Legislation Department, opinion of 11 February 1997, Parl Doc, Chamber of Representatives, 1997–1998, No 49-1700/1. See also Michel Leroy, Les règlements et leurs juges (Brussels, Bruylant, 1987) 57f. 128 Ibid, opinion of 25 October 1995, Parl Doc, Chamber of Representatives, 1995–1996, No 49-208/2. 129 ECLI:BE:CASS:1971:ARR.19710527.16. See also ECLI:BE:CASS:1989:ARR.19890510.8; ECLI:BE:CASS:200 2:ARR.20021217.12; ECLI:BE:CASS:2008:ARR.20080404.2. 130 International customary law does not need legislative approval: it is automatically part of Belgian law (ECLI:BE:CASS:2012:ARR.20121122.3). 131 Obligations for individuals laid down in a norm of European or international law are only binding if this norm is published in the Moniteur belge. The rights conferred to individuals under EU or international law against the state are, by contrast, also binding without such publication (Dries Van Eeckhoutte and Jan Wouters, ‘Le droit international devant le juge belge’ [2013] 19 Revue belge de droit constitutionnel 299, 308 ff). 132 ECLI:BE:CASS:2001:ARR.20010312.8. 133 Van Eeckhoutte and Wouters, ‘Droit international’ (fn 130) 307 f, specifying that the basic treaty establishing this organisation must nevertheless have received parliamentary approval.

Belgium  27 further state action being needed. The Cour de cassation sees this obligation to grant primacy to directly applicable international or EU law as a general principle of law.134 Ever since that landmark judgment, all Belgian ordinary and administrative judges systematically allowed directly applicable and binding norms of EU law and international law to trump Belgian legislation.135 The Convention, the free movement rights, and other directly applicable norms of primary or secondary international and supranational law are invoked on a daily basis before Belgian courts. These courts willingly examine such actions or exceptions, inspired by a thorough analysis of the CJEU’s or ECtHR’s jurisprudence. If possible, they will first try to uphold the legislative provision using the technique of ‘conform interpretation’, ie by choosing the interpretation of national law that conforms with the higher norms. This technique is based on the presumption that the legislature or the executive did not intend to violate these higher norms.136 If the legislative provision under scrutiny is not open for such a conform interpretation, however, the judges will not hesitate to respect the Anwendungsvorrang of EU law. According to a study concerning judgments delivered in 2009 and 2010,137 EU law was relevant in 5.3 per cent of the cases before the Cour de cassation, whereas in 15.5 per cent of the cases the Convention was invoked or applied. In 30.8 per cent of the EU law-related cases, it refers to the CJEU’s case law, whereas in 5.8 per cent of the ECHR-related cases it refers to the ECtHR’s case law. In the same period, it referred six cases to the CJEU for a preliminary ruling, and refused to refer in six other cases.138 The Charter is used in less than 1 per cent of the Cour de cassation’s case law.139 The same openness towards EU law, the Convention, and other international law can be seen in the jurisprudence of the Constitutional Court.140 In this regard, the Constitutional Court’s position is atypical in the European landscape of constitutional courts. Many of its counterparts uphold a watertight division between, on the one hand, constitutionality review, which is their exclusive competence, and, on the other hand, review of legislation against international and EU law, which normally falls under the competence of the ordinary and administrative judges.

134 ECLI:BE:CASS:1994:ARR.19941205.11; ECLI:BE:CASS:2000:ARR.20001103.5. The Conseil d’Etat’s Legislation Department even calls it a ‘general principle of constitutional law’ (opinion of 25 April 2000, Parl Doc, Senate, 2000–2001, no 2-897/1). 135 See Nicolas Cariat and Janek Nowak (eds), Le droit de l’Union européenne et le juge belge/Het recht van de Europese Unie en de Belgische rechter (Brussels, Bruylant, 2015); Patricia Popelier, ‘Judicial Conversations in Multilevel Constitutionalism. The Belgian Case’ in M Claes et al (eds), Constitutional Conversations in Europe. Actors, Topics and Procedures (Mortsel, Intersentia, 2012) 73; Piet Van Nuffel, ‘Technieken van doorwerking van EU-recht in het Belgische privaatrecht’, in I Samoy et al (eds), Invloed van het Europese recht op het Belgische privaatrecht (Mortsel, Intersentia, 2012) 1. 136 ECLI:BE:CASS:2001:ARR.20010928.6; ECLI:BE:CASS:2003:ARR.20030212.22; Van Eeckhoutte and Wouters, ‘Droit international’ (fn 130) 312 ff; Piet Van Nuffel, ‘Doorwerking van Europees Gemeenschapsrecht in de Belgische rechtsorde’ in J Wouters and D Van Eeckhoutte (eds), Doorwerking van internationaal recht in de Belgische rechtsorde. Recente ontwikkelingen in een rechtstakoverschrijdend perspectief (Mortsel, Intersentia, 2006) 337, 351-59. 137 Popelier, ‘Judicial Conversations’ (fn 134) 81, 83 and 85. 138 In the same period, the Conseil d’Etat referred a case to the ECJ for a preliminary ruling in 17 cases and refused to refer in 103 cases. 139 Sarah Lambrecht, ‘Belgium: the EU Charter in a Tradition of Openness’, in M Bobek and J Adams-Prassl (eds), The EU Charter of Fundamental Rights in the Member States (Oxford, OUP, 2021) 95. 140 See André Alen, Jean Spreutels, Etienne Peremans and Willem Verrijdt, ‘Cour constitutionnelle de Belgique’, in R Huppmann and R Schnabl (eds), The Cooperation of Constitutional Courts in Europe. Current Situation and Perspectives (Vienna, Verlag Österreich, 2014); Sarah Lambrecht, ‘The Attitude of Four Supreme Courts towards the European Court of Human Rights: Strasbourg has spoken …’, in S Besson and N Levrat (eds), Le juge en droit européen et international/ The Judge in European and International Law (Zurich, Schulthess, 2013) 301; Popelier, ‘Judicial Conversations’ (fn 134) 73–99; Marc Verdussen, ‘La Cour constitutionnelle, partenaire de la Cour de justice de l’Union européenne’, 17 Revue belge de droit constitutionnel (2011) 81; Willem Verrijdt, ‘Het Grondwettelijk Hof als rechter van EU-recht’, in N Cariat and J Nowak, Le droit de l’Union européenne et le juge belge/Het recht van de Europese Unie en de Belgische rechter (Brussels, Bruylant, 2015) 253.

28  Willem Verrijdt Formally, the Constitutional Court can only review legislation against some provisions of the Constitution and against Special Majority Acts distributing the legislative competences of the federal and federated levels. International and EU law is not part of the Court’s direct reference provisions. Nevertheless, the Court developed two techniques allowing it to use the latter norms as indirect reference provisions. The first technique, based on the principle of equality (Articles 10 and 11 of the Constitution), is based on the statement that these constitutional provisions prohibit any and all discrimination, irrespective of its origin. Consequently, the Constitutional Court can review legislation against the constitutional principle of equality and non-discrimination, in combination with rights guaranteed in international and EU treaties which bind Belgium.141 This technique allows the Court to review legislation against human rights in the Convention or the Charter, primary or secondary EU law, and even general principles of EU law. These indirect reference provisions do not need to be self-executing. The Constitutional Court’s adjudication being a contentieux objectif, in which a lower norm is reviewed against a higher norm, it is irrelevant whether or not that higher norm grants rights to individuals.142 A second technique of indirect review used by the Constitutional Court consists of taking into account international provisions guaranteeing analogous rights and freedoms when it reviews legislation against constitutional rights. In such instances, the analogous constitutional and international provisions (eg the freedom of opinion in Article 19 of the Constitution and in Article 10 of the Convention) form an ‘inextricable unity’.143 When applying this technique, the Constitutional Court upholds the principles of widest protection and uniform interpretation of human rights, demanding that limitations to a human right meet both the criteria laid down in the Constitution and the ones laid down in the Convention. This technique also allows the Court to amply cite and apply the jurisprudence of the CJEU and the ECtHR. Thus, the Constitutional Court maximises rights protection for citizens and ensures coherence of rights protection and legal certainty. This technique also enhances the evolutionary interpretation of the constitutional rights, most of which are unchanged since 1831. Both techniques allow the Constitutional Court to avoid falling below the minimum standard of human rights protection set by the ECtHR and, more generally, to uphold EU and international law. Because of this, EU law and the Convention play an important role in the Constitutional Court’s jurisprudence. In 2013, it used EU law rules as indirect reference provisions in 18 out of 183 judgments, and referred to the CJEU’s case law in 26 of its judgments. In 2014, legislation was reviewed against EU law in 17 out of 191 judgments and CJEU case law was referred to in 18 judgments. In the same year, the Convention was mentioned in 55 judgments, 41 of which contained references to ECtHR case law.144 The Constitutional Court makes greater use of the Convention than of the Charter: in 2018, it referred to the Convention in 33.8 per cent of its judgments145 and to the Charter in 17 per cent of its judgments.146 This is because, on the one 141 BECC No 23/1989, ECLI:BE:GHCC:1989:ARR.19891013.1. See Alen/Spreutels/Peremans/Verrijdt, ‘Cour constitutionnelle’ (fn 140) 295–300. 142 BECC No 41/2002, ECLI:BE:GHCC:2002:ARR.20020220.4; BECC No 106/2003, ECLI:BE:GHCC:2003:ARR.20030722.6. 143 BECC No 136/2004, ECLI:BE:GHCC:2004:ARR.20040722.9; BECC No 189/2005, ECLI:BE:GHCC:2005:ARR.20051 214.17; BECC No 195/2009, ECLI:BE:GHCC:2009:ARR.20091203.1. See Alen/Spreutels/Peremans/Verrijdt, ‘Cour constitutionnelle’ (fn 140) 300-03; Géraldine Rosoux, Vers une ‘dématérialisation’ des droits fondamentaux? Convergence des droits fondamentaux dans une protection fragmentée, à la lumière du raisonnement du juge constitutionnel belge (Brussels, Bruylant, 2015) 39. 144 See Alen/Spreutels/Peremans/Verrijdt, ‘Cour constitutionnelle’ (fn 140) 303 f and 308 ff. 145 André Alen and Willem Verrijdt, ‘The Dialogue Between the European Court of Human Rights and Domestic Constitutional Courts. The Belgian Example’ in K Lemmens, S Parmentier and L Reyntjens (eds), Human Rights with a Human Touch. Liber amicorum Paul Lemmens (Antwerp, Intersentia, 2019) 171. 146 Lambrecht, ‘Belgium’ (fn 139) 95.

Belgium  29 hand, the latter has a more limited scope and, on the other hand, the Convention has a longer tradition. Where the Charter is used, the Court makes ample use of the CJEU’s case law.147 The Constitutional Court’s openness towards EU law is also shown by its extensive preliminary reference dialogue with the CJEU. So far, the Court has referred 39 cases to the CJEU, accounting for 133 distinct questions, and has subsequently followed the CJEU’s judgments.148 Furthermore, the Constitutional Court fully respects the procedural rules flowing from ECJ case law on the full effect of EU law. Therefore it respects the obligations concerning EU conform interpretation,149 provisional measures,150 ex officio application of EU law in the context of procedural equivalence,151 prohibition on time-modulating the effects of judgments if this would jeopardise the uniform application of EU law,152 and accepting the Commission’s standing in certain matters.153,154

VI.  Conclusion: The Relation between EU Law and Belgian Law The Belgian Constitution and the way it is interpreted and applied, reflect the EU-friendliness which is generally present in Belgian society. Being among the six founding Member States, having favoured its further development, housing most of its institutions, and being an internationally oriented economy, Belgium has a significant interest in the EU and in its further development. Therefore, it comes as no surprise that EU law and EU integration are generally well perceived by Belgian politicians and lawyers alike. Likewise, this context explains why the Constitution and its interpretation and application currently do not pose high hurdles for further EU integration. Before 1970, there was no constitutional basis for Belgium’s membership of several international organisations, including the EC. These memberships were, therefore, deemed contrary to the principle of national sovereignty laid down in Article 33 of the Constitution. When this constitutional basis was created post factum in 1970, it was conceived as a general permission for the legislator to enter international organisations and to be bound by the decisions of their organs. This constitutional basis, currently laid down in Article 34 of the Constitution, applies to the EU and to other international and supranational organisations alike. On a theoretical level, this permission is strictly limited. It can only involve the conferral of the ‘exercise’ of ‘specific’ powers. However, the way in which it is applied – sometimes even contrary to the legal opinions 147 Eg BECC No 116/2011, ECLI:BE:GHCC:2011:ARR.20110630.1 (annulment of a legislative provision allowing distinct life insurance fees for men and women, for violation of the Arts 21 and 23 of the Charter); BECC No 57/2021, ECLI:BE:GHCC:2021:ARR.20210422.4 (annulment of the Belgian Data Retention Act). 148 See Alen/Verrijdt, ‘Relation’ (fn 44) 33–74. In one case, the BECC criticised the CJEU’s answer but nevertheless followed it (BECC No 89/2011, ECLI:BE:GHCC:2011:ARR.20110531.2). There is only one case in which the Constitutional Court went further than the ECJ’s judgment by applying a higher human rights standard (see fn 59). 149 BECC No 161/2012, ECLI:BE:GHCC:2012:ARR.20121220.9. 150 BECC No 96/2010, ECLI:BE:GHCC:2010:ARR.20100729.8; BECC No 183/2013, ECLI:BE:GHCC:2013:ARR.2013 1219.24; BECC No 167/2020, ECLI:BE:GHCC:2020:ARR.20201217.6. 151 BECC No 149/2010, ECLI:BE:GHCC:2010:ARR.20101222.1; BECC No 97/2011, ECLI:BE:GHCC:2011:ARR.2011 0531.10; BECC No 15/2015, ECLI:BE:GHCC:2015:ARR.20150205.5. 152 BECC No 18/2012, ECLI:BE:GHCC:2012:ARR.20120209.2; BECC No 154/2012, ECLI:BE:GHCC:2012:ARR.2012122 0.2BECC No 144/2013 (fn 21). 153 BECC No 161/2012 (fn 145, on the basis of Art 15.3 of the Regulation No 1/2003). 154 See Jan Theunis, ‘Het Grondwettelijk Hof en de procedurele verplichtingen uit het Europees Unierecht’, in W Pas, P Peeters and W Verrijdt (eds), Liber discipulorum André Alen (Bruges, Die Keure, 2015) 409; Michelle Daelemans, ‘Supranationale beperkingen op de mogelijkheid om de rechtsgevolgen van vernietigingsarresten te handhaven’ in W Pas, P Peeters and W Verrijdt (eds), Liber discipulorum André Alen (Bruges, Die Keure, 2015) 39.

30  Willem Verrijdt of the Conseil d’Etat’s Legislation Department – allows for rather extensive conferral of powers to the EU. Apart from the limits recently developed by the Constitutional Court relating to the Belgian national identity and the basic values of constitutional protection, the permission neither limits substantively the powers that may be assigned to the EU, nor requires difficult procedures such as a referendum. The ordinary legislative majority applies for approving new EU treaties, provided that the Parliament is kept posted during the negotiation process. The main hurdle to further EU integration is of a practical nature. Since a new conferral of competences may affect the substantive competences of both the federal level and all federated entities, legislative approval of up to eight distinct parliamentary bodies may be necessary before a new EU treaty can be ratified by Belgium. Having veto powers, these parliamentary bodies could, thus, use their power to delay or prevent further EU integration as leverage for obtaining concessions in purely domestic political discussions. Although, on a theoretical level, the hierarchy between the Constitution and international law is debated between the country’s three highest courts, EU law’s primacy is guaranteed in practice. Primary EU law’s constitutionality can only be challenged for 60 days from the publication in the Moniteur belge of its legislative approval. Moreover, the Constitutional Court takes a deferential approach in such cases. Legislation transposing or implementing secondary EU law can also be challenged before the Constitutional Court. However, this Court refers to Article 34 of the Constitution in order to grant the legislator a very broad margin of appreciation for meeting the obligations flowing from EU law. Furthermore, although the Constitutional Court has recently nuanced the primacy of primary and secondary EU law over the Constitution by referring to the Belgian national identity and to the basic values of human rights protection, it has not yet put these limits into practice. In case of doubt concerning secondary EU law’s constitutionality or its compliance with primary EU law or, through the prism of Article 6 TEU, the Convention, the Constitutional Court always refers the case to the CJEU for a preliminary ruling on the validity of the piece of secondary EU law. Be that as it may, the Constitutional Court has not yet agreed to the CJEUs Melloni jurisprudence, and at least one judgment suggests that it might give precedence to the Convention over secondary EU law. Furthermore, the Constitutional Court’s acknowledgement of the principle of national identity proves that it will not automatically accept all features of primacy and full effect of EU law. All potential problems in this regard would disappear if the Constitution was systematically amended before approving new EU treaties or transposing or implementing secondary EU law contrary to it. Although the Conseil d’Etat systematically advocates for this, it has not been applied consistently. Discrepancies between the Constitution and secondary EU law might also be avoided if Belgium systematically scrutinises the constitutionality of draft EU legislation and is actively involved in the drafting process, for example through the EWS. Figures show, however, that Belgium’s responsiveness in this regard has room for improvement. The same holds true for transposing and implementing secondary EU law. Belgium’s transposition deficit and yearly rate of infringement procedures are still above the EU average. By contrast, the courts’ involvement in the application and enforcement of EU law against the legislature and the executive is exemplary. The latter statement also includes the Constitutional Court, which clearly operates as a factor of EU integration. The conclusion should, therefore, be that further developments in EU integration are unlikely to meet significant constitutional resistance in Belgium. In spite of some procedural difficulties which may cause delays, all parliaments concerned are likely to approve such new steps. The Constitutional Court would presumably also not jeopardise these steps. This can be said based on the Court’s usual integrationist and deferential approach to the EU and given the absence of a strong anti-integrationist lobby in Belgium which systematically challenges such integrationist steps before it.

Belgium  31

References A Alen and K Muylle, Handboek van het Belgisch Staatsrecht (Mechelen, Kluwer, 2011). A Alen, J Spreutels, E Peremans and W Verrijdt, ‘Cour constitutionnelle de Belgique’, in R Huppmann and R Schnabl (eds), The Cooperation of Constitutional Courts in Europe. Current Situation and Perspectives (Vienna, Verlag Österreich, 2014). A Alen and W Verrijdt, ‘La relation entre la Constitution belge et le droit international et européen’ in I Riassetto et al (eds), Liber amicorum Rusen Ergeç (Luxembourg, Pasicrisie Luxembourgeoise, 2017) 31. A Alen and W Verrijdt, ‘The Dialogue Between the European Court of Human Rights and Domestic Constitutional Courts. The Belgian Example’ in K Lemmens, S Parmentier and L Reyntjens (eds), Human Rights with a Human Touch. Liber amicorum Paul Lemmens (Antwerp, Intersentia, 2019) 155. A Alen, D Haljan and W Verrijdt, Constitutional Law in Belgium (Alphen aan den Rijn, Wolters Kluwer, 2020). E Cloots, ‘Europese integratie en de eerbiediging van de nationale identiteit van de lidstaten’, in E Vandenbossche and S Van Drooghenbroeck (eds), Europese voorschriften en Staatshervorming/ Contraintes européennes et réforme de l’Etat (Bruges, Die Keure, 2013) 19. E Cloots, National identity in European Law (Oxford, Oxford University Press, 2015). M Daelemans, ‘Supranationale beperkingen op de mogelijkheid om de rechtsgevolgen van vernietigingsarresten te handhaven’ in W Pas, P Peeters and W Verrijdt (eds), Liber discipulorum André Alen (Bruges, Die Keure, 2015) 39. B Deseure, R Geenens and S Sottiaux, Sovereignty, Civic Participation, and Constitutional Law. The People versus the Nation in Belgium (London, Routledge, 2021). H Dumont, ‘L’intégration européenne et le respect de l’identité nationale des états (notamment fédéraux), in E Vandenbossche and S Van Drooghenbroeck (eds), Europese voorschriften en Staatshervorming/ Contraintes européennes et réforme de l’Etat (Bruges, Die Keure, 2013) 55. P Gérard and W Verrijdt, ‘Belgian Constitutional Court Adopts National Identity Discourse’, 13 European Constitutional Law Review (2017) 182. J Goossens and P Cannoot, ‘Belgian Federalism after the Sixth State Reform’ (2015) 7(2) Perspectives on Federalism 29. F Ingelaere, ‘De Belgische deelstaten en de Europese Unie’ in F Judo and G Geudens (eds), Internationale betrekkingen en federalisme (Ghent, Larcier, 2006) 147. Sarah Lambrecht, ‘The Attitude of Four Supreme Courts towards the European Court of Human Rights: Strasbourg has spoken …’ in S Besson and N Levrat (eds), Le juge en droit européen et international/The Judge in European and International Law (Zurich, Schulthess, 2013) 301. Sarah Lambrecht, ‘Belgium: the EU Charter in a Tradition of Openness’, in M Bobek and J Adams-Prassl (eds), The EU Charter of Fundamental Rights in the Member States (Oxford, OUP, 2021) 87. Y Lejeune, ‘Les rapports normatifs entre la Constitution belge et le droit international ou européen dans l’ordre interne à l’aune des relations entre ordres juridiques primaires’ (2012) 47 Revue belge de droit international 372. P Lemmens, ‘Opdrachten aan de Koning voor de omzetting van internationale handelingen in de interne wetgeving’, in J Krings (ed), Liber amicorum G. Baeteman (Deurne, Story-Scientia, 1997) 565. M Leroy, Les règlements et leurs juges (Brussels, Bruylant, 1987). Leuven Centre for Global Governance, The Subsidiarity Early Warning System of the Lisbon Treaty – the role of regional parliaments with legislative powers and other subnational authorities (2013), http://cor/europa. eu/en/documentation/studies/Documents/subsidiarity-warning-system-lisbon-treaty.pdf. M Pâques, ‘Décentralisaton, régulation et contrôle démocratique. L’arrêt 130/2010 en question’ in D D’Hooghe et al (eds), Liber amicorum Marc Boes (Bruges, Die Keure, 2011) 411. W Pas, ‘The Belgian “National Parliament” from the Perspective of the EU Constitutional Treaty’ in P Kiiver (ed), National and Regional Parliaments in the European Constitutional Order (Groningen, Europa Law Publishing, 2006) 57. P Peeters and D Haljan, ‘Belgium’s Sixth State Reform: The State of the Nation(s)’ (2016) 22 European Public Law 411.

32  Willem Verrijdt P Popelier, ‘Judicial Conversations in Multilevel Constitutionalism. The Belgian Case’ in M Claes et al (eds), Constitutional Conversations in Europe. Actors, Topics and Procedures (Mortsel, Intersentia, 2012) 73. P Popelier and W Vandenbruwaene, ‘De subsidiariteitstoets door de parlementen in België: politieke dialoog, lethargie en een gemiste kans’ (2015) 17 Tijdschrift voor Wetgeving 17. G Rosoux, Vers une ‘dématérialisation’ des droits fondamentaux? Convergence des droits fondamentaux dans une protection fragmentée, à la lumière du raisonnement du juge constitutionnel belge (Brussels, Bruylant, 2015). G Rosoux, Contentieux constitutionnel (Brussels, Larcier, 2021). E Slautsky, ‘De la hiérarchie entre Constitution et droit international’ (2009) 32 Administration Publique 227. J Theunis, ‘Het Grondwettelijk Hof en de procedurele verplichtingen uit het Europees Unierecht’, in W Pas, P Peeters and W Verrijdt (eds), Liber discipulorum André Alen (Bruges, Die Keure, 2015) 409. D Van Eeckhoutte and J Wouters, ‘Le droit international devant le juge belge’ (2013) 19 Revue belge de droit constitutionnel 299. L Van Looy, ‘Het Vlaams Parlement als “nationaal parlement” in de Europese Unie (Ceci n’est pas une fiction)’, 8 Tijdschrift voor Wetgeving (2006) 28. J Van Meerbeeck and M Mahieu, ‘Traité international et Constitution nationale’ (2007) 61 Revue critique de jurisprudence belge 42. P Van Nuffel, ‘Doorwerking van Europees Gemeenschapsrecht in de Belgische rechtsorde’ in J Wouters and D Van Eeckhoutte (eds), Doorwerking van internationaal recht in de Belgische rechtsorde. Recente ontwikkelingen in een rechtstakoverschrijdend perspectief (Mortsel, Intersentia, 2006) 337. P Van Nuffel, ‘Technieken van doorwerking van EU-recht in het Belgische privaatrecht’, in I Samoy et al (eds), Invloed van het Europese recht op het Belgische privaatrecht (Mortsel, Intersentia, 2012) 1. M Verdussen, ‘La Cour constitutionnelle, partenaire de la Cour de justice de l’Union européenne’ (2011) 17 Revue belge de droit constitutionnel 81. W Verrijdt, ‘Het Grondwettelijk Hof en het Unierecht: over constitutioneel pluralisme en rechterlijke dialoog’ in I Samoy et al (eds.), Invloed van het Europese recht op het Belgische privaatrecht (Mortsel, Intersentia, 2012) 41. W Verrijdt, ‘Het Grondwettelijk Hof als rechter van EU-recht’, in N Cariat and J Nowak, Le droit de l’Union européenne et le juge belge/Het recht van de Europese Unie en de Belgische rechter (Brussels, Bruylant, 2015) 253.

2 Bulgaria Structural Adjustments of the Bulgarian to the EU Constitutional Order: Between the Conditionality of Democracy in Transition and the Challenges of the Multilevel Constitutionalism of the EU MARTIN BELOV

I. Introduction This chapter analyses the effects that transition to democracy, introduction of the principle of open statehood,1 and European integration have had on the Bulgarian constitutional order. These three processes have been partially overlapping both substantively and temporally. They have changed profoundly the institutional design and the constitutional axiology of the Bulgarian constitutional system. They have also served as stimuli for profound structural reforms at normative and institutional levels. These reforms, however, were not always paralleled by massive changes in the constitutional and political culture and the ‘law in action’, as predetermined by the socio-legal context. The chapter is structured in the following way. Section II briefly introduces the main characteristics of the Bulgarian constitutional system. It presents Bulgaria’s hybrid parliamentarism, which resulted from internalising achievements of European constitutionalism and European constitutional civilisation as well as creatively rethinking lessons from Bulgarian constitutional history. Special attention is devoted to key constitutional reforms and the Bulgarian constitutional culture. The latter is separately discussed in section III. There it is observed how that culture has been predetermined by the idea of ‘Europe’ as part of the normative ideology formed throughout the process of modernisation via Europeanisation in Bulgaria since the nineteenth century. Sections IV to VI focus on the European Union (EU) domain. Section IV presents the constitutional foundations of Bulgaria’s EU membership. Special emphasis is placed on the lack of direct democratic sanction for the EU membership and on the active and proactive role played by the Constitutional Court (BGCC) in paving the way for the EU integration of Bulgaria, contrasted 1 See Stephan Hobe, ‘Statehood at the End of the 20th Century – The Model of the “Open State”: A German Perspective’ Austrian Review of International and European Law No 2 (Leiden, Brill, 1997) 127.

34  Martin Belov with the lack of engagement of the BGCC with judicial dialogue.2 The analysis continues with critical assessments of the substance, scope, intensity and effects of the amendments related to Bulgaria’s EU membership which were introduced in the 1991 Bulgarian Constitution (the Constitution). Attention is devoted also to the EU integration clause in the Constitution and the failure of that clause to clearly set out the relationship between the Bulgarian and the EU legal orders. The BGCC’s stance on that relationship is also presented and criticised. The effects of EU law on the Bulgarian legal order and the constitutional rules and practice on implementing EU law are the object of in-depth analysis. Section V focuses on the Bulgarian constitutional limits to EU integration and to the influence of the EU law on the Bulgarian legal order. They are qualified as being largely part of the ‘missing constitution’ in Bulgaria. This is due to the fact that the Constitution, the legislation and the jurisprudence of the BGCC do not provide a clear and explicit system of ‘counter-limits’ to EU law and EU integration.3 Section VI discusses constitutional rules and practice on implementing EU law. Section VII provides some conclusions regarding the influence of supranational, international, and comparative legal standards on the Bulgarian constitutionalism and constitutional design. The challenges flowing from Bulgaria’s participation in forms of supranational constitutionalism are critically assessed and presented. Finally, suggestions are made with respect to future developments of EU integration and their impact on Bulgaria’s constitutional order.

II.  Main Characteristics of the National Constitutional System A.  Hybrid Parliamentarism Structured with Regard to Shadows from the Past and Hopes for the Future in a Partially Consolidated Democracy The Constitution is the fundamental law of Bulgaria, a post-communist country. It was created in the context of European optimism regarding globalisation and growing pressure for open statehood. The Constitution was also influenced by some national stigma and domestic political constraints. It was developed predominantly by reformist elements of the Bulgarian post-communist (and former communist) elites. The Constitution was a compromise between socio-paternalistic ideas of the state as an important factor for public well-being and the new (at the time popular) wave of globalism, liberalism and neo-liberalism. The Constitution is the outcome of the amalgamation of fears regarding the total state which was the outcome of monarchical authoritarian regimes, sometimes interrupted by authoritarian regimes of the prime ministers (1879–1944) and the communist regime (1944–89) in Bulgaria as well as the hopes for integration in the European community of democratic states ruled by law. The latter was institutionalised by Bulgaria’s membership in the Council of Europe (1992) and in the EU (2007). 2 See Orieste Pollicino, ‘European Judicial Dialogue and the Protection of Fundamental Rights in the New Digital Environment: An Attempt at Emancipation and Reconciliation. The Case of Freedom of Speech’, in S Morano-Foadi and L Vickers (eds), Fundamental Rights in the EU. A Matter for Two Courts (Oxford, Hart, 2015) 93. On the lack of engagement of the BGCC with judicial dialogue, see Mihail Vatsov, ‘European Integration Through Preliminary Rulings? The Case of the Bulgarian Constitutional Court’ (2015) 16 German Law Journal 1591. 3 See generally Giuseppe Martinico, ‘Complexity and Cultural Sources of Law in the EU Context: From the Multilevel Constitutionalism to the Constitutional Synallagma’ (2007) 8 German Law Journal 205 and Pietro Faraguna, A Living Constitutional Identity: The Contribution of Non-Judicial Actors, Jean Monnet Working Paper 10/15 (New York, NYU School of Law, 2015) 27.

Bulgaria  35 The Constitution is part of modern European constitutionalism with regard to its axiological core and institutional design. The constitutional axiology comprises constitutional values and constitutional principles. The preamble of the Constitution proclaims the following constitutional values: freedom, peace, humanism, equality, justice and tolerance.4 It also provides for the following constitutional principles: people’s sovereignty, separation of powers, political pluralism, rule of law, democracy and welfare state. Hence, one can conclude that the Constitution is a value-oriented one. Not only does it not limit its scope of application to the institutional design of public power and the constitutional status of citizens, but it also asserts an axiological stance on key issues with socio-legal implications. It should be mentioned that the BGCC has also made relatively frequent references to the constitutionally enshrined principles and values in its jurisprudence. The Constitution is reflexive. Its reflexivity consists in the creative rethinking of some of the historical challenges and mistakes made in previous Bulgarian Constitutions. For example, delegated legislation was frequently misused by the Bulgarian monarchs, and especially by Tsar Boris III in the interwar period. Moreover, under the 1947 and the 1971 Constitutions the Presidium of the National Assembly and the State Council were also the de facto centres of legislative power adopting primary norms via normative decrees. Hence, the current Constitution does not provide for any form of delegated legislation. Furthermore, the former lack of constitutionality review in Bulgaria led to systematic violations of the first Bulgarian Constitution, the 1879 Tarnovo Constitution. As a result, the 1991 Constitution established a strong constitutional court. The Constitution is also reactive because it rejects some important elements and principles of the institutional design and the constitutional axiology of the previous Bulgarian constitutions. With regard to the form of governance, it is reactive both to the constitutional monarchy established by the Tarnovo Constitution and to the Soviet-type people’s republic provided by the 1947 and 1971 Constitutions. The 1991 Constitution replaces the principles of unity of state power, supremacy of the Parliament, political monism through the leading role of the communist party, and ‘socialist legality’ provided by the 1947 and 1971 communist constitutions with the principles of separation of powers, democracy, political pluralism and rule of law. However, with regard to some institutions, human rights and constitutional principles, the Constitution serves as a bridge from the past to the future. Examples of such constitutional tradition and continuity are the unicameral structure of the Parliament, the unitary form of territorial distribution of state power, and the permanent presence (sometimes only on paper) of civil and political rights in all Bulgarian Constitutions. Bulgaria has a hybrid parliamentary system. This means that elements of semi-presidentialism are enshrined in its predominantly parliamentary republican institutional design. The parliamentary election of the Council of Ministers, the system of instruments for parliamentary control of the Government, the competence of the National Assembly (the Parliament) to pass a no-confidence vote or to reject a vote of confidence in the Government, and the rather weak presidential powers, are typical features of a parliamentary republic. Yet, the Council of Ministers cannot dissolve the National Assembly and there is incompatibility between an MP’s mandate and a ministerial position. These are two deviations from the classical parliamentary form of governance.5 The lack of competence of the Government to dissolve the Parliament can be 4 See Martin Belov, ‘The Preamble of the Constitution in the European Constitutionalism’ (2013) No 20 Revista General de Derecho Romano and Мартин Белов [Martin Belov], ‘Функции на конституционния преамбюл’ [‘Functions of the Constitutional Preamble’] in Право, управление и медии през XXI век [Law, Governance and Media in XXI Century] (Blagoevgrad, South-Western University Press, 2012) 207. 5 For the specific features of the Bulgarian hybrid parliamentarism see Пламен Киров [Plamen Kirov], Президентът в българския конституционен модел [The President in the Bulgarian Constitutional Model] (Sofia, Sibi, 2004) 112 f; and

36  Martin Belov explained by the history of the Tarnovo Constitution. Dissolution of Parliament was frequently misused by the monarch as a tool for establishing his centrality and dominance of political life.6 The current Bulgarian parliamentarism differs importantly from pure parliamentary republican systems in the enhanced status of the head of state. The Bulgarian President is elected directly by the people and has some important competences. Such competences are a moderate legislative veto, the right to approach the Constitutional Court and extensive patronage power.7 The latter consists of the right to appoint one third of the Constitutional Court judges, the right to appoint members of the Electronic Media Council and the Governing Council of the Bulgarian National Bank, the right to reject the appointment of the presidents of the Supreme Court of Cassation, the Supreme Administrative Court and the state prosecutor general, and the right to appoint and dismiss diplomats and high-ranking army officers as well as police officers on (effectively binding) proposal of the Government.

B.  The Judiciary as Focal Point of the Constitutional Debate and Permanent Object of Constitutional Reforms The Bulgarian judiciary has been a constant target of political critics for impeding the normal functioning of the rule of law. Bulgaria developed a continental European system of judicial power. It adopted the French and Italian models for governing the judiciary via a Supreme Judicial Council – a system that is widespread in Europe. The only formal and institutional remnants from the socialist period relate to the position of the state prosecutor general. This office has enhanced standing because it heads the prosecutors, in a context of a very hierarchical structure of the state prosecutors’ office and where the state prosecutors have the power to exercise general control for legality for example of the administrative acts. The Constitution structures the judicial power in Bulgaria in three main pillars. The first pillar consists of the courts, the second pillar of the state prosecutors, and the third pillar of the state investigators. The second and the third pillars were merged in 2009 into an enlarged state prosecutors’ office. The BGCC is an institution that is separate from the judiciary. The civil and criminal courts are defined as courts of general competence. There are also three types of specialised courts – the administrative courts, the military courts, which deliver justice in criminal matters related to the military, and the specialised criminal courts for organised crime. Any extraordinary courts are constitutionally prohibited. The civil and criminal courts operate at three instances – first, appellate, and cassation. The administrative courts, however, operate at two instances – first and cassation. The hierarchy of the Георги Близнашки [Georgi Bliznashki], Парламентарното управление в България [The Parliamentary Government in Bulgaria] (Sofia, University of Sofia St Kliment Ohridski Press, 2009) 21 ff. 6 See Martin Belov, ‘Establishment of the Modern Constitutional Principles in Bulgaria. The Tarnovo Constitution of 1879’ in M Stolleis (ed), Konflikt und Koexistenz. Die Rechtsordnungen Südosteuropas im 19. und 20. Jahrhundert, Vol 1: Rumänien, Bulgarien, Griechenland (Frankfurt a M, Vittorio Klostermann Verlag, 2015) 859; Martin Belov, ‘Executive Power Metamorphoses in the History of the Balkan Constitutionalism: Between the Imperatives of the Constitutional Teleology and the Potentialities of the Political Reality’, Romanian Journal of Comparative Law No 5 (Bucharest, Universul Juridic, 2014) 299; Мартин Белов [Martin Belov], ‘Конституционният статус на монарха в балканския конституционализъм’ [‘The Constitutional Status of the Monarch in the Balkan Constitutionalism’], in В Мръчков [V Mrachkov] (ed), 135 от приемането на Търновската конституция [135 Years of the Adoption of the Tarnovo Constitution] (Sofia, Sibi, 2014) 262. 7 See Martin Belov, ‘Separation of Powers reconsidered: a Proposal for a New Theoretical Model at the Beginning of the 21st Century’ in A Geisler, M Hein and S Hummel (eds), Law, Politics, and the Constitution. New Perspectives from Legal and Political Theory (Frankfurt a M, Peter Lang, 2014) 47.

Bulgaria  37 ordinary courts as well as of the two types of specialised criminal courts is topped by the Supreme Court of Cassation. The Supreme Administrative Court is the highest court in the field of administrative law. The structure of the state prosecutors’ office parallels the structure of the courts with the state prosecutor general at the top. The reform of the judiciary has been a hot topic in the Bulgarian constitutional discussion during the past 30 years, following the fall of communism. Many of the constitutional amendments introduced to the Constitution concerned the judiciary. There was a twofold impetus behind the constitutional reform of the judiciary – improving its performance and meeting the criteria for EU membership. Hence, the judicial reform in Bulgaria was accomplished also under the EU banner. The predominant criticisms of the Bulgarian judiciary focused on the following issues. The judiciary was believed to be slow and inefficient with alleged instances of corruption and nepotism. Furthermore, the administration and management of the judicial power was accused of being under political influence. The criticism of the Supreme Judicial Council’s structure and functioning led to an amendment of Article 130 of the Constitution in 2015. The Supreme Judicial Council was restructured into two main branches – one for the judges and one for the state prosecutors and the state investigators. The problems of the Bulgarian judiciary do not really stem only from its institutional design or the civil, criminal and administrative law procedures. While further debate and improvement of the legal basis of the judicial power may be useful, a real reform of the judiciary should target much more profound problems with empirical character. Fighting corruption and nepotism is just one issue to be addressed. Other issues which are underestimated determinants of the judiciary’s malfunctioning relate to legal education, legal culture, the magistrates’ mentality, and the overload of some courts and state prosecutors’ offices. Hence, a substantive reform of the Bulgarian judiciary must inevitably consist of both legislative amendments and educational reform.

III.  Constitutional Culture The Bulgarian constitutional system has developed through legal transplants predominantly from Western and Central Europe.8 In order to correctly understand it one should take into account the massive recourse to ‘Europe’. Europe was a source of general normative ideology, holistic ideal and a legal tradition inspiring Bulgarian legal modernisation since the re-establishment of Bulgarian statehood in 1878.9 In other words, Europe, in its manifestation as a geopolitical factor, cultural and legal tradition, shaped modern Bulgarian statehood. Europe impacted the formation of the Bulgarian constitutional culture and legitimised modernity as a new conceptual framework for the development of the Bulgarian society. During the first (1878–1944) and the third (1990 until now) waves of Bulgarian modernisation, the constitutional systems of some key European states served as sources of inspiration. The Tarnovo Constitution took over many of the characteristics of the 1831 Belgian Constitution and its Balkan variants – the 1864 Greek, 1866 Romanian and 1869 Serbian constitutions. Thus, the Tarnovo Constitution possessed the characteristics of a legal transplant. The same is true of the 1947 and 1971 constitutions which were transplants of the Soviet constitutional model.

8 See generally Alan Watson, Legal Transplants: An Approach to Comparative Law 2nd edn (Athens GA, University of Georgia Press, 1993). 9 See H Patrick Glenn, Legal Traditions of the World. Sustainable Diversity in Law (Oxford, OUP, 2000).

38  Martin Belov In contrast to the three previous Bulgarian constitutions, the current constitution is not the result of such direct legal transplantation. Rather, it reflects the Bulgarian constitutional tradition and tries to remedy past constitutional dysfunctionalities. The Constitution combines institutional design borrowed from European constitutions of both established democracies and democracies in transition with elements of the previous Bulgarian tradition. The Bulgarian constitutional legislator adopted mainly the normative models of Western and Central European states. An ideal constitution – a combination of normative ideologies and ideas that form the intellectual context for the proper application of the ‘law in books’ as well as the ‘factual constitution’, namely durable patterns of behaviour consistent with the spirit and the text of the constitution – proved impossible to achieve.10 They were to some extent simulated by the Bulgarian political and intellectual elite in the form of ‘invented tradition’ or conscious or implicit Western-like behaviour.11 Last but not least, the socio-cultural and anthropological preconditions for the normal functioning of democracy, rule of law and power polycentrism resulting in separation of powers and political pluralism have been established very slowly. Following the fall of communism, Bulgaria had the chance to become part of European constitutional civilisation and function largely as a modern state with some degree of substantive constitutionalism, democracy and rule of law. However, governance problems of constitutional significance abound. Bulgaria still lacks a sufficiently developed civil society. Public power is still the main formant of the society. State actors often predetermine the conditions for democracy and market economy through instruments of nepotism and dependency of the formally private economy on national and European public funding. Legal modernisation in Bulgaria, frequently a result of purely elitist decision-making and acceptance of foreign legal traditions, produced a discrepancy between the official constitutional order and the unofficial but factual constitutional culture. The power conception and scheme in Bulgaria is characterised by huge patronage power concentrated in unofficial elites. The concept of statehood is based much more on a network of inherited and negotiated personal and group relations than on the formality of the state and law as abstract, rational and universal constructions. The constitutionally proclaimed market economy is functioning as a semi-dispersed state-dependent capitalism. Democracy is rather unstable and often fictitious. The discrepancy between the formal (official) and the informal (practised) Bulgarian constitutional law reflects its partial fictitiousness. For example, adopting the Italian and French model of judicial power governance via a Supreme Judicial Council and integrating the state prosecutor’s office in the judiciary did not automatically produce impartial, depoliticised and efficient justice in Bulgaria. Another example is adopting the parliamentary-republican model of representative democracy, which frequently functioned in practice as a form of oligarchical system masquerading as parliamentarism. This was due to the weak leadership of political parties and the underdeveloped civil society. However, it has to be recognised that Bulgaria has achieved real progress since the beginning of its transition to European modernity in the late nineteenth century. The very idea of the state and law as rational and systematic phenomena is a Western European transplant, and has been to some extent successfully adapted to the Bulgarian reality. The very belief in constitutional reforms as tools for shaping the political future of the society is proof of that fact. Moreover, 10 For the concept of ideal constitution see, Евгени Танчев [Evgeni Tanchev], Въведение в конституционното право [Introduction to Constitutional Law] (Sofia, Sibi, 2003) 112 ff with further references. 11 See Diana Mishkova, ‘Domesticating Modernity: Transfer of Ideologies and Institutions in Southeastern Europe’, in M Stolleis (ed), Konflikt und Koexistenz. Die Rechtsordnungen Südosteuropas im 19. und 20. Jahrhundert, Vol 1: Rumänien, Bulgarien, Griechenland (Frankfurt a M, Vittorio Klostermann Verlag, 2015) 723.

Bulgaria  39 the cultural code of European constitutional civilisation has been deeply and durably enshrined in the Bulgarian constitutional identity. Bulgaria has legal professions (magistrates, attorneys at law and other legal practitioners as well as legal scholars), with predominantly European attitudes towards the legal system and the different legal discourses (legislative, judicial etc). Last but not least, the Bulgarian constitutional system and specifically the Constitution are genuinely European. They comply with the general standards of power polycentrism, rule of law and democracy in the EU. Hence, Bulgaria is a peripheral European constitutional system in which Western legal ideas and institutions coexist with South European and post-communist attitudes and behavioural schemes. Adopting European constitutional prototypes has also had a formative function on the Bulgarian system. It led to the establishment of semi-democratic constitutionalism with many instances of real and functioning rule of law. All essential elements of the constitutional principles of democracy and rule of law are provided by the Bulgarian Constitution and legislation. There is also increasing awareness of the policy actors and citizens of the content and meaning of democracy and rule of law. Nevertheless, there are still socio-political factors hindering the development of a fully fledged democracy and rule of law in Bulgaria, which are discussed elsewhere in this chapter. Thus, for Bulgaria the glass is rather half-full than half-empty. Bulgarian constitutionalism started its development as a system of legal transplants introduced in the late medieval Bulgarian society bounded in the oriental context of the Ottoman Empire. Its proper development was subsequently stalled for almost half a century by the communist regime. Recently, the Bulgarian society has been able to relatively swiftly adjust to the challenges of democratisation and Europeanisation. It formally, and to some extent also substantively, managed to fulfil the requirements of European modernity in general and of EU integration in particular.

IV.  Constitutional Foundations of EU Membership A.  The Lack of Direct Democratic Sanction for the EU Membership of Bulgaria Bulgaria was one of the very few states to join the EU without any form of direct democratic approval. There was neither a referendum on the constitutional amendments related to the EU membership nor a referendum on the ratification of the accession treaty. There was not even an indirect sanction by the people (for example in the form of parliamentary elections with a preceding campaign on the EU accession and membership). There are several reasons that could explain the elitist way in which Bulgaria became part of the EU. These are mainly the weak direct democratic traditions of Bulgaria, the elitist patterns of policy-making, the rush in which the Bulgarian political class worked for EU membership and the fear of sudden anti-European changes in the mood of the Bulgarian society. Nevertheless, the Zeitgeist and opinion polls suggest that the predominant part of the Bulgarian society was indeed in favour of EU accession. No referendum on EU matters has been held since Bulgaria’s EU accession. This can be explained by several factors. No major reform of primary EU law took place after the Lisbon Treaty, on which Bulgaria took a stance during its pre-accession negotiations. Again, this was done without any form of consultation with the people. Moreover, Bulgaria is not part of the eurozone and still ranks relatively well in terms of debt levels and budgetary stability. Thus, Bulgaria was neither forced to implement the Treaty on Stability, Coordination and Governance nor to

40  Martin Belov take drastic steps such as austerity measures to reduce and control the state debt and achieve budgetary stability.12 Last but not least, currently the ‘permissive consensus’ for remaining part of the EU is rather high and there are no strong anti-European parties and movements that could be possible initiators of a ‘BulgarExit’ referendum. The concerns regarding the weak direct democratic legitimacy of Bulgarian EU membership are still predominantly theoretical and much more a matter of principle than having any practical importance. However, they should not be neglected. What is especially problematic is the fact that the Bulgarian Direct Participation of the Citizens in the State Power and in the Local Self Governance Act13 (the Referendums Act) prohibits constitutional referenda related to the most important elements of the constitutional design that are listed in Article 158 of the Constitution as well as referenda for denouncement of treaties. There is a peculiar combination of two factors which have led to a situation that can be defined as ‘elitist normative entrapment of the demos’. The first factor is the representative democratic way Bulgaria joined the EU. The second factor concerns the prohibition of the direct democracy as a path for eventual amendment of the fundamental constitutional provisions directly or indirectly related to the participation of the state in the EU and for termination of Bulgaria’s EU membership. The entrapment is normative because it is based on an Act of Parliament with doubtful constitutionality. This is so because either all types of constitutional amendments approved through a referendum should be deemed prohibited due to the fact that the direct democratic instruments are not provided as a tool for constitutional amendment by Chapter IX of the Constitution or they should all be permitted on the basis of the principle of people’s sovereignty proclaimed in Article 1, paragraph 2 of the Constitution. Accordingly, it is unclear whether future constitutional amendments related to profound changes in the EU institutional design should be done only through means of representative democracy or direct democracy instruments such as constitutional people’s initiative and constitutional referendum may be used as well. The Constitution, by proclaiming in Article 1, paragraph 2 the principle of people’s sovereignty and the permissibility of accomplishment of the public power directly by the people, is open in principle to direct democracy. However, there are no constitutional provisions on referenda or people’s initiatives. The problem is even greater. An interinstitutional battle may arise if the BGCC declares unconstitutional a decision of the National Assembly to convene a referendum after the referendum was held. This may evolve in battles between the National Assembly and the BGCC as well as in direct clash between the BGCC as genuinely elitist institution whose activity is justified through output legitimacy14 and the people as masters of their sovereignty in the context of supranational and multilevel constitutional games.15 This kind of clash between policy actors may be viewed also as conflict between the principles of people’s sovereignty and democracy on the one hand and rule of law, on the other.

12 For the lack of democratic involvement in the management of financial crisis see Mihail Vatsov, Constitutional Change Through Euro Crisis Law: Bulgaria (2015), www.academia.edu/11550630/Constitutional_Change_Through_ Euro_Crisis_Law_Bulgaria. 13 Закон за пряко участие на гражданите в държавната власт и местното самоуправление, Държавен Вестник [Dăržaven Vestnik – State Gazette, the official journal of the state (DV)] 44/2009. 14 See Paul Craig, Gráinne de Búrca, EU Law: Text, Cases and Materials, 4th edn (Oxford, OUP, 2008) 65 and Philipp Kiiver, ‘European scrutiny in national parliaments. Individual efforts in the collective interest?’, in J O’Brennan and T Raunio (eds), National Parliaments within the Enlarged European Union. From ‘victims’ of integration to competitive actors? (Abingdon, Routledge, 2007) 95. 15 See George Tsebelis, Nested Games. Rational Choice in Comparative Politics (Berkley/CA, University of California Press, 1990).

Bulgaria  41 The second problem is the prohibition in the Referendums Act on referenda for denouncement of treaties. Thus, Bulgarians are ‘stuck’ and bound to be EU citizens as long as the EU exists, without any chance to validly express their will on that matter. According to the National Assembly, Bulgaria can exit the EU only on the will of its MPs but never on the will of the people as expressed directly in a referendum. This legislative concept is based on an unjustifiable asymmetry from democratic point of view. It is difficult to explain why the people are allowed to vote in a referendum for ratification but not for denouncement of a treaty. The prohibition of referenda for denouncement of treaties may be unconstitutional due to breach of the people’s sovereignty principle. Any such conclusions on constitutionality, however, can only be tentative due to the bare existence of provisions on direct democracy in the Constitution. Moreover, it is unpredictable what would be the reaction of the political elite and specifically the National Assembly and the BGCC in case of a successful referendum on a treaty denouncement. If it is sufficiently representative of the will of the majority of the people a choice will have to be made between recognition and non-recognition of an illegal and doubtfully constitutional but popular referendum. Thus, a choice between the principles of rule of law and people’s sovereignty will have to be made.

B.  The Role of the Constitutional Court in Paving the Way for EU-related Constitutional Amendments The introduction of constitutional amendments related to Bulgaria’s EU membership was preceded by intense constitutional dialogue between the BGCC and three other institutions. The state prosecutor general, the President, and the Supreme Court of Cassation approached the Court with a demand for interpretative decisions designed to clarify in advance issues pertaining to Bulgaria’s EU membership. Such inter-institutional dialogue was necessary due to the two-track procedure for constitutional amendment. Amending foundational tenets of the Constitution (listed in Article 158) such as the form of governance, the form of territorial distribution of power, the supremacy of the constitution, the primacy of the international treaties over the sub-constitutional domestic legislation and the irrevocability of the human rights or drafting a new Constitution are in the competence of a special body, the Grand National Assembly. Constitutional amendments beyond the scope of Article 158 of the Constitution can be passed by the National Assembly acting through increased formal and procedural hurdles. The BGCC issued three interpretative decisions – Decision no 3 of 2003,16 Decision No 3 of 200417 and Decision No 8 of 200518 related to two fundamental issues. The first issue was procedural. It concerned the structure of the constituent power in Bulgaria. The BGCC clarified which institution is competent to introduce the necessary constitutional amendments for Bulgaria’s EU membership – the Grand National Assembly or the National Assembly. The second issue was substantive. It concerned the question whether EU membership would introduce profound structural changes to the Bulgarian constitutional design or, instead, its influence would only be functional or otherwise related to structural adjustments that would only be of secondary importance.

16 DV 36/2003 (all BGCC decisions are published in DV as well as on the BGCC’s web page www.constcourt.bg. Unfortunately, both are in Bulgarian only). 17 DV 61/2004. 18 DV 74/2005.

42  Martin Belov In Decision No 3 of 2003 the BGCC defined what form of governance and form of territorial distribution of power were established by the Bulgarian constitutional order. This was a quite challenging task for two reasons. First, this was an obvious case of judicial activism in which the BGCC acted as a quasi-constitutional legislator ascribing meaning to fundamental constitutional concepts.19 Such activism was met with no criticism and even with no real debate in the Bulgarian academic and political discourse, which itself seems to be problematic. The choice of the constitutional drafters to define the competence of the Grand National Assembly in such broad and abstract terms opens the path for their context-dependent interpretation by the BGCC allowing the Court to function as de facto institution of the constituent power.20 Second, the BGCC indeed made an excessive use of the opportunity to function as an ‘interpretative Leviathan’ defining the fundamental structural concepts of the constitutional order. It established a detailed list of elements which, according to the Court’s opinion, appertain to the form of governance: The form of governance should be interpreted extensively. This concept not only defines the form of the state as parliamentary or presidential republic or monarchy. It also includes the whole system of high state institutions established by the Grand National Assembly through many provisions that further develop the parliamentarism … The form of governance includes the very existence of these institutions, their allocation in the respective state power, their organisation and the conditions and ways for formation of their mandate. The form of governance includes also the competences and the activities of these institutions as granted by the Constitution in case their amendment will distort the balance between them and the main principles on which the state is based – people’s sovereignty, supremacy of the Constitution, political pluralism, separation of powers, rule of law and independence of the judiciary.21

The BGCC explicitly opted for judicial activism. At the same time, it confused the constitutional framework of the institutional design enshrining the concept of the form of governance with the main constitutional principles and with the meso- and micro-institutional design concerning the specific competences of the state institutions and the procedures for the election or appointment of their office holders. Moreover, it is clear that such an extensive interpretation of the form of governance would in fact render impossible almost any constitutional amendment by the National Assembly. Instead, a Grand National Assembly must be convened which is politically a very difficult task. Hence, the impossibility of the National Assembly to introduce the necessary constitutional amendments would have impeded the Bulgarian accession to the EU. That is why in less than a year the BGCC adopted Decision No 3 of 2004. It was another example of judicial activism and an instance of a breach of the principle of continuity and predictability of the jurisprudence of the BGCC, often considered as part of the principle of rule of law. Decision No 3 of 2004 clearly departed from the general idea of Decision No 3 of 2003. According to the Court, the primacy of EU law would not challenge the supremacy of the Bulgarian Constitution. Consequently, there was no need to convene a Grand National Assembly in order to allow for the primacy and the direct effect of EU law. The BGCC found the fact that Bulgarian institutions will participate in the adoption of EU acts to be a sufficient safeguard for the supremacy of the Constitution. Moreover, the BGCC equated primary EU law with classical treaties which are 19 See Noah Berlatsky, Judicial Activism (Opposing Viewpoints) (New York, Greenhaven Press, 2011). 20 It seems that the BGCC may play an important role in the future in the context of Bulgarian EU membership. Such instances of decisive BGCC’s influences may be related not only to the abstract interpretations of the constitutional provisions but also to the control for constitutionality of acts of Parliament implementing EU related measures. So far the BGCC has not engaged in judicial dialogue with the CJEU, eg by approaching it in the context of a preliminary reference procedure. 21 Decision No 3 of 2003 of the Bulgarian Constitutional Court available at: http://constcourt.bg/bg/Acts/ GetHtmlContent/4bc826fe-aa52-4590-8c0e-3ab698cd272a.

Bulgaria  43 ratified, published, and entered into force. Such treaties enjoy primacy over sub-constitutional legislation but not over the Constitution itself. Subsequently, the BGCC upheld its stance on the sub-constitutional status of treaties by accepting to exercise legality control over the acts of Parliament in light of primary EU law as part of the abovementioned category of instruments. Thus, the BGCC tacitly disregarded the case law of the Court of Justice of the EU (CJEU) (Costa v ENEL,22 Van Gend en Loos23 and Internationale Handelsgesellschaft24) by implicitly denying any primacy of EU law over the Bulgarian Constitution. It must be noted that the BGCC is yet to enter into a judicial dialogue with supranational courts and especially the CJEU.25 Moreover, the BGCC has also not considered the matter of ‘constitutional identity’.26 Doing so could clarify the relationship between the Constitution and EU law in a more moderate and flexible way that would also be appropriate in the context of supranational constitutionalism and constitutional pluralism.27 Furthermore, the BGCC insisted that the EU membership of Bulgaria would not change the form of territorial distribution of power. The Court argued in Decision No 3 of 2004 that ‘the EU is neither a federation nor any other form of statehood. The EU membership of Bulgaria does not change the elements which characterise the form of territorial distribution of the state power’. In neither of the two Decisions (Decision No 3 of 2003 and Decision No 3 of 2004) did the BGCC define these elements. Decision No 3 of 2003 just stipulates that the form of territorial distribution of state power comprises the territorial integrity and the unitary nature of the state with prohibition of territorial autonomies. In fact, there is no constitution either of a unitary state or of a federation which does not directly or indirectly provide for safeguards for the territorial integrity. Moreover, the mere repetition of the constitutional provision of Article 2, paragraph 1 of the Constitution, according to which Bulgaria is a unitary state with local self-governance in which any territorial autonomy is prohibited, cannot count as definition or element of the form of territorial distribution of state power. This analysis of Decision No 3 of 2004 leads to the conclusion that the BGCC believes that the form of territorial distribution of state power would be affected only if there are changes in the territorial structure of the state below the national level which lead, for example to the federalisation or confederalisation of Bulgaria or to the establishment of territorial autonomous regions. The BGCC focuses its analysis on the subnational territorial structure of the state. Following that line of argumentation one can come to the rather strange conclusion that the eventual accession of Bulgaria to a federation or a supranational system with federal, quasi-federal or confederal character will not constitute a change in the form of territorial distribution of power. In fact, the BGCC combines two rather contradictory concepts in its reasoning on the form of territorial distribution of power. On the one hand, it focuses on the national and subnational levels of the territorial structure of the public power. Thus, it perceives statehood as a ‘closed territorial container’, which is an outdated concept and analytical perspective.28 On the other 22 ECJ 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66. 23 ECJ 05.02.1963 26/62 (Van Gend en Loos v Administratie der Belastingen) ECLI:EU:C:1963:1. 24 ECJ 17.12.1970 11/70 (Internationale Handelsgesellschaft mbH v Einfuhr- und Vorratsstelle für Getreide und Futtermittel) ECLI:EU:C:1970:114. 25 See Pollicino, ‘European Judicial Dialogue’ (fn 2) 93 ff; David Law, Wen-Chen Chang, ‘The Limits of Global Judicial Dialogue’ (2011) 86 Washington Law Review 523; and Melissa Waters, ‘The Future of Transnational Judicial Dialogue’ (2010) No 104 American Society of International Law, Proceedings of the Annual Meeting 465. 26 See Alejandro Saiz Arnaiz and Carina Alcoberro Llivina (eds), National Constitutional Identity and European Integration (Mortsel, Intersentia, 2013). 27 See Evgeni Tanchev, The Treaty on the Constitution for Europe and National Constitutions in the Age of Constitutional Pluralism, in L Breitner-Czuma (ed), Das alte Europa in neuer Verfassung? (Baden-Baden, Nomos, 2007) 83, 88 ff. 28 See Neil Brenner, ‘Beyond state-centrism? Space, Territoriality and Geographical Scale in Globalization Studies’ (1999) 28 Theory and Society 39, 55.

44  Martin Belov hand, the BGCC’s decision is very generous and permissive towards the participation of the state in supranational constitutional systems. The BGCC uses the concept of ‘open statehood’ very broadly and believes that the fact that Bulgaria is represented in the EU institutions and can participate in the decision-making of the EU quasi-automatically serves as a sufficient safeguard of the Bulgarian sovereignty and the supremacy of the Constitution. The result is opening the way towards swift EU integration of Bulgaria through the accomplishment of the necessary constitutional amendments by the National Assembly but also the ultimate submission of EU law under the Constitution. Thus, the pro-European stance of the BGCC is counterbalanced with its preference to consider EU law as classical international law. The last conclusion is rather contentious in the context of the current stage of the EU integration. Furthermore, Decision No 3 of 2004 provides that a constitutional amendment allowing for extradition of Bulgarian citizens to a foreign state or to an international court does not concern either the form of governance or the form of territorial distribution of power. Hence, such amendment can be accomplished by the National Assembly. This obvious conclusion does not take into consideration the fact that the provision of extradition in fact contravenes the irrevocability of fundamental rights of the Bulgarian citizens provided by Article 57, paragraph 1 of the Constitution.29 Indeed, this is not an absolute irrevocability since it can be removed by the Grand National Assembly. However, the provision of extradition, for example to the International Criminal Court or the general permission of European arrest warrants has had to be accomplished by the Grand National Assembly, but was adopted by the National Assembly. In this case, the BGCC cannot be blamed for not protecting the irrevocability of the constitutional rights because it has not been approached with such a demand and the Court cannot act ex officio. However, Decision No 3 of 2004 has been implicitly used by the National Assembly as permission for introducing important exceptions to the constitutional prohibition on extraditing Bulgarian citizens. Last but not least, the BGCC stipulates in Decision No 3 of 2004 that the acquisition of EU citizenship by the Bulgarian citizens does not concern the form of territorial distribution of power. Indeed, at first glance the citizenship concerns the personal and not the territorial element of the statehood, according to the classical and still influential theory of Georg Jellinek.30 The BGCC does not take into consideration the fact that ‘two track citizenship’, or in other words two parallel citizenships allocated on two levels of the territorial structure of the public power, can be found only in federations and quasi-federal systems, with the EU being still the only example of the latter. Decision No 8 of 2005 finally cleared the way for the National Assembly to adopt the EU membership-related constitutional amendments. This decision concentrated on possible constitutional amendments of Chapter VI of the Constitution, which is devoted to the judiciary. Constitutional reform of the judiciary was perceived as a necessary precondition for EU membership of Bulgaria despite the fact that the normative model of the judicial power did not formally infringe the acquis communautaire. With Decision No 8 of 2005 the BGCC effectively approved the introduction of a majority of the constitutional reforms of the judiciary. According to that decision, constitutional reforms comprising the constitutional proclamation that the courts are the main bearers of the judicial power, the restructuring of the state prosecutor’s office by defining the state accusation in criminal cases in front of the courts as its main function, the restructuring



29 Art 30 See

57 para 1 of the Constitution provides that the basic rights of the citizen are irrevocable. Georg Jellinek, Allgemeine Staatslehre (Charleston NC, Nabu Press, 2011).

Bulgaria  45 of the state investigator’s office, the provision of investigative magistrates and the proclamation that the state prosecutor’s office, the state investigator’s office and the police, accomplish a single state policy for combating criminal activity, supervised by the National Assembly, that can be adopted by the latter. The BGCC clarified that such constitutional amendments must not infringe the balance between the institutions and the main constitutional principles. The conclusion that can be drawn is that the BGCC played an important and even crucial role for the EU accession of Bulgaria.31 Its jurisprudence was inconsistent and featured judicial activism.32 This activism redrew the lines between (1) constituent and constituted powers and (2) the different institutions and procedures exercising the constituent power. However, from the point of view of pro-European constitutional politics, the case law of the BGCC was indeed the only chance for a swift and smooth introduction of the constitutional amendments that were perceived as necessary for Bulgaria’s EU membership. The ‘curving’ jurisprudence and in some instances substantively unpersuasive decisions of the BGCC were partially the result of its manoeuvring through the jungle of interests of the political stake-holders. Thus, the inconsistency of the BGCC’s jurisprudence was partly the product of avoiding being taken hostage by particular domestic interests, albeit not antiEuropean, that could have undermined Bulgaria’s EU membership.

C.  The Substance, Scope, Intensity, and Effects of the Constitutional Amendments Related to the Bulgarian EU Membership The constitutional amendments relating to Bulgaria’s EU membership can be divided into two main groups. The first group covers EU membership-specific amendments that are necessary minimum requirements by imperative provisions of EU law and the acquis communautaire. These are the provisions on (1) EU integration, (2) the right of EU citizens to acquire land in Bulgaria, (3) compulsory parliamentary ratification of treaties that transfer competences stemming from the Constitution to the EU by a two-thirds qualified majority, (4) National Assembly competences to be informed of and exercise control over the Government’s activity at the EU level, (5) electoral rights of EU citizens with respect to EU parliamentary and local elections, and (6) extradition to states and international bodies on the basis of an international treaty that is ratified, promulgated, published and entered into force for the Republic of Bulgaria. Although the EU membership amendments concern numerous constitutional texts, they are incremental because many of them are of secondary importance and a result of political cautiousness. The second group covers amendments reforming the judiciary not only in the context of EU integration but also related to participation in other supranational organisations such as the Council of Europe. Many of these amendments were adopted in dialogue with the European Commission and the Venice Commission for Democracy through Law of the Council of Europe. The very process of amending the Constitution is dominated by an approach that can be defined as constitutional gradualism.33 This means that almost all constitutional amendments introduced so far concern mainly issues of secondary importance. The big exceptions are the EU integration

31 See Martin Belov, ‘Constitutional Courts as Ultimate Players in Multilevel Constituent Power Games: the Bulgarian Case’, in M Belov (ed) Courts, Politics and Constitutional Law. Judicialization of Politics and Politicization of the Judiciary (Abingdon, Routledge, 2019) 152. 32 See Martin Belov and Aleksandar Tsekov, ‘The Constitutional Court of the Republic of Bulgaria as Law Maker’, in M Florczak-Wator (ed) Judicial Law-Making in European Constitutional Courts (Oxford, Routledge, 2020) 91. 33 See Evgeni Tanchev and Martin Belov, ‘Constitutional Gradualism: Adapting to EU Membership and Improving the Judiciary in the Bulgarian Constitution’ (2008) 14 European Public Law 3.

46  Martin Belov clause and the possibility for extradition of Bulgarian citizens. In these fields, there are several reasons for a conservative and restrained approach of the Bulgarian constitutional legislator. The Bulgarian political elite still adheres to the general appropriateness of the constitutional model adopted in 1991. Consequently, it believes that it is not necessary, and might even be dangerous, to introduce radical reforms. This belief is rooted in the fact that the deficiencies of the Bulgarian constitutionalism concern much more ‘the law in action’ than the ‘law in books’. In other words, the main dysfunctionalities of the Bulgarian constitutional model stem out of the weakness of the civil society, its dependence on public power the lack of real and idea-based political parties resulting in a crisis of representation, the lack of independent media, the low quality of substantive education, poor professional culture, and lack of calling in the legal profession. Moreover, Bulgaria features oligarchic capitalism dependent on national and EU funding rather than market economy. Hence, radical constitutional reforms are not only unnecessary, but may actually reinforce the position of oligarchical elites by creating institutional opportunities for concentration of power (for example by adopting a superpresidential republic) or for oligarchic empowerment (for example by establishment of a second chamber which is not directly elected by the people). Another reason for the use of constitutional gradualism is the lack of political consensus for more profound constitutional reforms. There is a permanent debate on the reform of the judiciary. However, there is no systematic and in-depth analysis of current constitutional problems or future challenges emerging from EU membership. Moreover, Bulgaria has a generally pro-European stance and is rather swiftly and even uncritically transplanting institutions and provisions with EU origin. However, there is insufficient public and expert discussion and professional competence when it comes to fundamental issues relating to the influence of the EU supranational constitutionalism on Bulgarian constitutionalism. Bulgarian institutions and officials try to avoid taking a stance on pressing issues of current European constitutionalism. Examples of such issues are the hierarchical or any other possible interrelation between the Constitution, treaties and EU law, the lack of any consistent and implicit concept of counter-limits and constitutional identity and the pale and purely formal parliamentary control over the governmental activity at EU level. The failure of Bulgarian institutions to engage in debates on these issues, to make full use of their institutional capacities, and to provide possible solutions clearly demonstrate the cautious attitude of the Bulgarian political elite in general and the constitutional and ordinary legislator as well as the BGCC in particular in taking a more proactive stance on the constitutional role of Bulgaria in the EU. It is interesting to note that the very activist pre-accession behaviour of the BGCC on EU membership-related issues has not been replicated with respect to issues of constitutional identity, the substantive, formal, and procedural limits set by the Constitution on EU integration, and the influence of EU law on the Bulgarian legal order.

D.  The EU Integration Clause, the Principle of Open Statehood and the Concept of ‘Just International Order’ The EU integration clause was included in the Constitution in 2005 with a new paragraph 3 to Article 4. The paragraph provides that ‘The Republic of Bulgaria shall participate in the construction and the development of the EU.’ This rather austere and modest provision seems to have declarative and even decorative role. The constitutional legislator chose neither to assemble all provisions relating to the EU membership in one constitutional chapter nor to provide for extensive and detailed EU integration clause. Consequently, the constitutional provisions relating to

Bulgaria  47 the EU membership are limited to the bare minimum required by EU law. These provisions are allocated in different parts of the Constitution in accordance with the sedes materiae criterion. Despite its concise formulation, the EU integration clause serves three very important functions – opening, linkage and empowering functions. The opening and the linkage functions may actually be perceived and treated as two dimensions of a single broader function for the ordering of legal orders.34 The EU integration clause opens the Bulgarian legal order to the EU supranational constitutionalism. It allows for the integration of EU law in the Bulgarian legal order and may be used as an argument for the primacy of EU law. It also opens the Bulgarian legal order to both the axiology of the EU constitutionalism and to its pragmatics. Furthermore, the EU integration clause links the EU legal system with the Bulgarian legal system on normative, institutional, and functional levels. It serves as a linkage between the normative orders, as bound between axiological orders and as a link between institutional systems functioning in multi-layered normative and socio-legal context. The EU integration clause empowers the Bulgarian institutions to participate in the EU’s institutional infrastructure and decision-making. This is especially important and obvious with respect to the work of the Council of Ministers, the Prime Minister, the ministers, and the National Assembly. However, the empowering function of the EU integration clause has its implications also for the BGCC and the other courts by permitting their participation in judicial dialogue in general and in the preliminary reference procedure in particular. It is interesting to note that the BGCC in the motives to its Decision No 3 of 2004 anticipated the need to legitimise the EU integration clause included in the Constitution in 2005. The BGCC determined in advance the constitutional basis for the EU integration of Bulgaria to which the EU integration clause acts as a continuation and a further development. The BGCC exposed the constitutionally proclaimed engagement of Bulgaria in the establishment of a ‘just international order’ as a constitutional principle. In particular, the BGCC held that the current Constitution explicitly enshrines as one of the basic principles on which the state is founded the principle of engagement of the Republic of Bulgaria in the establishment of just international order (Article 24, paragraph 2). This constitutional provision for the achievement of a just international order sets the framework and the ground for the accession of the state to a united Europe and for allowing of the supranational, direct and universal effect of its legal acts towards the Republic of Bulgaria. The provisions of Article 5, paragraph 4, Article 85, paragraph 2 and Article 149, paragraph 1, point 4 of the Constitution are expression of that idea. Hence, a conclusion can be made for the openness of the Bulgarian statehood towards international law with the possibility of constitutional empowering to take actions and accomplish measures for the achievement of a united Europe.

Achieving a ‘just international order’ cannot be a fundamental constitutional principle. It should be viewed as either a constitutional goal or as a sector principle of the international policy, because it is defined by the Constitution as one of the goals of Bulgaria’s international policy. Moreover, it has never been defined as fundamental constitutional principle in comparative or in comparative historical perspective. Last but not least, achieving a ‘just international order’ does not possess the legal features of a general constitutional principle, which would be applicable to all or at least to most of the spheres of constitutional law.

34 See Евгени Танчев [Evgeni Tanchev], ‘Съвременният наднационален конституционен плурализъм или подреждането на конституционните правопорядъци’ [‘The Contemporary Supranational Constitutional Pluralism or the Ordering of Legal Orders’], in Пл Киров (ed), Конституционнноправни изследвания. 2012–2013 [Constitutional Studies 2012–2013] (Sofia, University of Sofia St Kliment Ohridski Press, 2013) 171.

48  Martin Belov However, one can speculate that, by proclaiming the achievement of a ‘just international order’ as a fundamental constitutional principle, the BGCC aimed at constructing a traditional legitimation background for the EU membership of Bulgaria in general and for the EU integration clause in particular. Such speculative assumption is hard to prove. However, the BGCC has to some extent objectively rooted the EU integration of Bulgaria and the EU integration clause in a broader concept – the pursuit of a ‘just international order’ presumably aimed at by the authors of the Constitution. In that case the linkage of the Bulgarian engagement with ‘united Europe’ – later institutionalised as ‘construction and development of the EU’ provided by the EU integration clause, through the achievement of a ‘just international order’– is proactively constructed by the BGCC. This is another example of EU-related judicial activism of the BGCC.

E.  Effects of EU Law on the National Legal Order The National Assembly has taken a very pro-European stance on the political rights of EU citizens. Bulgarian law goes beyond the minimum level of political rights granted to EU citizens by EU law and the Constitution. This minimum consists of active and passive electoral rights for the European Parliament and for local self-governance institutions, as well as the right to participate in European citizen’s initiatives. EU citizens also enjoy other rights related to the deliberative and participatory democracy such as the rights to petition, assembly, and form and join NGOs. In Bulgaria, everyone enjoys these rights, not only EU citizens. Since the 2009 amendment of the Political Parties Act35 EU citizens may form or join Bulgarian political parties. There are two possible rationales in granting this right to foreigners and, in particular, to EU citizens. First, the political parties are the main engines of the representative democracy and the key channels and instruments for policy-making. Thus, the formation and participation in political parties should be reserved to members of the political community, that is to say to the national citizens. Second, it is very difficult to effectively participate in any elections without political party support. Hence, the above-mentioned electoral rights of EU citizens would remain largely fictitious if they were not supplemented by the right to organise and join a political party. Since 2009 the Bulgarian legislator preferred to follow the second rationale and empower EU citizens, ignoring arguments based on the sovereignty principle. This legislative option can be considered as a further step by Bulgaria towards the concept of open statehood. The Referendums Act grants EU citizens the right to participate in national and local citizens’ and people’s initiatives in Bulgaria. The right is conditioned on them having a permanent address somewhere in Bulgaria in the case of national initiatives and in the respective administrative unit in the case of local initiatives. Hence, the Bulgarian legislator further enfranchised EU citizens with these important forms of direct and participatory democracy on the input side of the governance process which concern the decision on substantive issues.36 On the other hand, the Referendums Act does not permit EU citizens to participate in national or local referenda. The reason is that the referendum is a form of direct democracy on the output side of the governance process which concerns the decision on substantive issues. Consequently, the referendum is a device for exercising sovereignty which remains reserved to Bulgarian citizens only.

35 DV 39/2009. 36 See Мартин Белов, [Martin Belov] Пряка демокрация – национални и сравнителноправни аспектил [Direct Democracy – National and Comparative Aspects] (Sofia, Sibi, 2009) 189 ff.

Bulgaria  49

V.  Constitutional Limits to EU Integration A system of explicit constitutional limits to EU integration and EU law is part of the ‘missing constitution’ in Bulgaria. Bulgaria does not have an extensive historical tradition of participating in supranational organisations. After World War II, Bulgaria was allocated to the so-called ‘Eastern Bloc’ of communist countries in Central and Eastern Europe led by the Soviet Union, though it was never part of the Soviet Union itself. Relations in that bloc were institutionalised in the form of international organisations that never developed into a supranational legal order. The domination of the Soviet Union in that framework was so clear and superimposed that developing any form of constitutional limits to the influence of the international organisations for cooperation between the communist regimes was clearly impossible. The communist constitutions of 1947 and 1971 were proclaimed as the supreme law of the state, but in fact they were rather fictive. Their formal supremacy was not a barrier to the will of the leaders of the Eastern Bloc in general or of the Soviet leadership in particular. Following the fall of communism, Bulgaria’s integration in the system of supranational and international forms of cooperation of the European democracies was welcomed as almost the indisputably right choice. Bulgaria’s membership in the Council of Europe (1992) and the EU (2007) and to a lesser extent in NATO (2004) has been paralleled by permissive consensus. The general belief was that European integration is the most appropriate form of modernisation of the Bulgarian state and society. Hence, there was no real discussion either in the scientific or in the socio-political social discourses on possible limits to the influence of supranational legal orders in general and EU law and its constitutionalism in particular on the Bulgarian legal and constitutional order. Another reason for tolerating greater influence of EU law within the Bulgarian legal system is the low level of nationalism in contemporary Bulgarian society. This contrasts with other postcommunist countries such as Poland, Hungary and Romania. The rise of nationalism in those countries was partially a reaction to the factual loss of sovereignty during the communist period. Indeed, Bulgaria also experienced such sovereignty loss. However, Bulgaria had bad experiences with nationalism in its pre-communist history. This is the main reason that recourse to nationalism has been rather limited during the Bulgarian transition to democracy since 1989. Moreover, the Bulgarians’ belief in the ‘European idea’ and their pursuit of the (supposedly) natural belongingness of Bulgaria to the ‘big European family’ were paralleled by their naïve expectation (actually a hope) that EU membership would have an almost automatic healing effect on the social, political and economic weaknesses and problems of the Bulgarian society. ‘Europe’ as a normative ideology, and the EU as its more concrete institutional manifestation, became part of the core of the Bulgarian political identity during the post-communist transition to democracy. This is especially true for the pre-accession period at the beginning of the twenty-first century. There are also formal and legal reasons for the lack of coherent and clear concept of constitutional identity and for the absence of a developed system of counter-limits in the current Bulgarian constitutional model. The Constitution is a rather rigid one.37 Its fundamental provisions are difficult to amend, requiring a Grand National Assembly. The VII Grand National Assembly, which adopted the Constitution in 1991, did not include any unamendable provisions. Nevertheless, Article 158 of the Constitution, which lists the constitutional issues reserved to the Grand National Assembly, functions as entrenched clause. However, this list is incoherent, as it includes constitutional models of the statehood (the form of governance and the form of territorial distribution of power), human rights, provisions on the normative hierarchy of the

37 See

Evgeni Tanchev, Introduction (fn 10) 185 ff.

50  Martin Belov sources of law, and even requirements concerning the state territory. That is why this entrenched clause can hardly be conceived as an indicator for either the Bulgarian constitutional identity or for any substantive and systematically intertwined counter-limits to EU integration. Indeed, the provisions stipulated in Article 158 of the Constitution are supposed to be hardly amendable – a feature which was deliberately elaborated by the VII Grand National Assembly in order for the entrenched clause to stabilise the Constitution in the context of transition to democracy and to protect it from misuse by the governing elites. This rigidity of Article 158 of the Constitution in conjunction with its supremacy may seem to function as a kind of counter-limit to the influence of supranational law in general and EU law in particular. However, they are not true counter-limits, for two reasons. First, all parts of the Constitution are formally amendable and any limitations to supranational constitutionalism can be removed. Second, the BGCC shifts the boundaries between the competent institutions of the constituent power and imbues the broad concepts of Article 158 of the Constitution with context-dependent meaning. Such practice, despite its problematic nature and the concerns it raises, already exists as it has been demonstrated above with Decision No 3 of 2003, Decision No 3 of 2004 and Decision No 8 of 2005 of the BGCC. Hence, it can be concluded that the Constitution itself may serve the function of substantial limitation to the transfer of sovereignty to the EU and to the influence of EU law on the Bulgarian legal order. Such a thesis is based on the supremacy of the Constitution proclaimed in Article 5, paragraph 1 and on the general inclination of the BGCC to subsume primary EU law under the category of international treaties. Further support for the restraining function of the constitutional supremacy on the influence of EU law can be found in the fact that several pre-accession constitutional amendments were required in order to align the Bulgarian legal order to that of the EU. Without these amendments the Constitution would have prevented the implementation of any EU law provisions conflicting with the Constitution. However, there are also reasons to doubt that the Constitution would have had such an extreme restraining function. First, the absolute supremacy of the national constitution contradicts CJEU case law. While the constitutional identity doctrine suggests that the supremacy of the Member States’ constitutions is a valid constitutional principle recognised by the EU and its institutions, constitutional supremacy is not absolute. It is instead focused predominantly on the provisions that form the constitutional identity of the Member States. Second, the BGCC could have limited that restraining function. In particular, it may have decided to take an explicit stance on the relation between the Constitution and EU law and give limited or unlimited precedence to the latter through judicial dialogue on constitutional identity or an interpretative decision. Third, it is generally hard for constitutional texts to perform genuine restraining function on EU law due to their vagueness and amenability to adjustment and interpretation by parliaments and courts, including constitutional ones. The constitution as a holistic document lacks the sharpness and the concreteness of specific limitations to EU law shaped either as unamendable and entrenched clauses or as elements of constitutional identity. The lack of any clear evidence on the scope, composition, and content of the Bulgarian constitutional identity either in a normative act (the Constitution and the Bulgarian legislation) or in the jurisprudence (the case law of the BGCC, the Supreme Court of Cassation or the Supreme Administrative Court) exposes the Bulgarian legal order to eventually unrestrained axiological, teleological, and institutional influence of the supranational constitutionalism of the EU. The supremacy of the Constitution over EU law, as stipulated by the above-mentioned BGCC decisions, might seem to restrain open statehood in general and the opening function of the EU integration clause in particular. However, the BGCC’s views on the sub-constitutional status of EU law are contained in the Court’s motives to Decision No 3 of 2004 and are partially confirmed by its practice on admissibility of cases which are formally not binding.

Bulgaria  51 Moreover, the EU integration clause is included in the Constitution itself. Hence, one might suppose that the constitutional legislator has deliberately chosen to opt for the openness of the Bulgarian statehood, constitutional system, and legal order, thus allowing for possible derogations of the absolute supremacy of the Constitution in case of conflict with fundamental principles of EU law. In any case, the lack of clarity on the relationship between EU law and the Constitution, the almost inexistent discussion and practice on the Bulgarian constitutional identity, and the potential for conflicts between EU law and treaties to which Bulgaria is party, should they both enjoy the same level in the normative hierarchy, are issues that have to be explicitly answered in the Constitution or at least by the BGCC. To sum up, the core competences and the constitutional identity of Bulgaria are not formally enshrined in the Constitution. They clearly do not coincide with the entrenchment clause – Article 158 of the Constitution. Due to the underdeveloped constitutional identity case law of the BGCC, the Supreme Court of Cassation, and the Supreme Administrative Court, constitutional identity has to be defined by theoretical analysis of the Constitution and the Bulgarian constitutional tradition. The same is true for the limits to the transfer of sovereignty to the EU and to the influence of EU law on the Bulgarian legal order. The procedural limits to transferring sovereignty to the EU are rather weak. This is especially evident where the ratification procedures for treaties reforming EU primary law or treaties on the transfer of constitutional competences are compared with the constitutional amendment procedure. If a reform of EU primary law does not require any changes in the Bulgarian constitutional model and, especially, is not related to transfer of constitutional competences, then the reform treaty should be ratified by an absolute majority of the MPs in two readings. However, where transfer of constitutional competences is also required such a treaty should be ratified by a twothirds qualified majority in two readings. Both procedures are much more simplified than the procedure for amending the Constitution. The Constitution can be amended either through a burdensome procedure involving a Grand National Assembly or by a simplified procedure at the National Assembly. Even in the latter case, the National Assembly has to adopt the act for amendment of the Constitution with a threequarters qualified majority in three readings. In the event that the draft act for amendment of the Constitution receives less than three-quarters but more than two-thirds of the votes of all MPs, it can be reconsidered in a period between two and five months after the initial reading and can be adopted by two-thirds of the votes of all MPs. Consequently, it is harder to adopt amendments of the Constitution, even if they are of secondary importance, than to transfer constitutional, that is to say sovereign, competences to the EU. This awkward situation is a result of several factors. Most important among them are the very pro-European stance of the Bulgarian political elite, the continuous, and maybe even deliberate, misrecognition of the federal features of the EU, the preoccupation of the Bulgarian politicians with domestic constitutional concerns, and the lack of clear understanding of pressing contemporary topics such as flexible sovereignty, constitutional identity, and counter-limits. Finally, the issue of constitutionality review of EU secondary law needs to be addressed. Despite the importance of that issue, there is almost no practice on it in the case law of the BGCC, the Supreme Administrative Court and the Supreme Court of Cassation. The only important case in that regard concerned the implementation of the EU Data Retention Directive38 in Bulgaria through provisions of the Electronic Communications Act.39 38 Directive 2006/24/EC of the European Parliament and of the Council of 15 March 2006 on the retention of data generated or processed in connection with the provision of publicly available electronic communications services or of public communications networks and amending Directive 2002/58/EC, [2006] OJ L 105/54. 39 DV 41/2007 with numerous amendments.

52  Martin Belov These provisions were declared unconstitutional by Decision No 2 of 2015 of the BGCC.40 The BGCC was approached by the ombudsman with the demand that the provisions of the Act infringe the right to private life, the freedom and secrecy of correspondence, and other types of communications which are enshrined in the Constitution. The ombudsman mentioned only indirectly the fact that the EU Data Retention Directive and the national implementing law also infringe the EU Charter of Fundamental Rights (CFR). The reason is purely procedural – the ombudsman has the mandate to seize the BGCC only with cases alleging the infringement of constitutionally protected human rights by the National Assembly. Consequently, the ombudsman had to include arguments on the incompatibility of the Directive with primary EU law as only supporting his case on the unconstitutionality of the national implementing legislation. It is hard to predict whether the entrenched clause of Article 158 of the Constitution will serve as an impediment to Bulgaria’s engagement in future EU Treaties reforms. An explicit proclamation of the primacy of EU law over domestic constitutions or explicit federalisation of the EU may eventually trigger the need for convening a Grand National Assembly in order to amend the Constitution in advance of ratifying such EU Treaty reforms. However, in the context of constitutional pluralism and ‘l’Europe a plusieurs vitesses’, which might seem to be both the more viable solution and the possible choice of the current political elite for the future of the EU (as opposed to the clearly federalist option of Europe as a multilevel constitutional system), recourse to the Grand National Assembly seems unlikely. Such a prognosis seems to hold true for several reasons. The Bulgarian political elite are still rather pro-European and will not risk blocking Treaty reforms and discrediting Bulgaria’s EU membership by convening a Grand National Assembly. In Bulgaria, EU politics hold secondary importance to national politics. Hence, a Grand National Assembly is much more likely to be convened for domestic reasons than for EU Treaty reform. Constitutional pluralism cannot be very easily related to the classical Westphalian concepts such as sovereignty, constitutional supremacy, and form of territorial distribution of power. Hence, a Grand National Assembly will be convened in order to introduce constitutional amendments with preliminary importance for Treaty reforms only if the BGCC or a majority in the National Assembly adopts a position that such a reform touches upon the entrenched clause of Article 158. Unfortunately, it seems that there will be no definite or at least clearer solution regarding the relationship between the Constitution and EU law, the counter-limits provided by the Constitution to EU law, and Bulgarian constitutional identity. These conceptual issues can be clarified either by a Grand National Assembly or by the BGCC. Engagement of the BGCC in a judicial dialogue with the CJEU may serve as a stimulus for and precursor to such important conceptual clarifications.

VI.  Constitutional Rules and Practice on Implementing EU Law A.  The Influence of EU Membership on Bulgarian Parliamentarism Relations between the Council of Ministers and the National Assembly regarding Bulgaria’s participation in the EU decision-making processes are regulated by the Constitution and the



40 DV

23/2015.

Bulgaria  53 Rules of Procedure of the National Assembly.41 Both legal acts generally reproduce the EU legal provisions on that matter. More precisely, they observe the framework set by the Protocol on the Role of National Parliaments in the European Union and the Protocol on the Application of the Principles of Subsidiarity and Proportionality. According to the 2005 amendments of Article 105 of the Constitution, the Council of Ministers has to inform the National Assembly on matters concerning obligations stemming from Bulgaria’s EU membership. Moreover, the Council of Ministers has to inform the National Assembly in advance when it participates in the elaboration and adoption of EU acts. The Council of Ministers must also account to the National Assembly for such acts after their adoption. The Rules of Procedure of the National Assembly empower the Committee on EU matters and Control over EU Funds to impose a parliamentary reservation on EU draft acts that are included in the Annual Working Programme of the National Assembly on EU matters. The right to initiate the imposition of a parliamentary reservation is granted to this committee or to another standing committee of the National Assembly. The parliamentary reservation obliges the Government to refrain from making statements regarding the act concerned at the Council of the EU during a certain term, until the National Assembly pronounces on this issue. These provisions may create the impression that the National Assembly has an enhanced position towards the Council of Ministers compensating it for the structural advantages the Government enjoys in the multilevel EU constitutionalism. However, the National Assembly has never made real use of either the additional rights for anterior or posterior parliamentary control over the Government or the right to impose a parliamentary reservation. In fact the intensity of parliamentary control over the Government has been reduced in the recent years due to internal political factors. This ‘low intensity parliamentarism’ combined with the general tendency of the EU constitutional design to produce ‘executive federalism’ has led to an increasing trend towards marginalisation of the National Assembly in the decision-making process.42 Hence, the National Assembly has neither issued binding resolutions and mandates for the national ministers or for the Prime Minister with regard to their participation in the Council and the European Council, nor availed itself of the opportunities provided in the Protocol on the Role of National Parliaments in the European Union and the Protocol on the Application of the Principles of Subsidiarity and Proportionality.43 It is interesting to note that, especially in the pre-accession period and during the first post-EU accession years, often the implementing Bulgarian legislation strictly followed or even reproduced the text of the directive, occasionally to the extent of its mere translation. Thus, for several years Bulgaria experienced massive formal modernisation through legal transplantation. On the one hand, this process led to the harmonisation of Bulgarian legislation with EU law and in many instances brought innovative solutions to important problems. On the other hand, however, the scope, the formality, and the speed of this massive transplantation produced formal compliance with EU standards that was not necessarily paralleled either by their appropriate adaptation to the Bulgarian legal order or by their integration in the Bulgarian legal and constitutional culture.

41 DV 35/2017. 42 On executive federalism see Andreas Maurer, Parlamentarische Demokratie in der Europäischen Union, (BadenBaden, Nomos, 2002) and, Юрген Хабермас [Jürgen Habermas], За конституцията на Европа [On the Constitution of Europe] (Sofia, Kritika I Humanisam, 2011) 68 ff. 43 See Vatsov, Constitutional Change (fn 12).

54  Martin Belov

B.  The Participation of Bulgarian Courts in Multilevel Judicial Dialogue Bulgarian courts are rather reluctant in participating in judicial dialogue both within and outside the EU framework. The BGCC, the Supreme Administrative Court, and the Supreme Court of Cassation refrain from defining the Bulgarian constitutional identity either as a separate intellectual exercise or within the dialogue between the CJEU and the Constitutional and Supreme Courts of other EU Member States. Moreover, Bulgarian courts have still relatively modest practice in requesting preliminary rulings from the CJEU. Typically, the CJEU is approached by the administrative courts with preliminary rulings having strong emphasis on taxation issues. Equally rare are the cases of horizontal judicial dialogue in the form of migration of jurisprudentially elaborated concepts by the courts of other EU Member States. Bulgarian courts usually do not participate in such migration or mutual exchange of judicial standards. This attitude can be explained by the predominance of legal normativism in the education of Bulgarian lawyers, by the persistence of the understanding that the judge should be ‘the mouth of the law’ as prescribed by Montesquieu centuries ago, by the lack of training to make use of foreign judicial practice, and by the belief that such use may be even illegal. An example of a missed opportunity for engagement in multilevel judicial dialogue is the EU Data Retention Directive. The BGCC was approached by the ombudsman to consider the unconstitutionality of the implementing legislation and, thus, only indirectly its incompatibility with EU law, whereas the Supreme Court of Cassation and the Supreme Administrative Court could have asked the BGCC to declare the Bulgarian legislation not only unconstitutional but also contrary to the CFR. Bulgarian courts are also not very keen on applying the case law of supranational courts. According to a 2012 representative survey with judges and state prosecutors of the Justice Development Foundation, the proportion of magistrates (judges and state prosecutors) who never apply directly the case law of the CJEU and the European Court of Human Rights (ECtHR) was 17.1 per cent and 12.5 per cent respectively. Moreover, only 54.1 per cent of the magistrates declared that they regularly accord primacy to international treaties, 43.2 per cent the case law of the CJEU, 52.6 per cent the ECtHR case law. And this is so only if there is an established practice.44 The most popular jurisprudential source of law with supranational origin is the ECtHR case law. It is used by courts of different instances. It is a rather surprising fact that, in contrast to this general restrictive attitude of the Bulgarian courts towards engagement in multilevel and supranational judicial dialogue, they are eager to use the CFR as a source of legal argumentation. This finding is more interesting with regard to the fact that many ordinary courts of first and second instance quote the Charter and use it as indirect source of argumentation even in cases in which the Charter cannot be applied as valid source of law due to the fact that it is neither violated by an EU institution or a national institution implementing EU law (Article 51 CFR), nor is there a sufficient ‘degree of connection between the measure of EU law and the national measure at issue’ as required by the CJEU.45

44 For further analysis and data see Мартин Белов [Martin Belov], ‘Източниците на правото в действие – правосъдието между позитивизма и реализма’ [‘The Sources of Law in Action: the Judiciary between Legal Positivism and Legal Realism’] (2012) Съвременно право [Modern Law] No 4, 29. 45 CJEU 10.07.2014 C-198/13 (Julian Hernández and Others) ECLI:EU:C:2014:2055. See also Martin Belov and Maria Fartunova, ‘Bulgaria’ in L Burgorgue-Larsen (ed), The Charter of Fundamental Rights as Apprehended by the Judges in Europe (Paris, Pedone, 2017) 177.

Bulgaria  55

VII.  Resulting Relationship between EU Law and National Law During the almost 30 years of post-communist transition to democracy, and especially during the pre-accession and the EU membership periods, the Bulgarian legal order has been deeply influenced by supranational and comparative legal standards and has, thus, changed profoundly. The quality and the quantity of institutional and conceptual novelties which were introduced in the Bulgarian constitutional and legal system over recent decades amount to a legal revolution. The Bulgarian constitutional design was greatly restructured in the years immediately following the fall of the authoritarian regime. During the first decade of democratic transition a new Constitution was adopted and the legal framework was established for brand-new institutions such as the Constitutional Court, the President, the Supreme Judicial Council and the ombudsman, as well as for the reorganisation of pre-existing institutions with their specific institutional memory. In comparison, the novelties that were formally introduced in the Bulgarian constitutional system and especially in the Constitution with view of the EU accession might not seem great in terms of quality and quantity. However, a closer look shows that the evolving federalist features of the EU, the development of the constitutional identity debate, the primacy of EU law, the new roles performed by classical actors in multilevel games,46 the emergence of executive federalism, the crisis of parliamentarism and representative democracy, and the institutionalisation of a set of human rights in quasi-federal jurisdiction in the form of the CFR inevitably induce structural and functional changes, shifts, and fractions with the fundamental character of the Bulgarian constitutional order. The debate on adjusting the Bulgarian constitutional order to the EU has been focused on institutional issues such as the constitutional amendments required by the EU integration and the reform of the judiciary. However, the need to adapt the Bulgarian constitution to the European ‘ideal constitution’ and to enhance the socio-legal prerequisites for the emergence and development of genuine civic society, political pluralism, democracy and rule of law have been underestimated. The challenges to the EU membership of Bulgaria are produced by factors concerning both the Bulgarian state and society and the EU. The national issues that have to be resolved concern the transition from formal compliance of the Bulgarian legislation with EU standards to the real implementation of EU law requirements. Other aspects of potential reform deserving attention are the strengthening of the civil society and of the competence and legal consciousness of the different legal professions as well as the establishment of independent media and idea-based, and not just leadership-based, political parties. The fight against corruption, nepotism, misuse of state institutions for party political and oligarchic interests and the prevention and deconstruction of authoritarian trends are also key issues of immediate importance for reforming the Bulgarian political system. The challenges on the EU level concern the taking of a strategic decision on the further development of the EU integration, the overcoming of the democratic deficit,47 and the establishment of a real and functioning supranational democracy. There are also problems that have to be addressed at both national and EU levels. These are the crisis of representation, parliamentarism and democracy in the EU and its Member States, and the increased role of new factors of global governance that cannot be framed in the classical terms of modern representative democracy. 46 See Belov, ‘Multilevel Constituent Power Games’ (fn 30). 47 See Livia Door, The Democratic Deficit Debate in the European Union: Is There a Democratic Deficit in the European Union? (Saarbrücken, VDM Verlag, 2013).

56  Martin Belov From a conceptual point of view, the most challenging issue consists in the structural changes in the constitutional geometry produced by globalisation and supranationalisation.48 Constitutional geometry is the conceptual explanatory scheme for the adjustment and relationship between different legal orders. The demise of ‘closed statehood’ based on hierarchy and strict territoriality, and the emergence of forms of supranational constitutionalism, pose an important question – whether hierarchy is still the semi-universal explanatory paradigm for structuring the legal order and the public power or whether it has to be replaced or supplemented by the network. In fact, the debate which is the better explanatory paradigm for the EU supranational constitutionalism – the hierarchybased multilevel constitutionalism or the network-based constitutional pluralism – is a result of the conceptual issue of the post-modern institutional design in the age of global governance. These problems have to be addressed by both the EU constitutionalism and the Member States’ constitutions. The multilevel constitutional paradigm cannot be easily reconciled with the requirements of the entrenched clause of Article 158 of the Constitution. Further EU federalisation seems to require the sanction of a Grand National Assembly. Such sanction should introduce some explicit reflections of the changed EU constitutional infrastructure in the Constitution. This can be done either through a single Europe provision (following the example of the German Basic Law) or in the form of constitutional provisions spread throughout the relevant parts of the Constitution in accordance with the sedes materiae concept. An alternative option for the adjustment of the Constitution to future EU constitutional developments is the case law of the BGCC. It can provide flexible solutions in case of multilevel constitutionalism or constitutional pluralism trends in future EU Treaties reforms. The BGCC may adjust and adapt the Bulgarian legal order to the EU one by either engaging in judicial dialogue with the CJEU or by using its domestic competences such as abstract interpretation of the Constitution and abstract or concrete norm control. In the Bulgarian case, the permissive consensus for EU integration, the lack of extensive scientific and political debate on the issues posed by the EU supranational constitutionalism to the Bulgarian constitutional order, constitutional culture, and institutional design and the judicial self-restraint of the BGCC to enter into a judicial dialogue with the CJEU may have ambiguous and to some extent unpredictable effects. Indeed, they allow for the smooth and swift EU integration of Bulgaria and for the effective, albeit somewhat formal, adjustment of the Bulgarian legal order to the EU one. However, they do not foster the exposure of national constitutional identity in sources of valid law. Moreover, they prevent explicit counter-limits provisions with respect to EU law and mitigate the relationship between the Constitution and EU law. What is especially problematic, however, is the fact that they impede the genuine and profound inclusion of the EU constitutional axiology in the Bulgarian constitutional culture. No substantive and durable internalisation of European values and constitutional foundations in the Bulgarian constitutional culture would be possible without a profound and extensive critical debate, both informal and institutionalised, on the issues that EU integration is currently posing as part of the general trend towards emergence of supranational and even global constitutionalism. In light of current developments, it seems that EU leaders, as well as the leaders of the most influential EU Member States, are opting for a multispeed Europe. This means that attempting to establish a multilevel constitutional system following more traditional federalist models will eventually be replaced by an asymmetrical vision of the EU or even of Europe in general that is closer to constitutional pluralism. Although constitutional pluralism is yet to be clearly defined 48 On constitutional geometry see Martin Belov, ‘The Challenges to Westphalian Constitutional Geometry in the Age of Supranational Constitutionalism, Global Governance and Information Revolution’ in M Belov (ed), Global Constitutionalism and Its Challenges to Westphalian Constitutional Law (Oxford, Hart, 2018) 54; Martin Belov (ed), Global Governance and Its Effects on State and Law (Frankfurt a M, Peter Lang, 2016) 7 ff.

Bulgaria  57 as a concept, it seems to more properly reflect the diversity of constitutional and legal orders and of constitutional cultures in Europe in general and in the EU in particular. Seen from a Bulgarian perspective, this multipolar and multifaceted ‘Europe of different speeds, circles, networks and levels’ produces reasonable concerns. The most important concern among them, from a strategic point of view, is that Europe as a normative ideal has served formative function in the course of the Bulgarian modernisation since the mid-nineteenth century. Moreover, the EU as a specific institutionalised version of Europe has functioned as the most important stimulus and trigger for (sometimes uneasy) reforms in the Bulgarian legal order and society. Hence, a situation in which Bulgaria is placed in the EU’s periphery, that is to say outside of the ‘inner circles’ of EU integration, comprising for example the eurozone or a common European army, may definitely and durably discourage the impetus for further modernisation via Europeanisation. In the worst case, it may partially dismantle the formative influence of the European ideal on the Bulgarian ideal, written and factual constitution and produce a search for alternative sources of modernisation.

References М Белов, [M Belov] Пряка демокрация – национални и сравнителноправни аспектил [Direct Democracy – National and Comparative Aspects] (Sofia, Sibi, 2009). М Белов [M Belov], ‘Източниците на правото в действие – правосъдието между позитивизма и реализма’ [‘The Sources of Law in Action: the Judiciary between Legal Positivism and Legal Realism’] (2012) Съвременно право [Modern Law] No 4, 29. М Белов [M Belov], ‘Функции на конституционния преамбюл’ [‘Functions of the Constitutional Preamble’] in Право, управление и медии през XXI век [Law, Governance and Media in XXI Century] (Blagoevgrad, South-Western University Press, 2012) 207. M Belov, ‘The Preamble of the Constitution in the European Constitutionalism’ (2013) No 20 Revista General de Derecho Romano. М Белов [M Belov], ‘Конституционният статус на монарха в балканския конституционализъм’ [‘The Constitutional Status of the Monarch in the Balkan Constitutionalism’], in В Мръчков [V Mrachkov] (ed), 135 от приемането на Търновската конституция [135 Years of the Adoption of the Tarnovo Constitution] (Sofia, Sibi, 2014) 262. M Belov, ‘Separation of Powers reconsidered: a Proposal for a New Theoretical Model at the Beginning of the 21st Century’ in A Geisler, M Hein and S Hummel (eds), Law, Politics, and the Constitution. New Perspectives from Legal and Political Theory (Frankfurt a M, Peter Lang, 2014) 47. M Belov, ‘Executive Power Metamorphoses in the History of the Balkan Constitutionalism: Between the Imperatives of the Constitutional Teleology and the Potentialities of the Political Reality’, Romanian Journal of Comparative Law No 5 (Bucharest, Universul Juridic, 2014) 299. M Belov, ‘Establishment of the Modern Constitutional Principles in Bulgaria. The Tarnovo Constitution of 1879’ in M Stolleis (ed), Konflikt und Koexistenz. Die Rechtsordnungen Südosteuropas im 19. und 20. Jahrhundert, Vol 1: Rumänien, Bulgarien, Griechenland (Frankfurt a M, Vittorio Klostermann Verlag, 2015) 859. M Belov (ed), Global Governance and Its Effects on State and Law (Frankfurt a M, Peter Lang, 2016). M Belov, ‘The Challenges to Westphalian Constitutional Geometry in the Age of Supranational Constitutionalism, Global Governance and Information Revolution’ in M Belov (ed), Global Constitutionalism and Its Challenges to Westphalian Constitutional Law (Oxford, Hart, 2018) 54. M Belov, ‘Constitutional Courts as Ultimate Players in Multilevel Constituent Power Games: the Bulgarian Case’, in M Belov (ed) Courts, Politics and Constitutional Law. Judicialization of Politics and Politicization of the Judiciary (Abingdon, Routledge, 2019) 152. M Belov and M Fartunova, ‘Bulgaria’ in L. Burgorgue-Larsen (ed), The Charter of Fundamental Rights as Apprehended by the Judges in Europe (Paris, Pedone, 2017) 177. M Belov and A Tsekov, ‘The Constitutional Court of the Republic of Bulgaria as Law Maker’, in M FlorczakWator (ed), Judicial Law-Making in European Constitutional Courts (Oxford, Routledge, 2020) 91. N Berlatsky, Judicial Activism (Opposing Viewpoints) (New York NY, Greenhaven Press, 2011).

58  Martin Belov Г Близнашки [G Bliznashki], Парламентарното управление в България [The Parliamentary Government in Bulgaria] (Sofia, University of Sofia St Kliment Ohridski Press, 2009). N Brenner, ‘Beyond state-centrism? Space, Territoriality and Geographical Scale in Globalization Studies’ (1999) 28 Theory and Society 39. P Craig, G de Búrca, EU Law: Text, Cases and Materials, 4th edn (Oxford, OUP, 2008). L Door, The Democratic Deficit Debate in the European Union: Is There a Democratic Deficit in the European Union? (Saarbrücken, VDM Verlag, 2013). P Faraguna, A Living Constitutional Identity: The Contribution of Non-Judicial Actors, Jean Monnet Working Paper 10/15 (New York, NYU School of Law, 2015) 27. HP Glenn, Legal Traditions of the World. Sustainable Diversity in Law (Oxford, OUP, 2000). Ю Хабермас [J Habermas], За конституцията на Европа [On the Constitution of Europe] (Sofia, Kritika I Humanisam, 2011). S Hobe, ‘Statehood at the End of the 20th Century – The Model of the “Open State”: A German Perspective’ Austrian Review of International and European Law No 2 (Leiden, Brill, 1997) 127. G Jellinek, Allgemeine Staatslehre (Charleston NC, Nabu Press, 2011). P Kiiver, ‘European scrutiny in national parliaments. Individual efforts in the collective interest?’, in J O’Brennan and T Raunio (eds), National Parliaments within the Enlarged European Union. From ‘victims’ of integration to competitive actors? (Abingdon, Routledge, 2007) 95. Пл Киров [Pl Kirov], Президентът в българския конституционен модел [The President in the Bulgarian Constitutional Model] (Sofia, Sibi, 2004). D Law, W C Chang, ‘The Limits of Global Judicial Dialogue’ (2011) 86 Washington Law Review 523. G Martinico, ‘Complexity and Cultural Sources of Law in the EU Context: From the Multilevel Constitutionalism to the Constitutional Synallagma’ (2007) 8 German Law Journal 205. A Maurer, Parlamentarische Demokratie in der Europäischen Union (Baden-Baden, Nomos, 2002). D Mishkova, ‘Domesticating Modernity: Transfer of Ideologies and Institutions in Southeastern Europe’ in M Stolleis (ed), Konflikt und Koexistenz. Die Rechtsordnungen Südosteuropas im 19. und 20. Jahrhundert vol 1: Rumänien, Bulgarien, Griechenland (Frankfurt a M, Vittorio Klostermann Verlag, 2015) 723. O Pollicino, ‘European Judicial Dialogue and the Protection of Fundamental Rights in the New Digital Environment: An Attempt at Emancipation and Reconciliation. The Case of Freedom of Speech’ in S MoranoFoadi and L Vickers (eds), Fundamental Rights in the EU. A Matter for Two Courts (Oxford, Hart, 2015) 93. A Saiz Arnaiz and C Alcoberro Llivina (eds), National Constitutional Identity and European Integration (Mortsel, Intersentia, 2013). Е Танчев [E Tanchev], Въведение в конституционното право [Introduction to Constitutional Law] (Sofia, Sibi, 2003). E Tanchev, ‘The Treaty on the Constitution for Europe and National Constitutions in the Age of Constitutional Pluralism’ in L Breitner-Czuma (ed), Das alte Europa in neuer Verfassung? (Baden-Baden, Nomos, 2007) 83. Е Танчев [E Tanchev], ‘Съвременният наднационален конституционен плурализъм или подреждането на конституционните правопорядъци’ [‘The Contemporary Supranational Constitutional Pluralism or the Ordering of Legal Orders’], in Пл Киров (ed), Конституционнноправни изследвания. 2012–2013 [Constitutional Studies 2012–2013] (Sofia, University of Sofia St Kliment Ohridski Press, 2013) 171. E Tanchev and M Belov, ‘Constitutional Gradualism: Adapting to EU Membership and Improving the Judiciary in the Bulgarian Constitution’ (2008) 14 European Public Law 3. G Tsebelis, Nested Games. Rational Choice in Comparative Politics (Berkley CA, University of California Press, 1990). M Vatsov, ‘European Integration Through Preliminary Rulings? The Case of the Bulgarian Constitutional Court’ (2015) 16 German Law Journal 1591. M Vatsov, Constitutional Change Through Euro Crisis Law: Bulgaria (2015), www.academia.edu/11550630/ Constitutional_Change_Through_Euro_Crisis_Law_Bulgaria. M Waters, ‘The Future of Transnational Judicial Dialogue’ (2010) No 104 American Society of International Law, Proceedings of the Annual Meeting 465. A Watson, Legal Transplants: An Approach to Comparative Law 2nd edn (Athens GA, University of Georgia Press, 1993).

3 Czech Republic Assertive Judiciary and Baffled Administration LUBOŠ TICHÝ AND TOMÁŠ DUMBROVSKÝ

I.  Main Characteristics of the Constitutional System A.  Main Features of the Constitutional System of the Czech Republic The Czech Constitution (ÚČr) was adopted in 1993 following the dissolution of Czechoslovakia. Constitutional laws are organised into a constitutional order, which consists of several constitutional acts: the Constitution of the Czech Republic in the strict sense,1 the Charter of Fundamental Rights and Freedoms,2 and a number of other constitutional acts and their amendments.3 This constitutional order forms a referential framework for the Constitutional Court for reviewing the constitutionality of legislative and administrative acts. All constitutional provisions, disregarding in which constitutional act they are to be found, have the same legal status with one exception: the eternity clause in Article 9(2) of the Constitution grants ‘higher status’ to ‘the essential requirements for a democratic state governed by the rule of law’, changes of which are impermissible. Treaties ratified by the Parliament sit between constitutional law and ordinary law in the hierarchy of norms and enjoy application priority over statutory law in cases of conflict. Constitutional amendments can take the form of a separate constitutional act or a change to an existing constitutional act. There is no legal difference in these forms. Adopting a constitutional amendment requires qualified majorities in both chambers of Parliament.4 The Constitution must always be seen in the context of the constitutional order as a whole and understood as a value-based document. These values must be upheld by all public authorities 1 Constitutional Act No 1/1993 Sb (Sbírka zákonů, Collection of Laws). 2 Constitutional Act No 2/1993 Sb. 3 The concept of constitutional order is defined in Art 112 (1) ÚČr: ‘The constitutional order of the Czech Republic is made up of this Constitution, the Charter of Fundamental Rights and Basic Freedoms, constitutional acts adopted pursuant to this Constitution [eg Constitutional Act No 347/1997 Sb, on the Creation of Higher Territorial Self-Governing Units; Constitutional Act No 110/1998 Sb, on the Security of the Czech Republic; Constitutional Act No 515/2002 Sb, on the Referendum on the Czech Republic’s Accession to the European Union] and those constitutional acts of the National Assembly of the Czechoslovak Republic, the Federal Assembly of the Czechoslovak Socialist Republic, and the Czech National Council defining the state borders of the Czech Republic, as well as constitutional acts of the Czech National Council adopted after the sixth of June 1992’. The main reason for constructing the concept of constitutional order was to ensure that the Charter of Fundamental Rights and Basic Freedoms would have equal status to the Constitution. 4 Art 39 (4) ÚČr. Three-fifths majorities of all Deputies and of Senators present are required for the adoption of a constitutional act.

60  Luboš Tichý and Tomáš Dumbrovský whenever they are performing constitutional functions such as adopting laws, issuing administrative acts or adjudicating legal conflicts. The Constitution follows a specific objective and, together with the Charter of Fundamental Rights and Basic Freedoms, is designed to prevent the rise of a totalitarian regime. It features division of powers and a system of checks and balances between the respective branches. The system of public power of the Czech Republic is based on parliamentary and representative democracy with a prevailing proportional electoral system and some features of direct democracy. There are two chambers of parliament, the Government executes part of its powers in conjunction with the President of the Republic, and the Constitutional Court, inter alia, reviews the exercise of public power and adjudication of general courts in light of its constitutionality. Also, the Czech National Bank, the Supreme Audit Office and self-governing units, alongside the principle of selfadministration, play an important role in the constitutional system.

B.  The Relationship between the Constitutional Court and Ordinary Courts The Czech judiciary is composed of ordinary courts and the Constitutional Court. The system of ordinary courts is headed by two supreme courts – the general Supreme Court and the Administrative Supreme Court. The relationship between the ordinary courts and the Constitutional Court originates in Hans Kelsen’s concept of constitutional judiciary as a specialised institution with a unique task of supervising constitutionality of the legal system through abstract review of legislation. The first Constitutional Court in Czechoslovakia was established in 1921. In line with Kelsen’s idea, the Court exercised abstract review only. It decided whether a particular law is in conflict with the Constitution exclusively upon a motion of the Supreme Court of Justice, Supreme Administrative Court, the Electoral Court, the Chamber of Deputies, and the Senate or the Diet of Russinia5 within three years of promulgation. Due to the control of its case load by the legislators and supreme courts, the first Constitutional Court had little impact on the constitutional development and effectively ceased to function after its first 10-year term expired. The doctrinal leadership in the interwar period was assumed by the Supreme Administrative Court instead. Full restoration of the constitutional judiciary did not occur until the democratic transition following the 1989 revolution. Since the Supreme Administrative Court was effectively abolished in 1948, it was the general Supreme Court that was at the apex of the judiciary until 1989. The Supreme Administrative Court was re-established in 2003 and has been committed to restore its interwar reputation. It took a considerable effort, lasting two decades, before the Constitutional Court established its authority vis-à-vis ordinary courts, especially the two supreme courts. In the effort to place the Constitution in the centre of the legal order and enforce the rationale that the Constitution creates a specific duty on the general courts to provide full and effective protection of constitutional rights, the Constitutional Court clarified two interrelated issues: the extent to which its decisions bind the general courts; and the extent to which the general courts are obliged to interpret provisions of general law6 in the light of provisions of constitutional law. Establishing the authority of the Constitutional Court vis-à-vis ordinary courts represented an important pretext

5 The Diet of Russinia was the legislative body of an autonomous region situated in Sub-Carpathian Russia, which formed part of Czechoslovakia from 1918 to 1945. 6 A category of laws having lower than constitutional status within the Czech legal system.

Czech Republic  61 for a similar struggle, in which the authority of European Union (EU) law and its interpreter, the Court of Justice of the EU (CJEU), would be established. A struggle between the Constitutional Court and the Supreme Court on the question of which is the final arbiter within the Czech legal order occupied the highest Czech judicial institutions for most of the 1990s. At the heart of this struggle was the question whether the decisions of the Constitutional Court carry precedential force. In Article 89(2), the Constitution only stipulates that ‘[e]nforceable decisions of the Constitutional Court are binding on all authorities and persons’. While the Constitutional Court repeatedly declared an obligation of the Supreme Court to respect its case law, the Supreme Court opposed such an ‘absolute’ binding force of the Constitutional Court’s decisions. The Court acknowledged the problem in a decision of 1998: ‘[T]he binding force of Constitutional Court judgments which, in the present state of the law and in spite of the fact that it represents the sine qua non condition of the constitutional judiciary, brings no small amount of difficulties in its wake’.7 While the Court asserted ‘cassation’ binding force, it left the question of a ‘precedential’ binding force of its decisions open for further discussion. The general courts eventually accepted the unconditional ‘cassation’ binding force.8 In 2011, the Constitutional Court reviewed the discussion. It summed up that both supreme courts (the Supreme Court and the Supreme Administrative Court) had accepted the unconditional ‘cassation’ binding force and that a consensus had been reached on the functioning of the ‘precedential’ binding force: while the ‘cassation’ binding force is unconditional, within the tenets of the ‘precedential’ binding force, ‘there is a possibility for a general court to (not) reflect the legal opinions of the Constitutional Court when the general court in good faith outlines a dissenting reasoning and hence starts a constitutional dialogue with the Constitutional Court’.9 The Czech Constitutional Court’s critique of an ‘exalted formalism’ practised by the general courts represents the second area of the Constitutional Court’s effort to assert the effect of fundamental rights within the area of general law. This ‘exalted formalism’, according to the Constitutional Court, rested in a strict adherence by the general courts to textual interpretation of the provisions of general law. Various constitutional provisions on fundamental rights, the Constitutional Court repeatedly held, nonetheless obliged the general courts to search for the purpose and sense of the provisions of general law.10 In its view, the general courts ought to interpret provisions of general law in the light of constitutional provisions, especially those affording rights to individuals, which ‘radiate’ through the entire legal order (radiation doctrine and principle of constitutionally-conforming interpretation).11 7 Czech Constitutional Court (hereinafter ÚsČr) 02.12.1997, II.ÚS 425/97 ECLI:CZ:US:1997:2.US.425.97, N 42/10 SbNU (Sbírka nálezů a usnesení Ústavního soudu, Collection of Decisions and Resolutions of the Constitutional Court) 285, paras 6–8. 8 Cf inter alia decisions of the Czech Supreme Administrative Court (hereinafter NssČr), 6 Ads 62/2003 ECLI:CZ:NSS:2008:6.ADS.62.2003 and the Czech Supreme Court, 30 Cdo 610/2012 ECLI:CZ:NS:2013:30. 9 ÚsČr 08.11.2011, IV.ÚS 1642/11 ECLI:CZ:US:2011:4.US.1642.11.1, N 191/63 SbNU 219, para 21. 10 See inter alia ÚsČr 04.02.1997, Pl. ÚS 21/96 ECLI:CZ:US:1997:Pl.US.21.96, N 13/7 SbNU 87; ÚsČr 03.02.1999, Pl. ÚS 19/98 ECLI:CZ:US:1999:Pl.US.19.98.1, N 19/13 SbNU 131; ÚsČr 02.06.2010, I. ÚS 2232/07 ECLI:CZ:US:2010:1. US.2232.07.1, N 119/57 SbNU 467 (where the Court held that ‘a general court is not bound by the textual meaning of a statute absolutely; it may and shall diverge from [the textual meaning of the provision] if the purpose of the statute, history of its origin, systemic coherence, or a principle rooted in the constitutionally-conforming legal order understood as a value system require so … Fundamental rights and freedoms function within the area of general law as regulative ideas’. Ibid, part IV, paras 1–2). 11 In 2006 the Constitutional Court held that it ‘has repeatedly emphasised the duty of general courts to interpret individual provisions of procedural law from the point of view of purpose and sense of the constitutionally guaranteed fundamental rights and freedoms, which “radiate” through the entire legal order’. ÚsČr 13.06.2006, I. ÚS 287/05 ECLI:CZ:US:2006:1.US.287.05, N 121/41 SbNU 515, part III, para 4. For the genesis of the concept of radiation adapted from the German doctrine of Ausstrahlung (more or less correctly captured in decision I. ÚS 2232/07 [n 10]) and its misinterpretation by the ÚsČr (as well as the CJEU) in the area of EU law, see Luboš Tichý and Tomaš Dumbrovský,

62  Luboš Tichý and Tomáš Dumbrovský

II.  Constitutional Culture A.  The Importance of the Constitution The Constitution’s most important feature is its omnipresence with respect to the legal order. This is especially important within the protection of fundamental rights and freedoms and their regulation. The Constitution is not only granting and protecting rights and freedoms, designing institutional framework and governing the exercise of power by state authorities, it also vests state authorities with interpretation and application of law in a manner compliant with the protection of fundamental rights and freedoms.12 Similar to other democratic constitutions, the Czech Constitution is directly applicable. The doctrine of direct applicability of the Constitution is ensured by the constitutional judiciary. The Constitutional Court is ‘responsible for the protection of constitutionality’.13 It should be borne in mind that the Constitution is a specific legislative instrument. Many of its provisions are, in essence, legal principles and not legal norms in the strict sense, and thus may not be applied by analogy as is the case of ‘ordinary’ law. Principles are enforced through applying the law. They require implementation, that is, specification in order to become enforceable in the sense of applying the law. Also, the Constitution sometimes contains only torsos of legal norms, especially when regulating institutional framework and procedure, but does not clearly delimit subjects, parts of material elements, etc. Nonetheless, the interconnection between principles’ purpose and values is decisive. Therefore, a transparent and foreseeable interpretation of the Constitution is crucial. Rulings of the Constitutional Court are, in general, better received when Constitutional Court’s interference with the contested act or decision is only minimal and thus avoids judicial activism. This moderate approach is typical for the Constitutional Court’s adjudication on constitutional complaints in concrete review. Ruling on constitutionality of laws in abstract review is typically subject to sharp criticism. This is because, among others, the Constitutional Court deems not only the decision, but also its reasoning, to be generally binding, that is, a source of law. This approach of the Constitutional Court is warranted from the perspective of the protective function of the Constitution. The Constitution has asserted the central place in the functioning of public power in the Czech Republic through a number of constitutional claims and counterclaims advanced by the key constitutional actors. The most instrumental was the struggle between the Constitutional Court and the ordinary judiciary. As mentioned above, the Constitutional Court, employing the ‘radiation’ rationale, compelled the general courts to accept the supremacy of the Constitution and its specific methods of interpreting the Constitution (mostly teleological), which are aimed at advancing constitutional values. The relationship between the Constitutional Court and the supreme courts was settled in a way that aimed to foster a dialogue on new and contentious constitutional issues. This settlement was tested in the Slovak Pensions dispute, in which the Supreme Administrative Court involved the CJEU to counter the Constitutional Court’s case law on the matter (see more below).

‘Between Two Legal Orders: A Relativist Doctrine for a Member State Constitutional Court?’, in A Koch et al (eds), Europe. The New Legal Realism. Essays in Honour of Hjalte Rasmussen (Copenhagen, Djøf Forlag, 2010) 757. 12 ÚsČr 21.07.1998, III.ÚS 129/98 ECLI:CZ:US:1998:3.US.129.98; and ÚsČr 21.01.1999, III. ÚS 257/98 ECLI:CZ:US:1999:3.US.257.98. 13 Art 83 ÚČr.

Czech Republic  63 Vis-à-vis the Parliament, the Constitutional Court asserted the power to review the constitutionality of constitutional amendments from the standpoint of the material core of the Constitution. The 1993 Constitution favoured stability of the Parliament. While the Senate is indissoluble, the Chamber of Deputies could be dissolved only after failing three times to vote confidence in a new Government. The procedure was lengthy, and having failed to reach a consensus on a new Government, Parliament opted in 1998 to dissolve the Chamber of Deputies by an ad-hoc constitutional act shortening the parliamentary term. However, when this was repeated in 2009, the, Constitutional Court quashed the constitutional act. The Court held that the act violated the material core of the Constitution for two reasons: it lacked generality as one of the attributes of law, and by shortening the term of the Chamber of deputies retroactively, it violated the right to vote.14 The Parliament, in response, passed a constitutional amendment that empowered the President of the Republic to dissolve the Chamber of Deputies upon its request in the form of a resolution adopted by a three-fifths majority of all deputies.15 Despite the Czech Republic being a parliamentary system, the powers of the President of the Republic may become extensive especially during times of political transition and crisis. After the transformative presidency of Václav Havel, his successors increasingly tested the limits of presidential powers vis-à-vis both the Parliament and the Government. During the ratification of the Lisbon Treaty and the Article 136 TEU Amendment, President Klaus claimed a veto power over ratification of treaties. The Constitution states that the President ratifies treaties, while the systemic view indicates centrality of Parliament in the constitutional system. The Parliament gave its assent to both treaties and issued a resolution demanding the President’s ratification in both cases. The President eventually signed the Lisbon Treaty under the condition that the Czech Republic would be exempted from the Charter of Fundamental Rights of the EU (see below). However, he refused to sign the Article 136 TEU Amendment. The issue was resolved only with the expiration of the President’s term and ratification by his successor. The Parliament went as far as to charge the President with high treason, which the Constitutional Court dismissed for procedural reasons, since the President’s term expired in the meanwhile.16 The 2012 Amendment to the Constitution introduced direct election of the President.17 Despite presidential powers being slightly reduced as part of the reform, a newly acquired independence from the Parliament and boosted democratic legitimacy encouraged President Zeman to further test the limits of his constitutional powers. The President repeatedly asserted broader powers over the composition of the Government. In 2013, upon the demise of the Government, he unilaterally appointed a new Government that had no support in the Parliament. Despite a no-confidence vote in the Parliament, the extra-parliamentarian Government remained in power until the next general elections. Several other instances of highly contested constitutional interpretation advanced by the President have threatened to undermine some of the core constitutional values, such as independence of courts and universities on the executive power. This includes selective refusal to appoint university professors18 and lower court judges,19 a struggle 14 ÚsČr 10.09.2009, Pl. ÚS 27/09 (Melčák) ECLI:CZ:US:2009:Pl.US.27.09.1, N 199/54 SbNU 445. In response, the Parliament adopted an amendment to the Constitution that allows for the dissolution of the Chamber of Deputies by the President upon a request by a three-fifths majority of the deputies. Constitutional Act 319/2009 Sb. 15 Constitutional Act 319/2009 Sb. 16 ÚsČr 27.03.2013, Pl. ÚS 17/13 ECLI:CZ:US:2013:Pl.US.17.13.1, U 4/68 SbNU 521. The ratification issue was only one of the charges. The other charges included the failure to appoint judges to the vacant positions at the Constitutional Court, failure to appoint the judge to ordinary court, the amnesty decision and delaying a ratification of the Additional protocol to the European Social Charter by eight years. 17 Constitutional Act 71/2012 Sb. 18 See Czech Supreme Administrative Court, NSS 27.04.2018, sp. Yn. 10A 174/2016-143. 19 See Czech Supreme Administrative Court, NSS 21.05.2008, sp. zn. 4 Ans 9/2007-197.

64  Luboš Tichý and Tomáš Dumbrovský over appointments and recall of the Supreme Court President,20 a failure to nominate judges to the Constitutional Court to fill the vacancies, leaving the Court for a considerable time without the necessary number of judges to exercise its judicial review power, and undermining the credibility of intelligence services for political reasons. The constitutional culture has further deteriorated since the 2017 general elections, which concluded the transformation of the party system that had started following the financial crisis in 2009. The minority Government composed of populists and Social Democrats and supported by the Communists has taken a nationalistic, less liberal, and consequently increasingly Eurosceptic course, supported by the President. The Prime Minister, Andrej Babiš, has been investigated for a conflict of interest, involving misuse of European funds in his private capacity as a businessman, by both Czech and EU authorities.21 The process of redefining constitutional values of the 1989 revolution was further catalysed by the refocusing of Czech foreign policy, under pressure from President Zeman, regarding Russia, China and Visegrad cooperation, that includes the illiberal regimes of Poland and Hungary.22

B.  The Constitution and Its Interpretation The interpretation of the Constitution is characterised by the emphasis on the unity of the constitutional order. Provisions of the Constitution must be interpreted in their unity and in their context as they articulate a unified and consensual concept of the state. A significant contribution of the Constitutional Court to the development of the legal culture in the Czech Republic is its emphasis on teleological interpretation. The objective of interpretation of constitutional law is different from the interpretation of ordinary law. In contrast with the interpretation of ordinary law, the interpretation of constitutional law focuses more on the normative objective than on a textual interpretation of stipulated rights and duties. The proportionality, object of interpretation, and value the interpreter accords to rights must be the starting point. Interpretation of constitutional law depends on the following factors: a) consensus on the content of constitutional institutes, b) homogeneity of generally accepted constitutional values in a democratic society, c) professional and ethical criteria by which a democratic society chooses the people occupying the positions of authoritative interpreters. According to the Constitutional Court, law’s binding force does not imply a duty to apply the law literally or without critical analysis; the courts are bound by the objective and meaning of the law. When the wording of a law conflicts with its meaning and objective, the result of a teleological interpretation must prevail because it limits arbitrariness in the application of the law.23 Hence, the tension between textual and teleological interpretation is resolved in favour of the latter.24 20 See ÚsČr 12.09.2006, II. ÚS 53/06, ECLI:CZ:US:2006:2.US.53.06 and ÚsČr 11.7.2006, Pl.ÚS 18/06 (Removal of Chief Judges), ECLI:CZ:US:2006:Pl.US.18.06. 21 See the European Commission, Final Audit Report, Audit No REGC414CZ0133 of 23.11.2019; ÚsČr 18.02.2020, Pl. ÚS 4/17 (lex Babiš) ECLI:CZ:US:2020:Pl.US.4.17.2 (the Constitutional Court defended the Czech law on conflict of interest meant to prevent Prime Minister Andrej Babiš from using his business for political advantage and vice versa (lex Babiš) among others by highlighting common values on which the Czech and EU legal orders are based, see paras 101 et seq). In May 2021, the Czech police recommended to the prosecutor, for the second time, that Andrej Babiš should be charged with fraud. See also European Parliament resolution of 13 December 2018 on conflicts of interest and the protection of the EU budget in the Czech Republic (2018/2975(RSP)). 22 This process has recently been reversed, regarding Russia, following the revelation of extensive Russian intelligence operations on Czech soil (see the Vrbětice incident). 23 ÚsČr 29.11.1994, I. ÚS 89/94 ECLI:CZ:US:1994:1.US.89.94, N 58/2 SbNU 151. 24 ÚsČr 17.12.1997, Pl. ÚS 33/97 ECLI:CZ:US:1997:Pl.US.33.97, N 163/9 SbNU 399; and Pl. ÚS 27/09 (n 14).

Czech Republic  65 The teleological method of constitutional interpretation could not in itself make the ordinary courts to internalise the idea of Constitution’s omnipresence. The Constitutional Court, influenced by the German Federal Constitutional Court, therefore adopted a concept of constitutionally conforming interpretation. This method requires, where possible, to solve conflicts between ordinary law and constitutional law through interpretation rather than abrogation of the law (the principle of precedence of constitutionally conforming interpretation over abrogation).25 It also requires that, where several interpretations are possible, the interpretation that most closely corresponds with the Constitution must be given precedence. As mentioned above, the Constitution affords special position to international law. From Article 1(2) of the Constitution we can infer the principle of amicable relations towards obligations arising for Czech Republic from the international law. Further principles developed by the constitutional judiciary in cooperation with the theory include the principle of preservation of democracy, which prohibits an abuse of interpretation of law as stipulated by Article 9(3) ÚČr. According to the principle of plurality, which can be found in Article 22 of the Charter of Fundamental Rights and Basic Freedoms, the interpretation of norms governing political rights and freedoms must enable and preserve free competition of political forces in a democratic society. The prohibition of retroactive interpretation of the law in closed cases represents yet another fundamental principle. The principle in dubio pro libertate requires public authorities to give precedence to freedom of an individual when in doubt. Other rules of interpretation and principles, such as equality before law, can be inferred from the Constitution. For the Court’s approach, it is characteristic that it relies on the so-called ‘complex argument’. While a single interference may be constitutional, the result produced by all the interferences together may ‘in its totality’ conflict with the Constitution.26 The Constitutional Court’s case law also typically fills lacunae in the legal system, especially where there is a lack of legal instruments to remedy a case of unconstitutionality. Thus, the Court resolves the tension between gaps in the legislation and the principle of denegatio iustitiae. The Constitutional Courts admits the existence of normative unwritten constitutional principles and conventions that exist praeter constitutionem, provided that a written constitutional act does not stipulate the opposite.27 For the Court a modern democratic written constitution is a contract by which people exercising their pouvoir constituant established themselves into the state and governs the interactions among individuals and between individuals and the state authorities.28 The application of constitutional law is regularly confronted with a collision between constitutionally protected rights and freedoms. Fundamental rights and freedoms may be restricted only when they collide with another constitutionally protected value having a fundamental right or freedom nature (ie so-called public good). Resolving such conflicts requires conditions stipulating which fundamental right is to be given priority over the other. These conflicts are resolved through the principle of proportionality, which has become a fully integrated part of Constitutional Court’s reasoning.

25 This principle was first applied in ÚsČr 17.03.1998, II. ÚS 75/98 ECLI:CZ:US:1998:2.US.75.98.1, U 19/10 SbNU 407 and reappears in other rulings, for instance ÚsČr 01.12.1999, Pl. ÚS 17/99 ECLI:CZ:US:1999:Pl.US.17.99, N 174/16 SbNU 267. 26 ÚsČr 24.01.2001, Pl. ÚS 42/2000 (Elections Act) ECLI:CZ:US:2001:Pl.US.42.2000. 27 Pl. ÚS 33/97 (n 24). 28 Ibid.

66  Luboš Tichý and Tomáš Dumbrovský

C.  Constitutional Court: Its Significance and Influence As mentioned above, the Constitutional Court of the Czech Republic is a judicial body ‘responsible for the protection of constitutionality’.29 The Constitutional Court is a specialised state body of judicial nature that protects constitutionality and performs other adjudicative functions of constitutional importance. It follows from the Constitution’s wording that the Constitutional Court is not a part of the court system of the Czech Republic, although the Constitution’s general provisions governing general judiciary (Articles 81 and 82) are applicable to the Court and its judges. The Constitutional Court does not only protect the Constitution. Its powers are far more diverse. When executing its powers, the Court also reviews legality (not only constitutionality) of sub-statutory legal acts. The Constitutional Court constitutes the central body in the system of checks and balances and supervises the constitutionality and legality of legislative, executive and judicial state authorities. Constitutional review does not only protect individual rights guaranteed by the Constitution, including protection of democratic minority, but also fulfils its initial historical function – to prevent and to frustrate efforts of the legislature, of the executive or of the judiciary to gain undue power at the expense of other powers in the state. In relation to the legislature, the Constitutional Court may, upon a request submitted by an authorised party (by a requisite number of senators or deputies), review an Act of Parliament and quash it if it contradicts the Constitution. It may also quash an individual, that is non-legislative, Act of the legislature. In relation to the executive, the powers of the Constitutional Court include notably the power to quash normative and individual acts issued by the executive. With respect to the judiciary, the Constitutional Court may review decisions of ordinary courts. This review is restricted solely to the constitutionality of such decisions, that is, to their compliance with constitutionally guaranteed fair trial rights. Most constitutional complains target decisions of general courts in civil, administrative, and criminal matters, which causes a significant overload of the Constitutional Court. The Court has repeatedly emphasised its determination to minimise its interference with adjudicative practice of general courts.30 Nonetheless, its review is still rather extensive. Constitutional review is performed mostly in cases of so-called omitted evidence or when findings of general courts manifestly conflict with the established facts.31 The Constitutional Court also asserted its jurisdiction to review constitutionality of a constitutional amendment for compliance with the eternity clause, as mentioned above.32 29 Art 83 ÚČr. 30 See e.g. ÚsČr 25.01.1995, II. ÚS 45/94 ECLI:CZ:US:1995:2.US.45.94; ÚsČr 23.06.2009, II. ÚS 1634/08 ECLI:CZ:US:2009:2.US.1634.08.1; ÚsČr 07.12.2011, I. ÚS 1622/10 ECLI:CZ:US:2011:1.US.1622.10.1; ÚsČr 16.11.2011, I. ÚS 2420/11 ECLI:CZ:US:2011:1.US.2420.11.1. 31 The Constitution thus provides for both abstract review and concrete review. Under the abstract review the Constitutional Court has power to annul statutes or its individual provisions if they are in conflict with the constitutional order and to annul other legal enactments or its individual provisions if they are in conflict with the constitutional order or a statute. A petition proposing the annulment of a statute may be submitted by the President, by a group of at least 41 Deputies or a group of at least 17 Senators, by a Panel of the Court in connection with deciding a constitutional complaint, or by the Government if an international court finds that an obligation resulting for the Czech Republic from an international treaty has been infringed by the encroachment of a public authority (s 64 of the Constitutional Court Act). Under the concrete review, the Constitutional Court has a jurisdiction over constitutional complaints of natural or legal persons against final decisions or other encroachments by public authorities infringing constitutionally guaranteed fundamental rights and basic freedoms (s 72 of the Constitutional Court Act). A complainant (under concrete review) may submit, together with his constitutional complaint, a petition proposing the annulment of a statute or some other enactment, or its individual provisions, the application of which resulted in the situation which is the subject of the constitutional complaint, if the complainant alleges it to be in conflict with a constitutional act, or with a statute, where the complaint concerns some other enactment (s 74 of the Constitutional Court Act). 32 Pl. ÚS 27/09 (n 14).

Czech Republic  67

III.  Constitutional Foundations of EU Membership A.  Specific Provisions on EU Membership In anticipation of the Czech Republic’s accession to the European Union, the Constitution was amended. The proposed amendment, submitted by the Government to the Parliament in March 2001,33 contained an explicit reference to the effects of EU law in the Czech Republic. However, it was altered by the Parliament. The respective articles that were omitted from the final version read: Constitutional order, treaties, which form part of the legal order and shall have precedence over laws, laws and other legal regulations shall not be interpreted and applied to the contrary with obligations arising from the membership of the Czech Republic in an international institution …34 If the judge supposes that the law applicable to the decided matter is contrary to an international treaty, which forms part of the legal order and shall have precedence over laws, he or she shall apply provisions of such international treaty.35

Such a general interpretative clause36 with an explicit command to the judge how to solve a conflict between EU law and Czech law, if enacted, would mean giving EU law an effect fully controlled by the CJEU.37 Instead, a more moderate EU amendment was enacted. Its main tenets are found in Articles 1(2), 10, and 10a ÚČr. Unlike other EU Member States’ constitutions, the Czech EU amendment does not mention the EU as such. At the time of its enactment, the Czech Republic was in the process of ratifying the Rome Statute of the International Criminal Court, which it signed in 1999. The consequences of the accession to the EU and the ratification of the Rome Statute for Czech sovereignty were considered to be of such an extent that constitutional provisions on international law were deemed inadequate. Therefore, a special provision for accession to such types of international organisation was considered necessary.38 The EU amendment provides a general basis for accession to other international organisations, particularly organisations of regional integration whose decisions are directly applicable to Czech legal and physical persons. The EU amendment formed a basis for both, the Lisbon Treaty and the European Council Decision amending Article 136 TFEU.39 Since enactment of the EU amendment, the Constitution regulates the effect of international law in the Czech Republic by stipulating that ‘The Czech Republic shall observe its obligations resulting from international law’40 and that ‘Promulgated treaties, to the ratification of which 33 Chamber of Deputies, III. electoral term, press No 884/0. 34 Art 10b of the Government’s draft bill. 35 Art 95(2) of the Government’s draft bill. 36 See the special part of the Explanatory Report on Art 10b of the Government’s draft bill. 37 It is hard to speculate how such provision would alter the Court’s approach taken in Sugar Quotas III (ÚsČr 08.03.2006, Pl. ÚS 50/04 ECLI:CZ:US:2006:Pl.US.50.04, N 50/40 SbNU 443) and subsequent cases. Although it is possible that the Court would be able to reach its current doctrine despite such general interpretation clause (ie foremost to establish a limited conferral of competences), such conclusion of the Court would be much more contestable. 38 The Czech Republic finally ratified the Rome Statute in 2009. 39 The European Council Decision was considered a ‘presidential’ international treaty falling within Art 49 of the Czech Constitution. The ratification therefore required an assent of both chambers of Parliament and ratification in the narrow sense by the President (Art 63 ÚČr). There was little discontent about the required majority. Given that the Amendment does not transfer new competences to the EU, simple majority would suffice. However, the EC Decision was finally considered under Art 10a ÚČr (which has been used as a constitutional basis for the Lisbon Treaty) requiring a threefifths majority of all Deputies and of Senators present (Art 39 [4] ÚČr). The Parliament opted for this solution in order to build and declare broader consensus on the issue. 40 Art 1 (2) ÚČr.

68  Luboš Tichý and Tomáš Dumbrovský Parliament has given its consent and by which the Czech Republic is bound, form a part of the legal order; if a treaty provides something other than that which a statute provides, the treaty shall apply.’41 In two special provisions (Articles 10a and 10b) the Constitution provides for accession to an international organisation of supranational character and democratic control of the formation and implementation of its integrative programme. Article 10a ÚČr stipulates that ‘Certain powers of Czech Republic authorities may be transferred by treaty to an international organisation or institution’ and, in connection with Article 39(4) ÚČr, that the ratification of such a treaty requires three-fifths majorities in both chambers or ratification by referendum if a special constitutional act provides for it. Article 10b ÚČr creates an obligation for the Government to inform the Parliament on issues connected to obligations resulting from the membership in such organisation or institution and the Parliament’s right to deliberate and give its ‘views on prepared decisions of such international organisation or institution’. In line with Article 10a ÚČr a special constitutional act provided for a referendum-based ratification of the Accession Treaty.42 However, since there is no general law on referenda, any subsequent ratification of an EU Treaty in a referendum would require a special constitutional act.43 Only the Accession Treaty has so far been approved in referendum, while the Lisbon Treaty and the European Council Decision amending Article 136 TFEU were approved by the Parliament. Finally, the EU amendment afforded to the Constitutional Court a power to decide on the conformity of an international treaty with the constitutional order prior to its ratification.44 At first sight, Article 10a ÚČr authorises the transfer of competences to the EU and Article 10 in connection with Article 1(2) ÚČr regulate the effect of treaties within the Czech legal order (precedence in application based on the monist approach). However, a fierce scholarly debate on which of the two constitutional provisions would form the basis for EU law effects in the Czech legal order filled national legal journals.45 The majority of scholarship argued that Article 10a has a two-way effect regarding EU law.46 It is, at the same time, the constitutional basis for the transfer of competences to the EU institutions and the constitutional basis for validity and effects of EU law in the Czech Republic. If Article 10 in connection with Article 1(2) were to be considered the basis for EU law effects in the Czech legal order, then the EU Treaties would take precedence over Czech statutory law, though not over constitutional law. Also, such a solution 41 Art 10 ÚČr. 42 Constitutional Act concerning the Referendum on the Czech Republic’s Accession to the European Union 515/2002 Sb. 43 Art 2 (2) ÚČr provides for a possibility of general referendum. However, an implementing constitutional act, required by the Constitution, has never been adopted. 44 Art 87 (2) ÚČr. The jurisdiction extends to all international treaties ratified by the President of the Republic (that is treaties under Art 10a as well as under Art 49). It has been invoked so far only during the ratification of the Lisbon Treaty. 45 For the arguments favouring Art 10a ÚČr see Zdeněk Kühn and Jan Kysela, ‘Na základě čeho bude působit komunitární právo v českém právním řádu?’ [‘What will be the Basis for the Community Law Application in the Czech Legal Order?’], 12 Právní rozhledy (2004) 23; Zdeněk Kühn, ‘Ještě jednou k ústavnímu základu působení komunitárního práva v českém právním řádu’ [‘Once Again on the Constitutional Basis for Application of the Community Law in the Czech Legal Order’], 12 Právní rozhledy (2004) 395; Richard Král, ‘Uznání i rozpaky nad „komunitárním“ nálezem Ústavního soudu ve věci cukerných kvót’ [‘Acknowledgment and Perplexity Regarding the “Community” Decision of the Constitutional Court in Sugar Quotas’], 14 Právní rozhledy (2006) 411; Jiří Zemánek, ‘Otevření ústavního pořádku komunitárnímu právu potvrzeno, nikoli však nekontrolovatelné’ [‘Opening of the Constitutional Order to the Community Law Confirmed, Albeit Not Unconditional’] 15 Jurisprudence (2006) no 5, 47; Jiří Zemánek, ‘The Constitutional Courts in the New Member States and the Uniform Application of European Law’ in I Pernice et al (eds), The Future Of The European Judicial System in a Comparative Perspective (Baden-Baden, Nomos, 2006) 257. For the opposite side of the discussion see Jiří Malenovský, ‘Ve věci ústavního základu působení komunitárního práva uvnitř ČR nebylo řečeno poslední slovo’ [‘The Final Word Has Not Been Said Yet on the Issue of Basis for the Community Law Application in the CR’], 12 Právní rozhledy (2004) 227; Jiří Malenovský, ‘Vítězství “dogmatiků” nad “pragmatiky” se odkládá’ [‘The Victory of “Dogmatists” Over “Pragmatists” Delayed’], 13 Právní rozhledy (2005) 408; Jiří Malenovský, ‘The Relations between Constitutional Law: Community Law and International Law according to the Czech Constitution: Still Ambiguities to be Resolved’ in M Hofmann (ed), Europarecht und die Gerichte der Transformationsstaaten (Baden-Baden, Nomos, 2008) 31. 46 For the opposite account cf Malenovský, ‘Relations’ (n 45) 32 f.

Czech Republic  69 would require assessing whether all decisions of EU institutions, normative and individual, and their interpretation by the CJEU, form part of the ‘international treaty’. The Article 10a solution, on the other hand, would allow the EU legal order to function autonomously within the Czech Republic. The second option was eventually adopted by the Constitutional Court. The Court thus acknowledged the specific nature of EU law vis-à-vis international law in line with the established doctrine of the CJEU, which differentiates the Community legal order from international law to prevent the dependence of Community law application in the Member States on their internal rules on international law transposition.47

B.  Relevance of General Provisions on International Law and Treaties 1.  Duty to Respect Obligations Arising from Treaties The EU-amendment of the Constitution not only created a special regime for treaties requiring a transfer of competences from the Czech Republic, but also changed most constitutional provisions governing the effects of international law in the Czech legal order. From the constitutional prescription that the Czech Republic observes its obligations resulting from international law, the Court infers that EU law has to be given full effect in the Czech legal order. That means foremost that all structural principles of EU law have to be fully operational and would unilaterally solve any conflict with a Czech legal norm. However, such an uncontrollable effect of EU law is not, in the Court’s view, in compliance with other requirements of the Czech constitutional order and the Court has thereby posed certain limits to such uncontrollable effects of EU law. As a result, the Court came to the conclusion that EU law and its principles ‘radiate’ through the Czech constitutional order, as mentioned above. Such interpretation, to a certain extent, weakens the impact of Article 1(2) ÚČr. We argue that the view of the Court basing the effects of EU law in the Czech legal order on Article 10a ÚČr rather than the general provisions on the role of international law in the Czech legal order is correct, albeit for different reasons. The fact that Article 1(2) ÚČr obliges the Court (and all other state authorities) to observe obligations arising from ratified treaties, including decisions of authorities of an international organisation based on such treaties, cannot mean that an entire EU legal order is taken over to become part of the Czech constitutional order (since, formally, only then no breach of a treaty occurs). Many of the obligations stemming from EU law are of far lesser importance than constitutional norms (for instance, the scheme for assigning sugar quotas). The Czech Republic is then ‘merely’ obliged to observe such obligations. The provision of Article 1(2) ÚČr is, therefore, only secondary for giving EU law effect in the Czech Republic. Despite the Court’s struggle with this issue, it reached a similar conclusion on the consequences of the duty to observe treaties. The Court then couples this argument with its reasoning on other substantive provisions (especially Article 10a ÚČr).

2.  Constitutional Court Jurisdiction over International Treaties As mentioned above, the EU amendment of the Constitution entrusted the Constitutional Court with reviewing the constitutionality of treaties. Prior to the ratification of a qualified treaty (a

47 ECJ 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66; ECJ 07.03.1985 32/84 (Van Gend & Loos v Inspecteur der Invoerrechten en Accijnzen) ECLI:EU:C:1985:104.

70  Luboš Tichý and Tomáš Dumbrovský treaty transferring powers to an international organisation or institution and treaties requiring ratification of Parliament), the Constitutional Court has jurisdiction to decide on the treaty’s constitutionality. A treaty may not be ratified until the Constitutional Court gives its judgment.48 A petition to the Constitutional Court may be submitted by (1) one of the chambers of Parliament, as of the moment when the treaty is submitted to it for its consent, until the moment when the treaty receives that consent; (2) a group of at least 41 Deputies or a group of at least 17 Senators, from the moment when the Parliament has given its consent, until the moment when the President of the Republic ratifies the treaty; (3) a group of at least 41 Deputies or a group of at least 17 Senators, from the declaration of the results of a referendum in which consent to the ratification is given, until the moment when the President of the Republic ratifies the treaty; and (4) the President of the Republic, from the moment when the treaty was submitted to him for ratification.49 The Court reviews only provisions of the treaty that are contested by the petitioner, not the treaty in its entirety. That means, a decision on compliance with the constitutional order of a treaty does not represent res iudicata for the whole treaty and other authorised petitioners may submit their own petitions with different argumentation contesting the same provisions or for provisions not yet contested. So far, only the Lisbon Treaty was subject to such a review (twice).50

C.  Judicial Practice and Doctrinal Debate 1.  Formation of the Constitutional Court’s European Doctrine In the pre-accession period, the Court dealt with EU law indirectly through the application of the Association Agreement. It took a favourable approach to EU law and its interpretation by the CJEU.51 The first case in which the Constitutional Court was confronted with the new situation of direct application of EU law post-accession was the Sugar Quotas III case.52 The Czech Republic had regulated the sugar market for several years, taking into account relevant EU regulations. However, EU law did not provide for a ‘year zero’ allocation of sugar quotas and the Government, when issuing the first national regulation on the matter, had to decide on a formula for quotas allocation and implicitly decide which companies would have access the market. This formula was repeatedly found unconstitutional.53 Despite the Constitutional Court’s decisions, the Government continued to derive the formula for quota allocation from the production results of companies in the years where unconstitutional governmental orders were in force. This approach did not allow for a factual change as the companies that were discriminated under the unconstitutional quota system continued to be eo ipso discriminated under every subsequent system.

48 Art 87(2) ÚČr. 49 S 71a of the Constitutional Court Act. 50 ÚsČr 26.11.2008, Pl. ÚS 19/08 (Lisbon Treaty I) ECLI:CZ:US:2008:Pl.US.19.08.1, N 201/51 SbNU 445; and ÚsČr 03.11.2009, Pl. ÚS 29/09 (Lisbon Treaty II) ECLI:CZ:US:2009:Pl.US.29.09.3, N 233/55 SbNU 197. The submission of the Accession Treaty for the review was rejected by the Court (ÚsČr 04.03.2004, Pl. ÚS 1/04 [Accession Treaty] ECLI:CZ:US:2004:Pl.US.1.04, U 11/32 SbNU 519). 51 Cf ÚsČr 29.05.1997, III. ÚS 31/97 (Škoda auto) ECLI:CZ:US:1997:3.US.31.97, N 66/8 SbNU 149; and ÚsČr 16.10.2001, Pl. ÚS 5/01 (Milk Quotas) ECLI:CZ:US:2001:Pl.US.5.01, N 149/24 SbNU 79. 52 Pl. ÚS 50/04 (n 37). 53 ÚsČr 14.02.2001, Pl. ÚS 45/2000 (Sugar Quotas I) ECLI:CZ:US:2001:Pl.US.45.2000, N 30/21 SbNU 261; and ÚsČr 30.10.2002, Pl. ÚS 39/01 (Sugar Quotas II) ECLI:CZ:US:2002:Pl.US.39.01 N 135/28 SbNU 153. Cf also Pl. ÚS 5/01 (n 51).

Czech Republic  71 In the Sugar Quotas III case, the Court was confronted with the fact that upon the accession, it was an EC regulation that directly regulated the sugar market. Although its previous case law would dictate to strike down the current allocation system as well, since the discriminatory effect in favour of ‘year zero’ companies persisted and functioned as a barrier to entry, the Court re-evaluated its previous reasoning in order to give full effect to the directly applicable EC regulation and concluded that in the case of economic and social policy decisions the consequences of the equality clause of the Constitution54 for the right to pursue economic activity55 and the right to property56 were weakened and the (Community) legislator enjoyed broader discretion. The Court, however, struck down the governmental order implementing the EC Regulation on the organisation of sugar markets. Finding no authorisation for delegation in the Regulation, the Court abrogated the national implementation measure on the basis of unlawful implementation of a directly applicable Regulation based on the Variola doctrine.57 Constitutional legal basis was found through the rationale that the Government acted in the matter where it had no competence to act because this competence was transferred to the Community by the Accession Treaty. The Government, therefore, acted beyond its constitutional competences, that is, ultra vires. A subsequent case concerned the enlisting of medical products that were reimbursed from public health insurance, a matter regulated by the Directive 89/105/EEC.58 The national law required the Ministry of Health to issue a regulation in which it enumerated the medications that were fully reimbursed from health insurance, as well as the level of reimbursement for all individual medicinal substances. Pharmaceutical companies that sought to enlist their products were given an opportunity to give their views on the classification process. However, since the list of reimbursed medicines was enacted as an annex to a governmental regulation, and was, therefore, enacted in legislative process and not reached in the context of administrative proceedings, pharmaceutical companies had no recourse to judicial review. The Constitutional Court asserted that decisions on the reimbursement of medications were not founded on objective criteria, substantiated, and subjected to judicial review. It concluded that the respective provisions of the law were both substantively unconstitutional and in breach of the Constitution by ignoring the Directive59 that stipulated requirements for transparency in the listing process and for remedies available to the persons who sought the enlisting of their medicinal products.60 In these two early decisions, the Court developed its ‘European’ doctrine. It asserted that (1) principles (and other norms) of Community law ‘radiate through its interpretation of constitutional law’61 and (2) these principles shall be applied as interpreted by the CJEU, however, (3) this radiating effect is limited by the eternity clause of the Constitution.62 In the Reimbursement 54 Art 1 of the (Czech) Charter of fundamental rights and freedoms. 55 Art 26(1) of the Charter of fundamental rights and freedoms. 56 Art 11 of the Charter of fundamental rights and freedoms. 57 ECJ 10.10.1973 34/73 (Fratelli Variola S.p.A. v Amministrazione italiana delle Finanze) ECLI:EU:C:1973:101 1973. 58 ÚsČr 16.01.2007, Pl. ÚS 36/05 ECLI:CZ:US:2007:Pl.US.36.05.1, N 8/44 SbNU 83. 59 Council Directive 89/105/EEC of 21 December 1988 relating to the transparency of measures regulating the prices of medicinal products for human use and their inclusion in the scope of national health insurance systems, [1989] OJ L 40/8. 60 Art 6(2) of the Directive requires that a decision not to include a medicinal product in the list must contain a statement of reasons based upon objective and verifiable criteria, including, where necessary, any expert opinions or recommendations on which the decision is based. In such cases, the applicant should be informed of remedies available to her. 61 The Court reasoned that ‘[a]lthough [its] referential framework has remained, even since 1 May 2004 [date of ­accession], the norms of the Czech Republic’s constitutional order, the Constitutional Court cannot entirely overlook the impact of Community law on the formation, application, and interpretation of national law, all the more so in a field of law where the creation, operation, and aim of its provisions are immediately bound up with Community law. In other words, in this field the Constitutional Court interprets constitutional law taking into account the principles arising from Community law’, Pl. ÚS 50/04 (n 37). 62 Ibid.

72  Luboš Tichý and Tomáš Dumbrovský of Medications case, the Constitutional Court applied its Sugar Quotas III reasoning in a moderate way. As the Directive did not deal with other questions on the complex process of reimbursement of medications, the Court did not base its reasoning solely on the primacy of EU law and direct effect of the Directive, but coupled it with the substantive requirements of due process stipulated in the Constitution and its own case law. As mentioned above, the Constitutional Court resolved the discussion on the constitutional basis for EU law effects in the Czech legal order in favour of Article 10a ÚČr and the autonomous functioning of the EU legal order within its jurisdiction. It claimed the authority to resolve conflicts between EU law and national law from the standpoint of the Constitution. On the one hand, it afforded to EU law a constitutional effect through the radiation rationale. On the other hand, the Court asserted that it would review the EU law effects from the eternity clause standpoint (counter-limits). Despite accepting the autonomous nature of the EU legal order, the Court reasoned that upon accession, ‘the Czech Republic took over into its national law the entire mass of European law’.63 According to the Constitutional Court, the material core of the Constitution that is protected by the eternity clause includes also the protection of fundamental rights and freedoms. In line with the German Constitutional Court’s Solange doctrine, it asserted that if the standard of protection in the EU was unsuitable, the Czech state institutions would have to take over the transferred powers in order to ensure that the standard of protection was observed. In the Arrest Warrant case, the Constitutional Court concluded that ‘The contemporary standard for the protection of fundamental rights within the European Union does not give rise to any presumption that this standard … is of a lesser quality than the level of protection provided in the Czech Republic.’64 As in the previous two cases, the Court reiterated that the constitutional prescription to observe obligations under international law, in conjunction with the EU principle of loyal cooperation, requires that ‘domestic legal enactments, including the constitution, should be interpreted in conformity with the principles of European integration and the cooperation between Community and Member State organs’.65 This principle of Euro-conforming interpretation, which again is rooted in the Court’s similar ‘national’ doctrine of constitutionally conforming interpretation, requires that when the Constitution ‘can be interpreted in several manners, only certain of which lead to the attainment of an obligation which the Czech Republic undertook in connection with its membership in the EU, then an interpretation must be selected which supports the carrying out of that obligation, and not an interpretation which precludes it’.66 In the first constitutionality review of a treaty prior to its ratification, which concerned the Lisbon Treaty, the Court broadened its referential framework. It applied, as a point of reference, the constitutional order of the Czech Republic as a whole, not only its material core. The arguments were: (1) the subject of review is EU primary law, not EU secondary law as in the previous proceedings; (2) the Lisbon Treaty was not yet valid law and in its review the Court is not bound by the obligation of Euro-conforming interpretation; (3) an Article 10a treaty requires the same majority as is required for a constitutional amendment; (4) to limit the review only to the material core of the Constitution would, to a large extent, deprive the institution of preliminary constitutional review of treaties of its meaning; (5) the Constitution does not distinguish between ‘ordinary’ treaties under Article 49 and treaties under Article 10a; (6) the text of Article 87(2) ÚČr explicitly envisages the whole constitutional order as a point of reference for this review, not only its material core; (7) Parliament’s role is limited as it can only approve or disapprove the international treaty.

63 Ibid,

S A-3, para 4. 03.05.2006, Pl. ÚS 66/04 (European Arrest Warrant) ECLI:CZ:US:2006:Pl.US.66.04, N 93/41 SbNU 195.

64 ÚsČr 65 Ibid. 66 Ibid.

Czech Republic  73 This special review standard is confined to ex-ante constitutional review of a treaty. The Court, nevertheless, emphasised that the ultimate limits rest in the material core of the Constitution. The Court gave a limited answer to the question what is the content of the constitutional core and avoided a general definition. According to the Court, it follows from Article 10a(1) that only certain powers may be transferred by a treaty to an international organisation. This provision must be interpreted in connection with Article 1(1) and Article 9(2) ÚČr: The transfer of powers cannot go as far as to violate the very essence of the Czech Republic as a sovereign and democratic rule-of-law-based state, founded on respect for the rights and freedoms of human beings and of citizens, or to change these essentials. The Court admitted that the notion of a ‘sovereign state’ is not uncontested and it is difficult to give an abstract definition.67 Nevertheless, the transfer of sovereign powers under the Constitution has certain limits:68 These limits should be left primarily to the legislature to specify, because this is a priori a political question which provides the legislature wide discretion; interference by the Constitutional Court should come into consideration as ultima ratio, i.e., in a situation where the scope of discretion was clearly exceeded, and Article 1 paragraph 1 ÚČr was affected, because there was a transfer of powers beyond the scope of Article 10a ÚČr.69

The culmination of the Constitutional Court’s ultra vires doctrine came in the Slovak Pensions XVII case.70 After a long saga of struggles between the Constitutional Court, on the one side, and the Government and the Supreme Administrative Court, on the other, on whether Czech nationals receiving pensions from Slovakia for work performed during the existence of the common state, which are lower than the pensions paid by the Czech Republic in similar circumstances, are unconstitutionally discriminated against, the Supreme Administrative Court (SAC) involved the CJEU by submitting a preliminary question. Since 2003, the Constitutional Court has repeatedly ruled that Czech citizens with Slovak pensions resulting from work in the common Czechoslovak state are discriminated against in comparison with Czech citizens receiving Czech pensions for the same period.71 The CJEU’s ruling in the Landtová case seemed to infuriate the Constitutional Court, which plausibly considered EU law inapplicable.72 Consequently, the Constitutional Court proclaimed the ruling of the CJEU ultra vires and without any effect in the Czech Republic. The problem arose after the dissolution of the Czech and Slovak Federal Republic (CSFR). In the form of a treaty, the two succeeding states, the Czech Republic and the Slovak Republic, agreed to divide the financial burden of paying pensions according to the location of the employer at the

67 Petr Bříza, ‘The Constitutional Court on the Lisbon Treaty Decision of 26 November 2008’, 5 European Constitutional Law Review (2009) 143, 149 f. 68 The German Federal Constitutional Court in its Lisbon Treaty Judgment showed that a clearer delimitation of conditions (based on the evaluation of transferred competences), after which the sovereignty would be lost, was possible (BVerfG 30.06.2009 2 BvE 2/08 [Lissabon] ECLI:DE:BVerfG:2009:es20090630.2bve000208). 69 Pl. ÚS 19/08 (n 50) para 109. Cf Jiří Zemánek, ‘Přezkum ústavnosti Lisabonské smlouvy: obsahové otázky’ [‘Constitutionality review of the Lisbon Treaty: Questions of the Content‘], 18 Jurisprudence (2009) no 1, 32, 34 f. 70 ÚsČr 31.01.2012, Pl. ÚS 5/12 (Slovak Pensions XVII) ECLI:CZ:US:2012:Pl.US.5.12.1, N 24/64 SbNU 237. 71 ÚsČr 03.06.2003, II. ÚS 405/02 (Slovak Pensions I) ECLI:CZ:US:2003:2.US.405.02, N 80/30 SbNU 245. The Constitutional Court consolidated its case law on Slovak Pensions in the decision of the full Court ÚsČr 20.03.2007, Pl. ÚS 4/06 (Slovak Pensions V) ECLI:CZ:US:2007:Pl.US.4.06.1, N 54/44 SbNU 665, sanctioning the controversial idea of a supplement (constituted as obiter dicta already in its decision ÚsČr 25.01.2005, III. ÚS 252/04 [Slovak Pensions III] ECLI:CZ:US:2005:3.US.252.04, N 16/36 SbNU 173, part II, para 15), despite this supplement having no grounds in statutory law. In the Constitutional Court’s view, the Czech Social Security Authority (CSSA) was obliged to mitigate this discrimination by topping up pensions paid by the Slovak Social Security Authority to the Czech citizens so that the overall pension became what it would have been had the person entitled worked for an employer based in the Czech part of Czechoslovakia; a solution, the Czech Supreme Administrative Court has consistently opposed. 72 CJEU 22.06.2011 C-399/09 (Marie Landtová v Česká správa socialního zabezpečení) ECLI:EU:C:2011:415.

74  Luboš Tichý and Tomáš Dumbrovský time of the CSFR dissolution.73 This solution led to some unreasonable results. For example, an entitled person might have worked all his life in the Czech part of the CSFR for an employer based in the Slovak part of the CSFR, without being aware of the consequences for his future. After accession to the EU, the issue of who pays the pensions within the internal market was regulated by the EC regulation,74 which superseded bilateral agreements between Member States with several, often historically rooted, exemptions, one of them being Article 20 of the 1992 Czech-Slovak Agreement specifying the employer’s registered office as the decisive factor for determining which state is responsible for pensions. The responsible Czech agency opined that the EC regulation precluded it from paying the supplement. On a preliminary question submitted by the SAC, the CJEU held that the case law of the Czech Constitutional Court entitling Czechs with Slovak pensions to the supplement was discriminatory, since only Czech residents75 would receive such pensions.76 Although the SAC initially submitted another preliminary question,77 the legislator eventually adopted a law that gives right to the supplement to every person who is entitled to both Czech and Slovak pensions.78 Although the Slovak Pension XVII case was the first case where a national constitutional court pronounced an Act of an EU institution ultra vires, its jurisprudential value is of little significance.79 The conflict between the CJEU and the Constitutional Court was collateral damage in the struggle between the SAC and the Constitutional Court,80 and partly a result of miscommunication between the CJEU and the Constitutional Court. While the CJEU refused to allow the Constitutional Court to defend its views in the preliminary proceeding initiated by the SAC, the Constitutional Court did not consider submitting a preliminary question on its own (as a result of its doctrine regarding the preliminary question procedure, which we describe below).

2.  Constitutional Court and the Preliminary Question Procedure The Constitutional Court has been reluctant to accept its obligations under the preliminary question procedure. In the Sugar Quotas III case,81 the Court was confronted with the question, but 73 Art 20(1) of the Agreement on Social Security of 29 October 1992. 74 Art 3(1), Regulation (EEC) No 1408/71 of the Council of 14 June 1971 on the application of social security schemes to employed persons and their families moving within the Community [1971] OJ L 149/2. 75 The Constitutional Court was not entirely clear whether the nationality or residence is the criterion; however, the Court of Justice found discrimination in either case. 76 CJEU C-399/09 (n 72). See Richard Pomahač, ‘Soudní dvůr EU: Diskriminační podmínka pobytu v důchodovém pojištění’ [‘CJEU: Discriminatory Condition of Residence in Pension Insurance’] (2012) 20 Právní rozhledy 640; Pavel Kantořík, ‘Ukončí zákon č. 428/2011 Sb. letitý spor o československé důchody?’ [‘Would the Law No. 428/2011 Coll. Resolve the Long-term Conflict on Czechoslovak Pensions?’], 20 Právní rozhledy (2012) 254; Jan Komárek, ‘Slovenské důchody: druhá „válka soudů“ před Soudním dvorem EU’ [‘Slovak Pensions: Second “War of Courts” Before the CJEU’], 17 Soudní rozhledy (2011) 388; Filip Křepelka, ‘Českoslovenští důchodci v pasti práva Evropské unie’ [‘Czechoslovak Pensioners in the Trap of EU Law’], 19 Časopis pro právní vědu a praxi (2011) 131; Robert Zbíral, ‘Czech Constitutional Court, judgment of 31 January 2012, Pl. ÚS 5/12 – A Legal revolution or negligible episode? Court of Justice decision proclaimed ultra vires’, 49 CML Rev (2012) 1475. 77 The SAC asked a quite dangerous question: whether EU law precludes a Member State’s supreme court from being bound by a decision of the Member State’s constitutional court that is not in compliance with EU law as interpreted by the CJEU. However, the reference was later withdrawn. Decision of the SAC, 6 Ads 18/2012–82 ECLI:CZ:NSS:2012:6. ADS.18.2012, CJEU 27.03.2013 Order C-253/12 ECLI:EU:C:2013:212. 78 The law abandons any reference to nationality or residence and grants the entitlement to anybody who receives a Czech pension and simultaneously receives a Slovak pension for his work in the CSFR. S 106a-c of the Law No 155/1995 Sb as amended by Law No 274/2013 Sb. This solution basically entitles the same group of people who would be entitled to the supplement if the nationality/residence requirement were retained. 79 Despite being the only precedent in the EU, the judgment was not mentioned by the German Constitutional Court in its PSPP judgment; BVerfG, 05.05.2020, 2 BvR 859/15, ECLI:DE:BVerfG:2020:rs20200505.2bvr085915. 80 See the works of Zbíral and Komárek quoted in n 76. 81 Pl. ÚS 50/04 (n 37).

Czech Republic  75 managed to avoid a direct answer. It asserted that the issue at hand was, nevertheless, already clarified by the CJEU, and therefore the acte clair doctrine82 shall be, in eventum, applied.83 Soon the Court was confronted with the question directly. In the Pfizer case84 the petitioner argued that Supreme Administrative Court should have applied EU law85 and asserted that the SAC breached the constitutional right to fair trial under Article 36(1) of the Charter of Fundamental Rights and Freedoms by not submitting a preliminary question to the CJEU. Instead of the fair trial rationale, the Constitutional Court opted for the more specific constitutional provision on lawful judge,86 similar to the doctrine of the German Constitutional Court. The Court asserted that the function of the Constitutional Court is to protect constitutionality, and EU law is not part of Czech constitutional law and therefore does not form a referential framework for the Court. It is not the task of the Constitutional Court therefore to supervise whether national law is in conflict with EU law. This is the task of ordinary courts, which are bound by the cogent provisions of EU law, which form for them part of applicable law. It is up to ordinary courts to identify the conflict and apply a mechanism of how to solve it (in general to give primacy to EU law and in the case an interpretation of applicable EU law is needed to turn to the CJEU to obtain an authoritative interpretation).87 The Constitutional Court would intervene only when the breach of EU law is flagrant and serious, especially when the ordinary court disregards a cogent provision of EU law in such a way that the rule of law is in jeopardy.88 The Constitutional Court refuses to substitute ordinary courts in their obligation to refer a preliminary question. However, the Constitutional Court feels obliged to intervene when disregard of EU law is of such salience that it affects constitutionality of the Czech legal order as such, including arbitrariness in deciding whether to submit a preliminary question by ordinary courts.89 While the Constitutional Court agrees that ordinary courts are bound by the requirements of the preliminary ruling procedure, at the same time it tacitly denies being subject to the very same requirement of the Treaty. The reason is found in that the Constitutional Court assumes a specific role in the legal system. It supervises the constitutionality under its jurisdiction and a question of EU law cannot as a rule arise in front of it. EU law will play a role in the Court’s reasoning only insofar as the matter touches upon constitutionality. 82 The Court should have referred rather to the acte éclairé doctrine. 83 The Court seemed bothered little with the fact that an application of an exemption (acte éclairé doctrine) may follow only after the rule (that is that the Court is a referring court) is accepted. 84 ÚsČr 08.01.2009, II. ÚS 1009/08 (Pfizer) ECLI:CZ:US:2009:2.US.1009.08.2, N 6/52 SbNU 57. The case involved a registration of the medical product of a Pfizer competitor, which Pfizer deemed a copy of its own product, but was denied standing in the administrative proceeding on the registration. 85 Directive 2001/83/EC of the European Parliament and of the Council of 6 November 2001 on the Community code relating to medicinal products for human use, [2001] OJ L 311/67. 86 Art 36(1) ÚČr (fair trial) reads: ‘Everyone may assert, through the legally prescribed procedure, his rights before an independent and impartial court or, in specified cases, before another body’. Art 38(1) ÚČr (lawful judge) reads: ‘No one may be removed from the jurisdiction of his lawful judge. The jurisdiction of courts and the competence of judges shall be provided for by law’. Cf also the brief concurring opinion of Judge Dagmar Lastovecká in the Pfizer case, who insisted on a solution within the fair trial rationale rather than the lawful-judge one. 87 ÚsČr 02.12.2008, Pl. ÚS 12/08 (Application Primacy) ECLI:CZ:US:2008:Pl.US.12.08.1, U 12/51 SbNU 823. 88 ‘Although the referral of a preliminary question is a Community law matter, the failure, in conflict with Community law, to make a reference may, in certain circumstances, also entrain a violation of the constitutionally-guaranteed right to one’s statutory judge … A violation of the right to one’s statutory judge comes about in the case where a Czech court (against whose decision there is no longer any further remedy afforded by sub-constitutional law) applies Community law but fails, in an arbitrary manner, that is, in conflict with the principle of the law-based state (Art 1 para 1 ÚČr), to refer a preliminary question to the ECJ’, II. ÚS 1009/08 (n 84) para 21. 89 ‘The Constitutional Court asserts that it deems as arbitrary action such conduct by a court of last instance applying a norm of Community law where that court has entirely omitted to deal with the issue whether it should refer a preliminary question to the ECJ and has not duly substantiated its failure to refer, including the assessment of the exceptions which the ECJ has elaborated in its jurisprudence’, ibid para 22. See also ÚsČr 07.12.2020, I. ÚS 865/20 (on the reasoning for non-referral of a preliminary question to the CJEU) ECLI:CZ:US:2020:1.US.865.20.1.

76  Luboš Tichý and Tomáš Dumbrovský

D.  European Integration Outside the EU Legal Order The Czech Republic is not a member of the eurozone and is not a party to the ESM Treaty. The Constitutional Court, therefore, has not been given an opportunity to evaluate how such integration measures taken outside the EU legal order affect its ‘European’ doctrine. Preparing for ratification of the Fiscal Compact in 2014, the Czech Government and the Parliament concluded that, despite the fact that the Czech Republic did not intend to be bound by parts of the Fiscal Compact, on the basis of its Article 14(5) that is otherwise applicable to the euro area only, the Treaty shall be considered an integrative treaty in the sense of Article 10a ÚČr and ratified by constitutional majorities in both chambers of Parliament. The Czech legal academic discussion focused on whether the Fiscal Compact means a transfer of competences to an international organisation or institution. The Fiscal Compact has been found to create new competences, in particular the commitment of contracting parties to support the proposals and recommendations of the Commission prepared within the excessive deficit procedure.90 At the same time, the Fiscal Compact contains decision-making rules that as a consequence create a specific international institution consisting of cooperation between the Commission and the contracting parties. The conclusion was that the Fiscal Compact conferred new competences to an international institution, which needed to be transferred by the state, within the excessive deficit procedure that went beyond the mandate of the TFEU, the Protocol on the Excessive Deficit, and the Stability and the Growth Pact, rather than solely implementing competences already transferred to the EU. As a consequence, the Fiscal Compact required ratification by a three-fifths majority of both chambers of Parliament or through a referendum.91 The Fiscal Compact was approved by the Senate in 2014. However, the Chamber of Deputies sat on it until the ratification process eventually expired with the end of its electoral term. It was not until February 2018 that the Government adopted a new resolution on the accession to the Fiscal Compact and submitted the treaty to the Parliament for ratification. This time, the Government considered the Fiscal Compact an ordinary treaty, requiring a simple majority. This change appears to be reflective of the change in Parliament composition after the 2017 elections and a lack of three-fifths-majority support for the Fiscal Compact. This might create application problems once the Czech Republic accedes to the eurozone and the hitherto latent provisions of the Fiscal Compact become applicable.

IV.  Constitutional Limits to EU Integration A.  Limits to (Further) Transfer of Powers to the EU through Treaty Amendment One of the Constitution’s principles is the irreversibility of its core provisions. The Constitutional Court in its first ruling on an abstract review of legal norms set out the fundamentals on this issue.92 90 Art 7 TSCG. 91 See inter alia Aleš Gerloch, ‘Povaha Smlouvy o stabilitě, koordinaci a správě v Hospodářské a měnové unii sjednávané mezi zeměmi Eurozóny z hlediska požadavků stanovených ústavním pořádkem České republiky na proces její ratifikace’ [‘The Character of the TSCG in the EMU Negotiated Among the Eurozone Countries from the Perspective of Conditions Required by the Constitutional Order of the Czech Republic for the Ratification Process’], 58 Acta Universitatis Carolinae – Iuridica (2012) 99, 104. 92 ÚsČr 21.12.1993, Pl. ÚS 19/93 ECLI:CZ:US:1993:Pl.US.19.93, N 1/1 SbNU 1.

Czech Republic  77 It held that constitutive principles of a democratic society fell beyond the powers of the legislature (were ultra vires). Hence any effort of the Parliament to remove this principle regardless of the will of the majority must be interpreted as an effort to remove the constitutional state as such. In 2002 the Court invoked Article 9(2) ÚČr as a fundamental principle of constitutional interpretation bearing consequences for both the legislature and the Constitutional Court.93 The principle of impermissibility of amendments to fundamental features of a democratic state governed by rule of law prevents the Constitutional Court from interpreting subsequent constitutional amendment in a manner decreasing the already achieved level of protection of fundamental rights and freedoms. The Constitutional Court held that as a constitutional body it may not act arbitrarily and held itself bound by its own rulings, which may be overruled only in exceptional circumstances. This principle forms a fundamental element of democratic state governed by rule of law. Subsequently, the Constitutional Court held that the essential features of the right to vote formed one of these elements as well. Several conclusions may be drawn from this. First, the Constitutional Court clearly articulated the necessity to protect the material core of the constitutional order. Second, it gradually delimitated its scope on a case-by-case basis. Third, it expressly drew consequences from this not only for the legislature, but also for itself. The Court developed its material constitutional core test in the Sugar Quotas III judgment: [The] conferral of a part of its powers is naturally a conditional conferral, as the original bearer of sovereignty, as well as the powers flowing therefrom, still remains the Czech Republic … [T]he conditional nature of the delegation of these powers is manifested on two planes: the formal and the substantive plane. The first of these planes concerns the power attributes of state sovereignty itself, the second plane concerns the substantive component of the exercise of state power. In other words, the delegation of a part of the powers of national organs may persist only so long as these powers are exercised in a manner that is compatible with the preservation of the foundations of state sovereignty of the Czech Republic, and in a manner which does not threaten the very essence of the substantive law-based state … According to Art. 9 par. 2 of the Constitution of the Czech Republic, the essential attributes of a democratic state governed by the rule of law, remain beyond the reach of the Constituent Assembly itself.94

While Article 10a ÚČr provides for transfer of competences to the EU, the Court in the quoted crucial part of its judgment talks about conferral, indicating that there is a somewhat closer bond between the competences so conferred and the Czech Republic than the Constitution suggests. In the Court’s view, the competences are only ‘conditionally delegated’ to the EU institutions. The Court seemed to attempt to put itself in a better position for the supervision of the exercise of such ‘conferred’ competences whose ‘original bearer’ remains the Czech Republic. The material core of the Constitution protects the nature of the Czech Republic as a democratic, sovereign state based on the rule of law and respecting fundamental rights and freedoms, which cannot be changed. The core’s content is defined by Article 1(1) ÚČr and its effects (impermissibility of change) by Article 9(2) ÚČr (the eternity clause). The Court has acknowledged that the level of human rights protection in the EU is comparable to the level of protection afforded by the Czech Constitution. However, it did not exclude the possibility of ad-hoc evaluations but has equally not set conditions under which it would engage in such review (as would fully correspond to Solange II doctrine of the German Federal Constitutional Court).95



93 ÚsČr 94 Pl. 95 Pl.

25.06.2002, Pl. ÚS 36/01 ECLI:CZ:US:2002:Pl.US.36.01, N 80/26 SbNU 317. ÚS 50/04 (n 37) Part A-3, B. ÚS 19/08 (n 50) para 196.

78  Luboš Tichý and Tomáš Dumbrovský The other essential requirements, such as democracy, sovereignty, and rule of law are developing concepts in the Court’s view and the Court would evaluate them on a case-by-case basis. This provides the Court with considerable flexibility for future re-evaluations without constraining itself to its own conditions for a reversal of its jurisprudence. This openness of the content of the material core of the Constitution at the same time means that there are no concrete limits to further European integration, in the sense of the German Federal Constitutional Court’s recent jurisprudence.96 The Constitutional Court in its Lisbon Treaty I judgment, nevertheless, underscored that Article 10a ÚČr and the eternity clause authorised only transfer of certain competences, and therefore competences cannot ever be given up fully. The general openness of the Czech constitutional system, through Article 10a ÚČr, to integrative projects between states means that there are no apparent reasons why the established ‘European’ doctrine on transfer of competences to an international organisation or institution (Article 10a ÚČr), role of treaties in the constitutional order (Article 1[2]), and eternity clause (Article 9[2]) should not apply to integration outside the EU legal order as well. Given that the Court in the European Arrest Warrant case applied its ‘European’ doctrine without much alteration to the then third, prevailingly intergovernmental, pillar of the EU only confirms this view. In sum, the same limits that apply to further integration within the EU legal order shall apply to any integration outside the EU legal order. The Constitutional Court has afforded itself a power to review comparability of the level of protection of fundamental rights and freedoms and it might invoke it if such integrative organisation does not have sufficient instruments, specifically judicial review mechanisms, and at the same time is able to directly affect individual rights of legal or natural persons. Another issue that might concern the Court, which has been repeatedly looked to the German Federal Constitutional Court’s doctrines for inspiration, is the right to vote and the necessity to increase a democratic input in the decision-making in the areas of European integration that fall outside of the EU legal order. This may be an issue due to the Court’s view that the essential requirements of the right to vote formed one of the elements of the material core of the Constitution. However, as neither the ESM, nor the Fiscal Compact have been ratified in the Czech Republic, the Constitutional Court did not have an opportunity to pronounce itself on the matter. This fact has also impacted the intensity of academic discussion.

B.  Scrutiny of Secondary Legislation and the Ultra Vires Doctrine As discussed above, the constitutional basis of the ultra vires doctrine is the same regarding acts of Czech authorities and EU institutions. It rests in the material core of the Constitution, change to which is impermissible. The essential requirement that is breached in such circumstances is the rule of law. Public power authorities may act only when they are endowed with a competence. However, not every breach of this requirement is as grave as to reasonably threaten the rule of law. Therefore, only flagrant and serious ultra vires acts will trigger constitutional review. The relatively high standard for triggering the review is warranted by the existence of other moderation of public power excesses concerning EU law. In the case of a conflict between EU law and an Act of Czech authorities, the latter act is inapplicable. If such solution suffices and the inapplicability does not undermine the principle of legal certainty, the Constitutional Court would

96 Cf BVerfG, 2 BvE 2/08 (n 68) and BVerfG 06.07.2010 2 BvR 2661/06 (Honeywell) ECLI:DE:BVerfG:2010:rs20100706. 2bvr266106.

Czech Republic  79 not intervene. In the case of acts of EU institutions, the EU legal order has its own mechanisms to review legality of the acts. The Constitutional Court intervenes only when such mechanisms fail. The ultra vires doctrine is a specific standard for addressing the most flagrant excesses of competences by public power authorities. Most lesser violations are reviewed by ordinary courts, by either applying ordinary law (since most of the violations would breach competences granted by a statute) or by applying the Constitution directly. The Constitutional Court therefore intervenes, as in all other situations, only when the violation threatens the constitutional order.

V.  The Constitution and the Implementation of EU Law A.  Decision of the Parliament or of the Government Binding Ministers Attending Meetings of the Council 1.  Role of the Parliament a.  Relevant Regulation The relationship between the Parliament and ministers of the Government regarding EU decision-making is regulated by Article 10b ÚČr, rules of procedure of each chamber of Parliament, and by the Government Directive on Dealing with EU Documents that form an annex to the Government Resolution No 665 from 6 August 2014. Article 10b ÚČr obliges ‘the Government to inform the Parliament, regularly and in advance, on issues connected to obligations resulting from the Czech Republic’s membership [of the EU]’. The chambers of Parliament shall give their views on prepared EU decisions in the manner laid down in their rules of procedure. The Constitution also envisages that a statute that governs the cooperation between the chambers and their external relations may entrust these competences to a body common to both chambers. Such a statute was adopted in 2017 establishing a common working group to both chambers. However, in relation to EU matters this option has not yet been used.97 The Parliament may exercise only an indirect influence over a member of the Government representing the Czech Republic at a Council meeting. The Senate’s and the Chamber of Deputies’ positions must be ‘reflected’ by the Czech authorities, especially by the guarantor (a Ministry or an executive agency responsible within the given area) and by the Committee for the EU, an organ of the Government, when articulating positions in Council meetings, especially during the preparatory stage of the decision-making process. This does not apply to the so-called Parliament’s reservation (see below). Only a decision of the Committee for the EU can directly impact representatives of the Czech Republic with respect to negotiations at the Council (see below). b.  So-called Reservation of the Parliament Parliament’s reservation is the most powerful instrument of influence on the actions of the Government with respect to EU activities.98 It expresses Parliament’s will to discuss a draft of



97 Act 98 Art

No. 300/2017 Sb. 28 of the Government Directive.

80  Luboš Tichý and Tomáš Dumbrovský EU legislative act during its preparation at the EU or at the Czech Republic level. A responsible state administrative body (a Ministry or state agency) is required to submit the reservation to the respective EU institutions, to which such a reservation signals a possible negative position of the Parliament or of the Czech Republic as a whole towards the draft. If the Parliament initiates the discussion on an EU legislative act, this precludes a Government member from participating in the decision-making process on the matter at the EU level.99 Furthermore, a committee of the Chamber of Deputies can summon a member of the Government and demand information regarding the position of the Czech Republic at the Council on the matter at hand.100 In such a case the responsible state administrative body must so inform the Chamber of Deputies (but not the Senate). c.  Prior Consent of the Senate and Chamber of Deputies The prior consent of the Chamber of Deputies and the Senate is required in specific matters listed in the Rules of Procedure of the two chambers.101 A responsible state administration body summarises a framework position and informs the Presidents of the Senate and of the Chamber of Deputies on commencement of the procedure of prior consent. The Government then submits the proposal for prior consent to both chambers without delay, indicating the current state of the law and proposed amendments and its own position. The proposal is distributed to all members of the Parliament and is assigned to the Senate and Chamber of Deputies Committees for European Affairs, which submit their resolution alongside an opponent report (if any) on prior consent to the plenary of the chambers. The Senate and the Chamber of Deputies may only grant the prior consent or refuse to grant it. In the latter case, the President of the respective chamber informs without delay the Government, the President of the other chamber, and the Presidents of the Council, the European Council and the European Commission about the refusal to grant consent to the intended legislation.102 A specific procedure must be followed when discussing a draft act amending third part of the TFEU under the simplified procedure of Article 48(6) TEU.103

2.  Binding Decisions of the Government This matter is regulated by the Statute of the Committee for the EU (SCEU) that forms an annex to the Government Resolution No 664 from 6 August 2014. The Committee for the EU has a key position within the Government. Its conclusions are binding upon the members of the Government for discussions and negotiations in the Council and the European Council. The members of the Committee are members of the Government. The Permanent Representative of the Czech Republic to the Council is an associated member of the Committee.104

99 § 119d of the Procedural Rules of the Senate, Act No 107/1999 Sb. 100 § 109b Procedural Rules of the Chamber of Deputies, Act 90/1995 Sb; Art 29 of the Government Directive. 101 § 109i Procedural Rules of the Chamber of Deputies (n 100) and § 119k and § 119m of the Procedural Rules of the Senate (n 99). This includes decisions of the European Council under Art 31(3) TEU, Art 48(6) and (7) and decisions of the Council under Arts 81(3), 153(2), 192(2), 312(2), 333(1) and (2), and 352 TFEU (if not concerning measures necessary for the functioning of the internal market). 102 § 109j and § 109k Procedural Rules of the Chamber of Deputies (n 100); § 119l and § 119n of the Procedural Rules of the Senate (n 99). 103 § 109l of the Procedural Rules of the Chamber of Deputies (n 100); Art 32 of the Government Directive. 104 Arts 1 and 2 SCEU.

Czech Republic  81

B.  Parliament’s Involvement in Drafting of Secondary Legislation 1.  Relevant Instruments The involvement of the Parliament in drafting secondary EU law is governed by the abovementioned Government Directive on Dealing with EU Documents, issued as an annex to Government Resolution No 665 from 6 August 2014. The Directive stipulates the procedure for the ministries and other public administration bodies and it forms part of secondary legislation preparatory works.105 Articles 17 et seq of the Directive governs the debate on EU matters in the Senate and the Chamber of Deputies.

2.  Parliament’s Involvement A responsible state administrative body (the guarantor) delivers a framework position on every draft of EU legislation. Regarding non-legislative drafts as defined by the EU law, the guarantor delivers a framework position only if the Senate or the Chamber of Deputies decides to discuss the respective instrument.106 The guarantor shall seek an approval of the framework position before the respective documents are discussed in EU institutions or in the Senate or the Chamber of Deputies.107 Hence, the Senate or the Chamber of Deputies may become actively engaged in the matter after the discussion of the respective legislative draft in EU institutions begins. The guarantor is responsible for due fulfilment of the Government’s duties during the discussion of the framework position in the chambers of Parliament in accordance with their Rules of Procedure.108 Furthermore, the guarantor is required to update the chambers of Parliament on the framework position if it is amended during the negotiations and if these amendments may change the position of the Czech Republic.109 The guarantor must inform the Senate on draft EU legislations and relevant EU documents as well as on the current position of the Government, including the impact on national legal order when the Senate or its relevant committee request so.110

C.  Transposition of Secondary Legislation 1.  Legal Framework The transposition process is regulated by two legal instruments that are binding upon administrative bodies participating in the transposition process. They stipulate the procedure and its institutional organisation. The more general one of the two regulations is the Government Rules on Legislative Drafting, adopted in the form of Government Resolution of 9 March 1998 (the last amendment was adopted on 17 January 2018). The second regulation is the Guidance Notes for Work on Fulfilment of the Legislative Obligations arising from the Membership of the Czech Republic in the European Union, issued pursuant to the Government Resolution of 12 October 2005 (the last amendment was adopted on 27 February 2018).



105 See

Art 1 of the Government Directive. 17 and 18(1) of the Government Directive. 107 Art 23 of the Government Directive. 108 See the Acts on Rules of Procedure of the Senate (n 99) and the Chamber of Deputies (n 100). 109 Arts 24 and 25 of the Government Directive. 110 Art 26 of the Government Directive. 106 Arts

82  Luboš Tichý and Tomáš Dumbrovský

2.  Institutional Organisation The transposition process is carried out by the Government, which executes its powers also through other state administrative authorities, and the Parliament, whose chambers adopt the implemented provisions in the form of laws. The President of the Czech Republic is involved in the legislative procedure through his power and duty to sign the transposing bill and disposes of a veto, which may be overruled by absolute majority in the Chamber of Deputies. Finally, the Constitutional Court may review the transposing law and quash it if it conflicts with the Constitution. ‘Technical’ work, that is the actual transposition, is conducted by the Department of Compatibility with EC/EU Legislation of the Office of the Government, by the guarantor, and by the Government of the Czech Republic.

3.  Transposition Procedure In the first phase, the Department of Compatibility with EC/EU Legislation assigns the guarantor with the text of the Directive within three days after the publication of the Directive in the Official Journal of the EU. The guarantor may refuse the assignment within 10 days and suggest another state administrative body to be assigned. Potential disputes are resolved by the Government. Subsequently, the guarantor prepares a transposition analysis. It is an analysis of the content of the Directive addressing fundamental issues, ie how the Directive should be transposed and what measures its transposition requires. The analysis also contains an assessment on whether the transposition may amount to a breach of Czech Republic’s obligations under international law within the meaning of Article 351 TFEU. The guarantor not only analyses the Directive in question, but considers and reflects other relevant sources, including the case law of the CJEU and primary EU law. Guidance Notes stipulate a specific period in which the transposing bill must be submitted to the legislative process.

4.  Standards for EU Law Transposition Standards for transposition of EU law aim to ensure that the transposing bill meets all requirements of EU law with respect to its content, form, and adaptation of the Directive to the specifics of the Czech legal order.111 Transposed Directives require national laws transposing them to contain information indicating that the law transposes a Directive (so-called information reference). It is necessary to differentiate between situations where a national legal instrument completely transposes a Directive and when it transposes only a portion of it. In the former case, the information reference is essentially a general informative clause in the normative text, whereas in the latter it is merely a footnote related to the respective transposed provision.

D.  Enforcement through the Courts and the Application of the EU Charter of Fundamental Rights The Czech Constitutional Court in its first rulings on the application of EU law in the Czech Republic found the Government and the Parliament in breach of their obligations not to 111 For the role of the CFR in the law-making process when EU law is implemented see Bohumil Peterka, ‘Reflexe Listiny základních práv EU při tvorbě vnitrostátní implementační legislativy v České republice’ [‘The CFR in the process of Czech legislation drafting’], Acta Universitatis Carolinae – Iuridica (2018) no 4, 19.

Czech Republic  83 transpose a directly applicable EC Regulation (Sugar Quotas III) and wrongly transposing an EC Directive (Reimbursement of Medications). In the Pfizer case the Constitutional Court made it clear that primary and secondary EU law, as interpreted by the CJEU, was applicable law for ordinary courts. The Constitutional Court itself would strike down decisions of ordinary courts in concrete review cases if their ignorance of EU law in the given case amounts to a breach of the Constitution as follows from Article 1(2) and Article 10a ÚČr. As in non-EU-related cases, the Court would not replace the ordinary courts in their obligation to apply ordinary law. That means that when an ordinary court fails to apply EU law, the Constitutional Court will not intervene where that failure does not amount to a breach of constitutional rights. However, if the failure amounts to a breach of constitutional rights, the Constitutional Court will strike down the decision of the ordinary court and remand the case back to that court for a new ruling. This applies also to failures to make a reference for a preliminary ruling from the CJEU as mentioned above. The effect of the EU Charter of Fundamental Rights on the Constitutional Court’s doctrine and obligations of ordinary courts cannot be fully assessed yet. The Court does not consider the EU Charter to be part of the Czech constitutional order and thus part of its referential framework for constitutionality review, however the Court applies the Charter as an additional criterion.112 The Constitutional Court applied Article 50 of the EU Charter to extend the territorial scope of the principle ne bis in idem;113 the Court found the protection of consumers in Article 38 of the EU Charter as lacking the quality of subjective rights that can be directly enforced;114 and in interpreting Article 39 of the EU Charter, the Court concluded that the provision did not preclude the 5 per cent threshold rule specified by Czech law for elections to the European Parliament.115

E.  Critical Assessment of the Implementation Process At first glance, the participation of the Czech Republic in the EU legislative process, presented above, seems sophisticated, well-organised and balanced enough to create conditions for effective involvement of the Czech Republic in EU law-making. However, experience with the functioning of the system does not support such assessment.

1.  Implementation Regime Experience often reveals a fundamental misunderstanding of the substance and purpose of EU directives by the implementing authorities due to lack of expertise within the given area generally and the law in question in particular. This is the case especially in the area of private law. The reasons are simple enough. Civil servants responsible for the Czech Republic’s input in EU 112 ÚsČr 19.05.2015, Pl. ÚS 14/14 (5 per cent threshold in elections to the European Parliament) ECLI:CZ:US:2015:Pl. US.14.14.1. See Magdaléna Svobodová, ‘Působnost Listiny základních práv EU v kontextu judikatury Ústavního soudu ČR’ [‘The scope of application of the CFR in the context of the case law of the Constitutional Court of the Czech Republic’], Acta Universitatis Carolinae – Iuridica no 4, 53; Magdaléna Svobodová, ‘Listina základních práv EU v judikatuře Ústavního soudu ČR’ [‘The CFR in the case law of the Czech Constitutional Court’], 25 Právní rozhledy (2017) 823; Ondrej Hamuľák, ‘Listina základních práv Evropské unie jako okolí ústavního pořádku České republiky’ [‘CFR in the Environment of the Czech Constitutional Order’], Acta Iuridica Olomucensia (2015) 7; J Grinc, ‘Listina základních práv EU jako měřítko pro vnitrostátní legislativu?’ [‘The Charter of Fundamental Rights of the EU as a Yardstick for National Legislation?’], Acta Universitatis Carolinae – Iuridica (2018) no 4, 11. 113 ÚsČr 14.04.2016, II. ÚS 143/16 (principle ne bis in idem in criminal proceeding in the Schengen area) ECLI:CZ:US:2016:2.US.143.16.1. 114 ÚsČr 10.04.2014, III. ÚS 3725/13 (bank fees) ECLI:CZ:US:2014:3.US.3725.13.1. 115 Pl. ÚS 14/14 (n 112).

84  Luboš Tichý and Tomáš Dumbrovský law-making and for the subsequent implementation are insufficiently trained. It is often the case that they do not understand well the substance of the legislation, especially its purpose. As a consequence, in connection with other circumstances mentioned below, the civil servants often neglect the purpose of new legislation and, notably, its contextual relationship with other instruments within the EU as well as national legal orders. If the civil servants, who are primarily and directly responsible for the Czech Republic’s input in EU law-making and its subsequent implementation, lack the needed expertise, it is illusory to expect such expertise from others involved in the legislative process. The lack of expertise covers almost all political levels, both governmental and parliamentary, with few exceptions. Unfortunately, even within the Government’s Legislative Council, composed of legal experts, there is often lack of members specialising in the area discussed.

2.  Deficiencies of Expert Discussion The reality of under-trained civil servants and the practice in other Member States calls for the involvement of external experts, especially academics, into both the drafting of the law at EU level and the implementation process. This is particularly urgent within the area of private law, in which the EU exercises its competence mainly in the area of consumer protection. In this particular area, responsibility in the Czech Republic is entrusted to the Ministry of Industry and Trade, which creates a paradoxical situation where civil servants responsible for advancing conditions for industrial development and trade shall ensure a protection of consumers. As a consequence, little attention is paid to consumer protection. In general, civil servants responsible for each area of legislation have almost no contact with the outside world. Not only do they rarely cooperate with experts in the given area, but they also lack any initiative towards developing contacts with their counterparts from other Member States. They are nonetheless involved in the legislative process at the EU level through informal negotiations with civil servants of the European Commission. The result is that nobody at the national level in fact takes care of the process in which the EU law is created and the attention that is occasionally given to an EU legislative initiative is mostly accidental, depending on a personal initiative of a civil servant.

3.  Cooperation with Other Member States and European Commission Officials The passivity of civil servants regarding EU law-making means also that the Czech Republic does not take any initiative towards legislative changes at the EU level. There is neither a discussion on possible changes or new regulations that might be pushed by the Czech Republic, nor a discussion on legislative initiatives already introduced by the European Commission.

4.  Case Study on the Implementation of the Directive on Consumer Disputes Experience with the drafting and implementing of the Directive on Alternative Dispute Resolution (ADR) for consumer disputes supplies evidence confirming the reality of the Czech Republic’s participation in EU law-making and implementation of EU law described above.116 116 Directive 2013/11/EU of the European Parliament and of the Council of 21 May 2013 on alternative dispute resolution for consumer disputes and amending Regulation (EC) No 2006/2004 and Directive 2009/22/EC (Directive on consumer ADR), [2013] OJ L 165/63.

Czech Republic  85 The Czech Republic did not participate in any way in the legislative process, which had been relatively lengthy and contentious. There had been no discussion or critique on the Commission recommendations from 1998, 2001, and 2013. Deficiencies in expertise in the implementation process described above in general terms come to light when we look into the draft of the proposed implementing legislation. The core of the Directive lies in creating dispute resolution organs that will be independent and impartial. Despite this clear obligation, the Czech draft legislation chooses as ADR entities the Czech Trade Inspection Authority, the Czech Telecommunication Office, and the Energy Regulatory Office, authorities whose primary task is to control and regulate the providers. Even if these authorities create a special, separate department that functions as the mediator, it is difficult to imagine that providers and consumers would entrust mediation and the necessary information to an entity that has the power to inspect and sanction them. The transposition process should have raised these questions and led to the creation of an independent institution that would fulfil the Directive’s material objective.

VI.  Conclusion: Ongoing Dynamic Relationship between EU Law and the Czech Constitutional Order Experience of totalitarian regimes and foreign occupations between 1938 and 1989 has defined three main tenets of the Czech constitutional system – a high level of protection of fundamental rights and freedoms, with an emphasis on individual rights, safeguards against reversion of the democratic system based on rule of law, and the special value of sovereignty. Despite the last tenet, the constitutional system has been very open to the effects of international law, especially to human rights treaties. The Constitutional Court formulated the fundamental principles of its constitutional adjudication prior to the accession of the Czech Republic to the EU: primacy of teleological interpretation, material core of the Constitution and ultra vires doctrine, radiation doctrine, principle of constitutionally conforming interpretation (indirect effect), and principle of high level of protection of fundamental rights and freedoms. The Court found a delicate balance between judicial restraint and activism during the transition to democracy. Its rather restrained approach towards abstract review, where it would clash with political branches, was compensated by its activist approach to protection of fundamental rights and freedoms within the concrete review, where the Court aimed to ‘teach’ the ordinary judiciary to apply the Constitution directly and on regular basis using the interpretative methods mentioned above that were rather foreign to ordinary judges. Following the Czech Republic’s accession to the EU, the Court adapted these principles to the new situation. The development of its European doctrine has been influenced by the German Federal Constitutional Court and the Court’s Kelsenian understanding of the constitutional judiciary as occupying a special position in the constitutional system with its main task being the protection of constitutionality. The Court developed two general review standards. It reviews existing EU law and acts of EU institutions in light of the eternity clause, the content of which is subject to evolution (first review standard).117 In the case of ex ante review of treaties under Article 10a ÚČr, the Court reviews their compliance with the entire constitutional order (second review standard). Yet, the Court signalled that it would block a ratification of an EU Treaty Amendment only when the constitutional core is threatened.118

117 Pl. 118 Pl.

ÚS 50/04 (n 37). ÚS 19/08 (n 50).

86  Luboš Tichý and Tomáš Dumbrovský This, at first sight, progressive and self-confident approach to European integration and EU law becomes grimmer when we look into everyday practice. Authorities of the state administration often lack sufficient expertise, approach their tasks formalistically, and are unwilling to engage in discussion with and learn from legal academia at home or their colleagues in other Member States, and are utterly unaware of the European legal academic discourse even in their areas of expertise. Czech legal academia, with notable exceptions, suffers from a lack of comparative approach and struggles to grasp changing realities due to their insistence on rigid legal categories and concepts of the past with little functional value. The courts follow this trend. All these issues add to a worrying decline in public support to further integration. The constitutional authorities, from political to judicial branches, simultaneously affect and adapt to the public attitude to integration. As a result, the Czech Republic is moving to the periphery of the European integration project, as indicated by its hesitant attitude to ratifying the Lisbon Treaty and abstaining from the eurozone and most of the euro-crisis measures. In the future, we may expect further refinement of the Constitutional Court’s European doctrine. First, within the second review standard, the Court has not clarified how to assess which parts of the treaty under ratification are really new and which are already enshrined in existing EU law (in particular when a treaty incorporates CJEU case law) and how to approach this problem. Second, the Constitutional Court will have to further refine its second review standard once it considers treaties under Article 10a ÚČr which are negotiated outside the EU framework and once it is asked to review ordinary treaties that require ratification. The Court would need to consider that while in the case of integrative treaties (Article 10a ÚČr), the Parliament gives its consent by a three-fifths majority of both chambers, the same majority needed for a constitutional revision, in the case of ordinary treaties, simple majority suffices. Third, the Constitutional Court will need to develop a broader scale of measures that could be employed in case of conflict with the Constitution or its core (depending on the review standard). In particular, the Court will have to find procedural ways to require the Parliament to adapt the Constitution to avoid the conflict, request special guarantees for the interaction between the Government and the Parliament, or set interpretative conditions for the problematic passages of the evaluated provisions (either of a treaty or secondary EU law).

References P Bříza, ‘The Constitutional Court on the Lisbon Treaty Decision of 26 November 2008’ (2009) 5 European Constitutional Law Review 143. A Gerloch, ‘Povaha Smlouvy o stabilitě, koordinaci a správě v Hospodářské a měnové unii sjednávané mezi zeměmi Eurozóny z hlediska požadavků stanovených ústavním pořádkem České republiky na proces její ratifikace’ [‘The Character of the TSCG in the EMU Negotiated Among the Eurozone Countries from the Perspective of Conditions Required by the Constitutional Order of the Czech Republic for the Ratification Process’] (2012) 58 Acta Universitatis Carolinae – Iuridica 99. J Grinc, ‘Listina základních práv EU jako měřítko pro vnitrostátní legislativu?’ [‘The Charter of Fundamental Rights of the EU as a Yardstick for National Legislation?’] (2018) 4 Acta Universitatis Carolinae – Iuridica 11. O Hamuľák, ‘Listina základních práv Evropské unie jako okolí ústavního pořádku České republiky’ [‘The CFR in the Environment of the Czech Constitutional Order’] (2015) Acta Iuridica Olomucensia 7. P Kantořík, ‘Ukončí zákon č. 428/2011 Sb. letitý spor o československé důchody?’ [‘Would the Law No. 428/2011 Coll. Resolve the Long-term Conflict on Czechoslovak Pensions?’] (2012) 20 Právní rozhledy 254. J Komárek, ‘Slovenské důchody: druhá „válka soudů“ před Soudním dvorem EU’ [‘Slovak Pensions: Second “War of Courts” Before the CJEU’] (2011) 17 Soudní rozhledy 388.

Czech Republic  87 R Král, ‘Uznání i rozpaky nad „komunitárním“ nálezem Ústavního soudu ve věci cukerných kvót’ [‘Acknowledgment and Perplexity Regarding the “Community” Decision of the Constitutional Court in Sugar Quotas’] (2006) 14 Právní rozhledy 411. F Křepelka, ‘Českoslovenští důchodci v pasti práva Evropské unie’ [‘Czechoslovak Pensioners in the Trap of EU Law’] (2011) 19 Časopis pro právní vědu a praxi 131. Z Kühn, ‘Ještě jednou k ústavnímu základu působení komunitárního práva v českém právním řádu’ [‘Once Again on the Constitutional Basis for Application of the Community Law in the Czech Legal Order’] (2004) 12 Právní rozhledy 395. Z Kühn and J Kysela, ‘Na základě čeho bude působit komunitární právo v českém právním řádu?’ [‘What will be the Basis for the Community Law Application in the Czech Legal Order?’] (2004) 12 Právní rozhledy 23. J Malenovský, ‘Ve věci ústavního základu působení komunitárního práva uvnitř ČR nebylo řečeno poslední slovo’ [‘The Final Word Has Not Been Said Yet on the Issue of Basis for the Community Law Application in the CR’] (2004) 12 Právní rozhledy 227. J Malenovský, ‘Vítězství “dogmatiků” nad “pragmatiky” se odkládá’ [‘The Victory of “Dogmatists” Over “Pragmatists” Delayed’] (2005) 13 Právní rozhledy 408. J Malenovský, ‘The Relations between Constitutional Law: Community Law and International Law according to the Czech Constitution: Still Ambiguities to be Resolved’, in M Hofmann (ed), Europarecht und die Gerichte der Transformationsstaaten (Baden-Baden, Nomos, 2008) 31. B Peterka, ‘Reflexe Listiny základních práv EU při tvorbě vnitrostátní implementační legislativy v České republice’ [‘The CFR in the process of Czech legislation drafting’] (2018) 4 Acta Universitatis Carolinae – Iuridica 19. R Pomahač, ‘Soudní dvůr EU: Diskriminační podmínka pobytu v důchodovém pojištění’ [‘CJEU: Discriminatory Condition of Residence in Pension Insurance’] (2012) 20 Právní rozhledy 640. M Svobodová, ‘Působnost Listiny základních práv EU v kontextu judikatury Ústavního soudu ČR’ [‘The scope of application of the CFR in the context of the case-law of the Constitutional Court of the Czech Republic’] (2018) 4 Acta Universitatis Carolinae – Iuridica 53. M Svobodová, ‘Listina základních práv EU v judikatuře Ústavního soudu ČR’ [‘The CFR in the case law of the Czech Constitutional Court’] (2017) 25 Právní rozhledy 823. L Tichý and T Dumbrovský, ‘Between Two Legal Orders: A Relativist Doctrine for a Member State Constitutional Court?’, in A Koch et al (eds), Europe. The New Legal Realism. Essays in Honour of Hjalte Rasmussen (Copenhagen, Djøf Forlag, 2010) 757. R Zbíral, ‘Czech Constitutional Court, judgment of 31 January 2012, Pl. ÚS 5/12 – A Legal revolution or negligible episode? Court of Justice decision proclaimed ultra vires’ (2012) 49 CML Rev 1475. J Zemánek, ‘Otevření ústavního pořádku komunitárnímu právu potvrzeno, nikoli však nekontrolovatelné’ [‘Opening of the Constitutional Order to the Community Law Confirmed, Albeit Not Unconditional’] (2006) 15(5) Jurisprudence 47. J Zemánek, ‘The Constitutional Courts in the New Member States and the Uniform Application of European Law’ in I Pernice et al (eds), The Future Of The European Judicial System in a Comparative Perspective (Baden-Baden, Nomos, 2006) 257. J Zemánek, ‘Přezkum ústavnosti Lisabonské smlouvy: obsahové otázky’ [‘Constitutionality review of the Lisbon Treaty: Questions of the Content’] (2009) 18(1) Jurisprudence 32.

88

4 Denmark The Hesitant European? The Constitutional Foundation of Denmark’s EU Membership and its Material Reality MARLENE WIND*

I. Introduction The establishment over the past 60 years of a supranational legal order in Europe with powerful European courts like the European Court of Human Rights (ECtHR) in Strasbourg and in particular the Court of Justice of the European Union (CJEU) in Luxembourg constitutes a rather dramatic challenge to Danish legal and political tradition. In Denmark, the Parliament has traditionally had the upper hand compared to other branches of government and is regarded as the ultimate symbol of democracy.1 Similarly to several other Nordic countries, Denmark has no tradition of judicial review2 and few national politicians, judges or civil servants over the years have challenged this state of affairs, despite the process of European integration.3 Denmark has been regarded as a bastion of Scandinavian legal positivism4 where statutes and political decision-making together with judicial self-restraint on the part of the national judiciary are

* I owe a lot of people thanks for helping me with this chapter. The greatest thanks go to cand jur Gustav Schaldemose, at the Danish State attorney, Kammeradvokaten, who made sure that the formal part on Danish constitutional law was in place and correct. Secondly, cand jur and former advisor to the Danish Government in EU matters in the MFA, Per Lachmann, who helped negotiate Danish EU membership in 1972 and take Greenland out of the EU. He read through the manuscript and corrected quite a few of my misunderstandings. As the chapter builds on my previous research, I would also like to thank scholars from iCourts, Centre of Excellence for international Courts in Copenhagen for constructive comments and criticism. Remaining misunderstandings or mistakes are of course entirely my own responsibility. 1 Jens Elo Rytter and Marlene Wind, ‘In Need of Juristocracy? The Silence of Denmark in the Development of European Legal Norms’ (2011) 9 International Journal of Constitutional Law 470; see also Henrik Zahle, ‘Grundlovens menneskerettigheder: Sammenstødet mellem legalistiske og dynamiske retstraditioner’ [‘Human rights of the Constitution: The Clash Between Legalistic and Dynamic Legal Traditions’], in M Kjærum et al (eds), Grundloven og menneskerettigheder – i et dansk og europæisk perspektiv (Copenhagen, Djøf, 1997) 361, 376 and n 39. 2 Speaking about lacking tradition for judicial review in this context constitutional review is primarily meant. 3 It could be argued that due to the decreasing standing of the political class and maybe also the influence of globalisation on the EU, no political party today is likely to argue against the principle of constitutional review, even though it is still not part of Danish political culture. In 1953, the issue of constitutional review was too disputed to be included in the new Constitution. 4 See further Marlene Wind, ‘Do Scandinavians Care About International Law? A Study of Scandinavian Judges’ Citation Practice to International Law and Courts’ (2016) 85 Nordic Journal of International Law 281.

90  Marlene Wind absolutely central. Thus, European constitutional development has both been rather profound but also much more daunting than is ordinarily recognised. Denmark lacks the constitutional court and constitutional council to deal with EU law and the concurring potential clash between European and national law, that other European countries have. Rather, the Ministry of Justice has played a strong gate-keeping role at all levels dealing with EU law and regulation, including implementation of EU law and advice to the Government, the ministries and the Danish Parliament’s European committee. In addition to this, the Ministry is also engaged in infringement cases against the state and with preliminary references, where the Judicial Committee,5 made up of civil servants from different ministries but headed by a high-ranking civil servant from the Ministry of Justice, also plays a central role. The Judicial Committee together with the Ministry of Justice has enormous influence in deciding (1) whether or not European case law is to be considered relevant in Danish law and at the general EU law implementation stage, (2) when new legislation is to be transposed into Danish law. The latter almost always happens through ministerial circulars (executive orders). Is one for instance to opt for a narrow or a broader interpretation of new case law or directives, and how does a new EU regulation correspond to its (potentially already existing) national counterpart? These are essential questions very often left to a small but influential group of civil servants. It goes without saying that the following description and analysis is condensed and will be unable to describe all the peculiarities of the Danish legal system in this respect. It is deliberately also much more critical than most formal textbooks on the Danish relationship with EU law, which are often mainly descriptive and undertheorised.

II.  Main Characteristics of the National Constitutional System The Kingdom of Denmark comprises Denmark, Greenland and the Faroe Islands. The Danish Constitution6 applies to all these parts,7 although Greenland and the Faeroe Islands both have a far-reaching system of self-government.8 Danish membership of the EU does not apply to the Faroe Islands, and Greenland is covered by the treaties only in respect to the association of overseas territories. The Constitution dates back to 1849. It was the result of peaceful political compromise between the King and the People that transformed Denmark from an absolute monarchy to a constitutional monarchy with a representative parliamentary system, separation of powers and a bill of rights. In reality, however, the King preserved an influential role until 19019 when a broad political compromise accepted that the Government had to reflect the political will of the lower house only. The King also decided on the election dates and retained a voice in foreign policy matters. Women and servants did not get a right to vote until 1915. Moreover, due to the peaceful transition from absolutism to democracy from the mid-nineteenth century onwards, the courts 5 In Danish Juridisk specialudvalg. It consists of civil servants from several ministries and has a high-ranking EU lawyer from the Ministry of Justice as its chair. 6 In Danish, Grundloven (hereinafter DKGL). 7 Sec 1 DKGL. 8 After the introduction of Home Rule in Greenland, a consultative referendum was held in 1982 on EC membership, after which Greenland chose to leave the EC on 1 February 1985. Art 204 TFEU specifically addresses Greenland. Art 355(5)(a) explicitly provides that the EU Treaties shall not apply to the Faeroe Islands, which is thus a third country from the point of view of EU law. EU-Faroe Islands trade is governed by a 1997 agreement ([1997] OJ L 53/2–135). 9 Tim Knudsen, Fra Enevælde til Folkestyre. Dansk demokratihistorie indtil 1973 [From Absolutism to People’s Government. Danish Democracy History until 1973] (Copenhagen, Akademisk Forlag, 2006).

Denmark  91 and large parts of the ministerial apparatus were taken over by the new rule. Where the civil servants and judges previously served the King, they now served the Government, and later the parliamentary majority. This smooth transferal also in terms of the judges’ loyalty, from absolutist rule to the Parliament, has been referred to as one of the main reasons for the absence of a tradition of judicial review, which was not mentioned in any of the following Danish constitutional amendments.10 Even today, the Danish Constitution is one of the most difficult constitutions to amend in Europe.11 Since 1849, it has only been revised four times, most recently in 1953 (plus an amendment of the Act of Succession to the throne in 2009, which put an end to male-preference primogeniture).12 The Constitution has also preserved some very old language and still talks about ‘The King’ as the law-giver instead of the Government. In a European context, it is important to note that in the 1953 revision, a provision was added to the Constitution to make it easier for Denmark to join supranational organisations. Twenty years later that provision was applied when Denmark acceded to the European Communities (EC).13 At the same time, however, due to the difficulty of amendment, many important features of the Danish constitutional setting today have no or little support in the written Constitution.14 For example, there is no mention of the EU in the Danish Constitution, as opposed to many other national constitutions in Europe, including that of close neighbour Sweden.15 This does not mean, however, that Danish EU membership is unconstitutional. Section 20 was – as indicated above – inserted in the Constitution in 1953 and ensures that national competences under specific circumstances, if specifically defined, can be transferred to international organisations. As a result of this inertia and unwillingness to discuss constitutional issues more broadly,16 practice and interpretation have over time become important in shaping the Danish constitutional setting.17 Thus, the old, very short and almost unamendable Danish Constitution leaves considerable room for manoeuvre to political and judicial actors. As we will return to later, this room was mainly filled by political actors and civil servants due to Denmark’s tradition for a strong Parliament and reticent courts.

A.  The Political System Denmark is a representative parliamentary democracy. The unicameral national Parliament comprises 179 members.18 Members are elected by proportionate representation for a four-year 10 Constitutional review was not inserted in the most recent amendment to the Danish Constitution in 1953, due to opposition by the main parties in the Folketing. 11 An obvious reason for the rarity of formal constitutional revisions – and the existence of outdated provisions – is the very rigid amendment procedure of Sec 88 DKGL. This procedure requires the adaptation in Parliament of the amendments in question, the calling of a general election, re-adaption of the same proposal in the newly elected Parliament, and approval by referendum in which the majority and at least 40 per cent of those entitled to vote must vote in favour of the proposed amendments to the constitution. 12 According to s 2 DKGL, Denmark is a constitutional monarchy. However, much has changed since 1849. Formally, the monarch (who is presently a queen) is the Head of State with a number of powers enumerated in the Constitution (ss 12–14). In today’s constitutional reality, however, the monarch has lost most of his/her power and is merely a rather popular head of state (with 83 per cent approval rate according to a poll from 2013). 13 We will analyse this provision in detail below, as it is essential for Denmark’s constitutional relationship to the EU. 14 For further analysis on the point, see Helle Krunke, ‘Constitutional Identity – Seen through a Danish Lens’, (2014) 37 Retfærd No 4/147, 24, 37 f. 15 Cf, eg, s 10, Ch 1 of Sweden’s Instrument of Government, and the contribution of J Nergelius in this volume. 16 In 2015, the Government decided to drop the proposed establishment of a commission to draft a new Constitution. 17 Krunke, ‘Identity’ (n 14) 27. 18 S 28 DKGL.

92  Marlene Wind period, but it is the prerogative of the Prime Minister to call a new election at any time during his/her electoral term, thus initiating a new four-year period.19 Similarly, the Government sits for a four-year period, unless it loses a vote of confidence in Parliament or if individual ministers choose to step down.20 Whereas, for instance, Germany adheres to a positive parliamentary principle, Danish parliamentarianism applies the negative form, ie a vote of no confidence is necessary to deselect the Government, headed by the Prime Minister, after an election.21 Denmark has mostly been run by minority governments. Therefore, it is highly dependent on supporting parties in Parliament. This has created a strong Parliament compared to many countries in the EU, and a tradition of firm parliamentary control of the Government. For instance, the European Committee in the Danish Parliament has traditionally been considered one of the most influential national European Policy Committees in the EU because the relevant Minister always had to ensure a mandate before negotiating in the Council of Ministers in Brussels.22 New research has, however, questioned this dogma as the rather small and understaffed European Committee can no longer overlook and relate to all legislation coming from the EU.23 Moreover, a study by the Danish Ministry of Foreign Affairs from January 2014 on the past five years of committee work demonstrated that mandates given from the Committee to different Ministers had on 14 occasions been given too late. This means that the mandate from the Committee to the Minister was given after the Government had already negotiated and decided on new legislation in the Council of Ministers. The analysis also showed that every fourth case was presented to the European Committee only a week before the final negotiation in the Council of Ministers, almost making it impossible for the Committee to influence the decision and exercise democratic control.24

B.  The Judicial System The judiciary is formally completely independent of the executive and legislative powers.25 Unlike many other European countries, Denmark does not have separate constitutional, administrative or criminal courts. The ordinary courts take on all these different roles and in this way the

19 Ss 31–32 DKGL. 20 S 15 DKGL. 21 The negative parliamentary principle in the Danish multi-party system has paved the way for the important feature that Denmark has had a quite consistent tradition of minority governments – however, sometimes based on a political coalition, which means that the Government together with one or more supporting parties has the majority in Parliament. As a result, the Parliament’s position has been strengthened vis-à-vis the Government. This has, accordingly, fertilised a highly consensus-oriented political culture and common practice of co-operation across the left/right-axis in Parliament. See further, Peter Munk Christiansen and Lisa Togeby, ‘Power and Democracy in Denmark: Still a Viable Democracy’ (2006) 29 Scandinavian Political Studies 1. The Constitution also provides Parliament with the possibility of legal action, as ministers can be impeached at the Court of Impeachment. There have only been five cases so far, however. Most of them date back to the nineteenth century; the last one, the so-called Tamil case, was as recent as 1995. 22 See below and Krunke, ‘Identity’ (n 14) 29. 23 See Hjalte Rasmussen, Folkestyre, Grundlov og Højesteret: Grundlovens § 20 på prøve [Government, Constitution and Supreme Court: Section 20 of the Constitution on Test] (Copenhagen, Christian Ejlers, 1996). See also Kasper Kaasgaard, ‘Forsker: Overlad den demokratiske kontrol til Parlamentet’ [‘Researcher: Leave Democratic Control to Parliament’], Altinget.dk 18.11.2014 (www.altinget.dk/eu/artikel/forsker-overlad-den-demokratiske-kontrol-til-parlamentet). 24 See further Michael Hjøllund and Rikke Albrechtsen, ‘Undersøgelse afslører huller i Folketingets EU-kontrol’ [‘Survey Reveals Gaps in Parliament’s EU Control’], Altinget.dk 22.01.2015 (www.altinget.dk/eu/artikel/undersoegelse-afsloererhuller-i-folketingets-eu-kontrol); see also Kasper Kaasgaard, ‘Tænketank: Europaudvalgets mandat er et gummistempel’ [‘Think Tank: The Mandate of the European Committee is a Rubber Stamp’], Altinget.dk 11.11.2014 (www.altinget.dk/eu/ artikel/taenketank-europaudvalgets-mandat-er-et-gummistempel). 25 See in particular ss 61 and 64 DKGL as well as s 3, embodying the principle of divisions of powers.

Denmark  93 Danish judiciary system is a unified one with the Supreme Court at the top.26 This competence is, however, not mentioned in the Constitution and constitutional scrutiny is not part of Danish constitutional culture. To the contrary, the Supreme Court has only once in 170 years (from 1849 to 2019) set aside a piece of legislation which was not in accordance with the Constitution.27 The right – in principle at least – to exercise judicial review was established in case law in 1921. As Henning Koch has pointed out, the Danish Supreme Court curiously emphasised its own full support for the supremacy of the Parliament in the same ruling when it – for the first time (in 1921) – formally stated its competence to review the constitutionality of legislative acts. As Koch noted: ‘The Supreme Court equally resolved that in principle there is no one over or next to the Rigsdag.’28 Concurrently, Danish courts have a strong tradition of self-restraint, leaving huge discretion to the Parliamentary majority.29 Jens Elo Rytter emphasises the primacy of the elected Parliament in Danish constitutional culture and other Nordic states below: Common to the constitutional tradition of the Nordic countries … there is an emphasis on the preferred position of Parliament in the constitutional power structure based on its democratic mandate through elections. The courts have no similar democratic mandate and, therefore, the judicial review of legislation is either problematic in principle or should at least be kept within rather narrow limits.30

The courts also cannot conduct abstract review of EU or international law. However, in practice abstract review takes place in the Ministry of Justice. This also means that it often takes three to four years and sometimes longer from enacting a law to a verdict of the Danish Supreme Court. The absence of a national constitutional court turns the Ministry of Justice into a true ‘power-house’ and the main mechanism ensuring the constitutionality of legislation, including international obligations and EU legislation. Thus, the Ministry de facto carries out an (abstract) ex ante review scrutinising the coherence between European law (case law as well as directives) before practical implementation takes place. To be more concrete, practical implementation mostly takes place as executive orders handed down by the different ministries with the Justice Ministry as the main responsible actor overseeing the process.31 Below, we will get back to exactly how central the Ministry of Justice is in most Danish EU-related matters.32 Danish courts also exercise self-restraint and hesitation when it comes to citing international law or human rights law in national cases.33 Self-restraint also accounts for the rather low number 26 S 63 DKGL empowers the courts to review the compliance of the administration with the law. Courts cannot conduct abstract review, and thus the plaintiff must have sufficient legal standing to have a case tried before the courts. There are exceptions, however: the Supreme Court has – on an exceptional basis – widened the scope of admittance in two cases – the Maastricht and the Lisbon case – concerning the constitutionality of Denmark’s ratifications of EU amendment treaties, see below. 27 This was in the so-called Tvind case, see Danish Supreme Court (hereinafter DKSC), 19.02.1999, Case I 295/1998, Ugeskrift for Retsvæsen [Law Journal, hereinafter UfR] 1999, 841, U.1999.841.H. We return to the question of Danish judicial restraint below. 28 Henning Koch, ‘Dansk Forfatningsret i Transnational Belysning’ [‘Danish Constitutional Law in Transnational Light’] (1999) 81 Juristen 213, 226, my translation. 29 See Rytter/Wind, ‘Need of Juristocracy’ (n 1). 30 Jens Elo Rytter, Grundrettigheder, domstolens fortolkning og kontrol med lovgivningsmagten [Fundamental Rights, the Court’s Interpretation and the Control of Legislative Power] (Copenhagen, Thomson-Gad Jura, 2000) 138. 31 Note how much EU legislation is transferred to Danish law through executive orders. On the Ministry’s expert lawyers’ important role in testing EU treaty amendments against the Constitution, see below. 32 For this view, see among others, Jens Peter Christensen, Jørgen Albæk Jensen and Michael Hansen Jensen, Dansk Statsret [Danish Constitutional Law] (Copenhagen, Djøf, 2012) 234; Ross, however, takes the position that courts’ competence to judicial review can be repealed by an ordinary legislative act, see Alf Ross, Dansk Statsforfatningsret [Danish Constitutional Law], ed O Espersen (Copenhagen, Nyt Nordisk Forlag, 1983) 184. 33 See Wind, ‘Do Scandinavians’ (n 4); and Marlene Wind, ‘The Scandinavians: The Foot-dragging Supporters of European Law?’, in M Derlén and J Lindholm (eds), The Court of Justice of the European Union: Multidisciplinary Perspectives (Oxford, Hart, 2018) 191.

94  Marlene Wind of preliminary references to the CJEU in the first decades of membership.34 Both issues may be attributed to the principle of dualism which is influential in Denmark. It governs the relationship between the national and the international judicial realms. Dualism, which still plays a role, has generally implied that international law is not automatically a part of Danish law, but only becomes so when the Danish legislator chooses to implement international law into domestic law through legislation.35 This means that international obligations do not have a constitutional but only statuary status governed by ordinary legislative acts and principles. As Christoffersen and Madsen argue, until well ‘into the late 1990s, human rights law was not a common discipline for Danish lawyers’.36 Thus, the dualist principle that ‘in the case of conflict – it is assumed that international law yields to national law, even if Denmark thereby violates international law’ has been a common understanding for the past decades of Danish lawyers.37 Denmark is, however, not the only dualist legal system, and for others it has clearly not been interpreted as rigidly as it sometimes appears in the Danish context. The principle of dualism is also, to some extent at least, modified by a number of more recent unwritten principles of Danish constitutional law. Firstly, the so-called ‘rule of interpretation’ states that Danish law, to the fullest extent possible, is to be interpreted in accordance with Denmark’s obligations under international law. Secondly, according to the so-called ‘rule of assumption’ it is generally to be assumed that Danish legislation is adopted in respect of Denmark’s obligations under international law, unless the Danish legislator has explicitly stated otherwise. In addition, the legal theory has identified a third rule, coined as the ‘rule of instruction’. That rule provides that when administrative authorities apply the law, they should do so in accordance with Denmark’s international obligations.38 However, despite these modifying elements, they are often considered difficult to accommodate when conflicts with national legislation arise exactly because statutes are seen as a ‘primary source of law’.39

III.  Constitutional Practice and Culture: The Parliament as the Ultimate Symbol of Democracy40 The strong emphasis on the Danish Parliament as elevated above other branches of government and the most significant symbol of democracy has also influenced the behaviour, attitudes and practice of Danish courts and judges.41 The doctrine of judicial self-restraint is central here and

34 See more about this below. See also Marlene Wind, ‘The Nordics, the EU and the Reluctance Towards Supranational Judicial Review’ (2010) 48 Journal of Common Market Studies 1039. 35 Christensen et al, Statsret (n 32) 195. However, some legal scholars have argued that the dualist approach (as modified by the rules of interpretation and assumption) ought to be abandoned in light of the development of international law, see, eg, Jonas Christoffersen, ‘Folkerettens virkning i dansk ret’ [‘The Effect of International Law in Danish Law’] (2001) UfR 143. 36 Jonas Christoffersen and Mikael Rask Madsen, ‘The End of Virtue? Denmark and the Internationalisation of Human Rights’ (2011) 80 Nordic Journal of International Law 257. 37 Ibid, 264. 38 Henrik Zahle, Dansk Forfatningsret – Studieudgave [Danish Constitutional Law – Textbook] (Copenhagen, Christian Ejlers, 2006) 364–67. 39 Cf Alexander M Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics (Indianapolis IN, Bobbs-Merrill Educational Publishing, 1962) 3. 40 The following section draws to some extent on section 2.1 in Rytter/Wind, ‘Need of Juristocracy’ (n 1). 41 See Henrik Palmer Olsen, Magtfordeling: en analyse af magtfordelingslæren med særligt henblik på den lovgivende magt [Distribution of Power: An Analysis of the Doctrine of Power Distribution with a Special Focus on the Legislative Power] (Copenhagen, Djøf, 2005). See also Marlene Wind, ‘When Parliament Comes First – The Danish Concept of Democracy Meets the European Union’ (2009) 27 Nordic Journal of Human Rights 272.

Denmark  95 emphasises a hesitancy to review the constitutionality of acts of Parliament.42 What self-restraint means is then that whenever judicial review is initiated, often on the basis of some imprecise constitutional provisions, the courts will give significant scope to the assessment of the legislator, primarily because of the direct democratic mandate of the elected politicians. Thus, to be more specific, a court should avoid insisting on having the final say as to the specific contents of broad constitutional norms.43 A self-restrained court will in other words most often accept the assessment of the legislator, even if its own constitutional interpretation might have ended in a different outcome. In Denmark the tradition of judicial self-restraint and of avoiding judicial review has been absolutely central. This was not only in political and legal life, but also as a mantra in education in law and political science, which have produced generations of national civil servant elites.44 As mentioned above, the Danish Constitution from 1849 did not allow or even refer to judicial review of legislation as a possibility. The tradition of retracted courts can, as indicated earlier, be traced back to the case from 1921 when the Danish Supreme Court first ruled on its own power to exercise judicial review in Danish constitutional law. Upholding Parliament’s land-ownership reform,45 the Supreme Court argued that a citizen’s claim that the act had not provided full compensation for his loss of property could not be affirmed ‘with the certainty which is required for the courts to set aside an Act of Parliament as unconstitutional’.46 It is indeed interesting that the Danish Supreme Court itself applied and established such a clear judicial self-restraint doctrine, thereby legitimising ‘majoritarian democracy’ and the political branch’s supremacy over the other branches of government.47 With the exception of Norway, where a solid judicial review tradition has been established, the vision of democracy as mainly contained in the Parliament is characteristic of the Nordic self-image.48 As we are to see below, the hesitance to overrule Parliament is closely linked to the legal and political philosophical tradition dominant in Denmark (and the other Nordic states) since the beginning of the twentieth century. Danish Supreme Court judge, Mogens Munch, emphasised this back in 1989, saying that As is known, the Danish Supreme Court has not, to this date, set aside an act of Parliament as unconstitutional, although it seems, at times, there was occasion to do so. The willingness of the courts to set aside acts of Parliament may be compared to an emergency brake: only if the machine runs wild, leaving the population and its general sense of justice behind, may one expect the use of the emergency brake.49 42 The power of judicial review was established by the famous opinion delivered by Chief Justice John Marshall in US Supreme Court, 23.02.1803, Marbury v Madison, 5 U.S. (I Cranch) 137 (1803). See also James B Thayer, ‘The Origin and Scope of the American Doctrine of Constitutional Law’ (1893) 7 Harvard Law Review 129. See also Ronald Dworkin, Taking Rights Seriously (Cambridge MA, Harvard University Press, 1977) 137. 43 Cf also Bickel, Least Dangerous (n 39) 3. 44 See below where we discuss the centrality of Scandinavian legal positivism but also Magtudredningen, an analysis of social power structures, commissioned by the Danish Parliament and conducted in 2003 by political scientists. The fact that political science students are still taught about politics and society through a text by the Norwegian political scientist Johan P Olsen with no mention of courts is rather telling. See Johan P Olsen, Politisk organisering [Political Organisation] (Oslo, Universitetsforlaget, 1978). 45 Studies have shown that this result was reached by the smallest possible margin: five of the 11 judges participating in the case voted in favour of striking down the act as incompatible with the constitutional protection of private property (one more judge had even been inclined to do so until the very last minute). Poul Andersen, Rigsdagen og Domstolene [Parliament and the Judiciary] (= Den Danske Rigsdag 1849–1949 Vol V) (Copenhagen, JH Schultz, 1953) 548 f. 46 ‘den Sikkerhed, som maatte kræves … for at tilsidesætte en lov som grundlovsstridig’, DKSC 1921 (Fæsteafløsningsdommen), UfR 1921, 644 (my translation and emphasis). 47 See Wind, ‘Parliament Comes First’ (n 41). 48 See also Koch, ‘Dansk Forfatningsret’ (n 28). 49 Mogens Munch, ‘Grænser for domstolenes retsskabende virksomhed i civile sager’ [‘Limits of Judge-made Law in Civil Cases’] (1989) 71 Juristen 43, 45 (my translation).

96  Marlene Wind The fact that the Danish Supreme Court only once, in 1999, set aside a parliamentary act bears witness to a strong ‘majoritarian political culture’, as Ronald Dworkin has dubbed it.50 In the Scandinavian version of legal positivism, where positive law seen as parliamentary statutes is absolutely central, courts, thus, play a very reticent role in political life, at least when it comes to being active in striking down and scrutinising law-making.51 Courts generally see themselves as ‘Bouches de la loi’ and not as law-makers, and Supreme Court judges sometimes criticise the Luxembourg and Strasbourg Courts for ‘policy making’. As Børge Dahl, a now retired Supreme Court Judge put it recently, ‘we’ (meaning the Danish judges) see it as our role to ‘find the law – not to invent it’.52 As we are to see below, this also impacts on the Danish relation to the EU where, in particular, the CJEU with its strong review powers has been conceived as a somewhat foreign element in Denmark. Scandinavia is also known for its positivist legal tradition53 based on the idea that mainly positive sources of law are considered valid due to the quest for predictability in legal adjudication. The legal text as represented by statutes and other legislative documents like preparatory works is here the most important framework of interpretation for judges. Any reliance on considerations other than the actual wording and the political intention of the legal provision is regarded with suspicion and often as judicial policy-making which judges ought to abstain from as far as possible. This is also why the travaux préparatoires become quite central for the judges because they inform about the intentions and the will of the legislature and, thus, the political majority in Parliament. Thus, the reticent approach by the Supreme Court in 1921 has been regarded as the general doctrine in Danish constitutional law.54 Following former Professor Jens Peter Christensen – now judge at the Danish Supreme Court – the Danish legal philosopher and legal positivist Alf Ross can to a large degree be attributed the idea of an unconstrained legislature as the ultimate symbol of legitimacy and power, a philosophy which has dominated law students in Copenhagen for generations.55 As Christensen puts it: ‘For almost 40 years did this idea dominate the law faculty at the University of Copenhagen’.56 As he further puts it in a footnote: ‘Not until the 2nd edition of Henrik Zahles Dansk Forfatningsret from 1996 did anyone claim that the §3 in the constitution limits the power of the legislator’.57

50 See Ronald Dworkin, Freedom’s Law: The Moral Reading of the American Constitution (Cambridge MA, Harvard University Press, 1996). Dworkin opposed majoritarian and constitutional democracy, taking the absence or presence of judicial review as the core defining feature. 51 It could be said that in a positivist legal system, legal actors would always know what the law is and thus not be surprised by an unwritten obligation or rule. Legal positivism in the Scandinavian version would also be sceptical of judicial discretion, expecting judges to decide cases in accordance with the law, and not their personal preferences or moral considerations. 52 Børge Dahl, ‘Dynamiske domstole, retssikkerhed og demokrati: skal menneskerettigheder udvikles af polikere eller dommere?’ [‘Dynamic Courts, Legal Certainty and Democracy: Should Human Rights be Developed by Politicians or Judges?’] (2017) 99 Juristen 145. 53 See Jes Bjarup, ‘The Philosophy of Scandinavian Legal Realism’ (2005) 18 Ratio Juris 1. Bjarup argues that the metaphysics and normative aspects of law should be destroyed and a scientific theory of law created, to abolish the distorting influence of values. Central figures were Axel Hägerström and the Danish philosopher and jurist Alf Ross. They both focused on positive law as a system of rules in terms of behavioural regularities among human beings. It builds on the idea of positive law as based on statutes and leaves no room for the normativity of law. 54 Cf, eg, Ernst Andersen, Forfatning og Sædvane [Constitution and Customs] (Copenhagen, Gads, 1947); Bent Christensen, ‘Rettens forhold til regeringen tiden efter 1849’ [‘The Court’s Relationship with the Government after 1849’], in P Bagge, L Frost and B Hjejle (eds), Højesteret 1661–1961, Vol I (Copenhagen, Gad, 1961) 401; Alf Ross, Dansk Statsforfatningsret Bind I [Danish Constitutional Law Vol I] (Copenhagen, Nyt Nordisk Forlag, 1980). 55 Jens Peter Christensen, ‘Ross og Statsforfatningsretten’ [Ross and the Constitutional Court], in J Holtermann and J Ryberg, (eds), Alf Ross. Kritiske Gensyn (Copenhagen, Djøf, 2006) 79, 81 (my translation). 56 Ibid. 57 Ibid, 103.

Denmark  97 There is little doubt that one of the important reasons why, until 1999, no Act of Parliament was ever deemed unconstitutional by a Danish court was the strong majoritarianism – institutionalised in the mindset of the elites – which also became central to the neighbouring discipline of political science. To this comes the scepticism regarding natural law, which Ross also campaigned against in many of his writings. This positivist philosophy of legal science which still dominates much legal and political thinking in Denmark is, however, a far cry from the workings of dynamic courts like the CJEU or the ECtHR drawing on normative legal reasoning and societal developments in their interpretation.58 From the above perspective, a more autonomous and dynamic style of interpretation is essentially tantamount to law-making and should be avoided by judges except when absolutely necessary for the reasonable functioning of the legal system.59 The same principles prevail when looking at treaties from a positivist approach. Here positive ‘consent’ is central and treaties (as well as established customary law) are the most important sources. If we look at the European Union and the European Convention on Human Rights (ECHR) it becomes obvious that the treaties are characterised by rather broad principles, which can often only be realised if the interpreter is open to look behind the actual wording of the text and principles, and is willing to base herself on a combination of past law, case law of the Member States and present-day conditions and norms. In a rigid positivist universe, many provisions – in particular human rights provisions – may be declared non-justiciable and often even political statements. In the Nordic countries in general, legal positivism combined with legal realism was prominent throughout the nineteenth and twentieth centuries, represented by (apart from Ross) thinkers like Karl Olivecrona and Alex Hägerström. As Rytter and Wind put it: The realist perspective asks the legal interpreter to interpret legal norms in their political and societal context, to take reality into account. Realism, thus, will even allow for those rights that are justiciable to be restricted by political needs, that is, realism will accept a restrictive political practice as relevant to the legal interpretation of rights.60

However, this does not imply that, as in the present day condition-interpretative style, a judge may reach a conclusion based on normative considerations. Rather, it implies that interpretation should be kept within strict textual limits or restrictions conveyed by political practice or decision-making.61 Or to put it more precisely: ‘Thus, the positivist-realist approach to legal – including constitutional – interpretation more or less takes away whatever potential might be left for judicial review under a strong doctrine of self-restraint’.62 The link between the retracted approach to judicial review and legal positivism/realism comes out clearly in the following citation, which originates in the Danish Supreme Court’s 1921 landmark decision mentioned earlier: ‘the courts may only set aside an Act of Parliament as unconstitutional, if the act’s incompatibility with the text of the Constitution is certain and manifest’.63 Since most other European countries after the Second World War have amended their constitutions to encompass a central judicial review mechanism as part of a living democracy (and as the EU and the ECHR system represent the same philosophy), the Danish hesitance and 58 Olsen, Magtfordeling (n 41). 59 See, eg, Bernhard Gomard, ‘Et retspolitisk program for dommerskabt ret’[‘A Legal Policy Programme for Judge-made Law’], in UfR. Special Edition Højesteret 1661–1986 (Copenhagen, Gad, 1986) 45, 59. 60 Rytter/Wind, ‘Need of Juristocracy’ (n 1) 476. 61 See, with numerous illustrations, Zahle, ‘Sammenstødet’ (n 1); Rytter, Grundrettigheder (n 30) 138–47. 62 Rytter/Wind, ‘Need of Juristocracy’ (n 1) 476. 63 Ibid, 476 f.

98  Marlene Wind commitment to a majoritarian political philosophy with a strong emphasis on the national legislative output, sticks out. As this author has demonstrated elsewhere, there is a close link between this restraint towards challenging and testing the legislature in court at the national level, and the Danish relationship with the increasingly powerful CJEU. The ever-increasing power of the CJEU has – it is demonstrated – influenced the Danish courts’ willingness to refer cases to the CJEU because doing so would ipso facto introduce the institution of judicial review in Denmark that lawyers, politicians and civil society together had fought so hard to avoid.64 We will get back to the Danish courts and their approach to the preliminary ruling procedure in the following sections.

IV.  Constitutional Foundations of EU Membership A. Overview The Danish Constitution does not contain any explicit references to the EU. Unlike many Member States, Denmark’s participation in the EU has (so far) not led to an amendment of the Constitution, even though the amendment in 1953 provided for future transfers of competences to international organisations. This reform of the Danish Constitution predated Denmark’s EC accession with 20 years and none of the constitutional provisions are designed to fit EU cooperation. However, the 1953 amendment of the Constitution did, as noted above, introduce section 20 in order to make it easier for Denmark to join (future) supranational organisations, and it was by virtue of that provision that Denmark could accede to the EC in 1973.65 Section 20 of the 1953-amended Constitution is based on the understanding66 that under the previous version of the Constitution, Denmark could only delegate powers from the national to the supranational level by amending the Constitution in accordance with the cumbersome procedure in section 88 (then section 94). With the establishment of the European Coal and Steel Community (ECSC) in 1951, section 20 was added in order to enable Denmark – if it were to become a member – to delegate powers to supranational organisations without having to amend the Constitution.67 Section 20 has since remained ‘the axis around which any ratification debate revolves in Denmark’.68 Sections 19 (the general foreign affairs provision) and 88 (prescribing the amendment procedure) of the Constitution are, however, the relevant sections when it comes to the

64 See Marlene Wind, Dorte Sindbjerg Martinsen and Gabriel Rotger, ‘The Uneven Legal Push for Europe: Questioning Variation when National Courts go to Europe’ (2009) 10 European Union Politics 63. 65 To be precise, the legal basis of Denmark’s membership to the EC was Law No 447 of 11 October 1972 (tiltrædelsesloven [Accession Act]). This legal act had its constitutional basis in s 20 DKGL. 66 Max Sørensen, ‘Responsum vedrørende de problemer, der knytter sig til grundlovens § 18’ [‘Opinion Regarding the Problems Related to Section 18 of the Constitution’], Rigsdagstidende 1952–53, Tillæg A [Rigsdag Gazette, Appendix A], 3709–38 (part of the travaux préparatoires to s 20). It has been debated whether this so-called ‘Section 20 assumption’ had any real basis in the earlier constitution or whether it should be viewed as a creation of Max Sørensen, who drafted the wording of this section. See eg Jens Hartig Danielsen, ‘One of Many National Constraints on European Integration: section 20 of the Danish Constitution’ (2010) 16 European Public Law 181. 67 Cf travaux préparatoires of the 1953 constitutional amendment, Rigsdagstidende 1952–53, Tillæg A, 3545 f. 68 Hjalte Rasmussen, ‘Denmark’s Waning Constitutionalism and Article 20 of the Constitution on Transfer of Sovereignty’, in A Albi and J Ziller (eds), The European Constitutions and National Constitutions: Ratification and Beyond (Alphen aan den Rijn, Kluwer Law International, 2007) 149.

Denmark  99 formal accession to treaties. Together, these three provisions set out the full range of procedures by which Denmark may enter into international cooperation. Section 88, however, is an integral part of the discussion of Denmark’s relationship with the EU.69 In what follows, we will elaborate a bit on the most important constitutional foundations for Danish membership.

B.  Section 19 Section 19 is the provision on international affairs, including treaty-making powers. From section 19(1) it follows that it is the prerogative of the Government (the executive branch) to act on behalf of Denmark in intergovernmental matters.70 However, the second sentence of section 19(1) requires the Government to obtain the consent of Parliament for the ratification of treaties that impose obligations upon Denmark. Government ratifications of the EU treaties, including the founding treaties and the Lisbon Treaty, have all required the consent of Parliament.71 Parliament’s consent requires a simple majority of the votes cast. The consent of the Parliament is not bound to a particular form, but is typically given by way of a parliamentary resolution or in a statute.72 Section 19 is based on a dualist philosophy.73 International obligations ratified by Denmark, at least as a point of departure,74 only establish obligations for national authorities. Thus, a treaty accession or a treaty amendment, including those concerning the relationship with the EU, must be implemented into Danish law in order to produce internal effects. Section 19 does not deal with the implementation. Therefore, international obligations must be adopted following the ordinary procedure for legislation, ie by simple majority in the Parliament.75 However, if a treaty accession or amendment entails a delegation of powers (or an extension of already delegated powers) from the institutions created by the Constitution – often but misleadingly called a delegation of sovereignty – the qualified procedure of section 20 must be followed.

C.  Section 20 Whereas section 19, inter alia, concerns the assumption by the state of Denmark of international obligations, section 20 concerns the implementation of a certain category of such obligations in Danish law. That is, it concerns obligations that presuppose a delegation of Danish sovereignty.76 Again, the Constitution reflects that the Danish legal system traditionally has adhered to the 69 As, for example, evident from the Maastricht case, where the legal question was whether the Maastricht Treaty should be implemented into Danish law by a section 20 or section 88 procedure, see DKSC 06.04.1998, Case I 361/1997 (Maastrichtsagen. Realiteten), UfR 1998, 800 H, para 9.2. 70 S 19(1) DKGL states: ‘The King acts on behalf of the Kingdom in international affairs.’ As mentioned above, the reference to the ‘King’ is generally a reference to the ‘Government’. 71 Jens Hartig Danielsen, Suverænitet [Sovereignty] (Copenhagen, Djøf, 2010) 185. For example, the Danish Parliament adopted the Act on ratification of the Lisbon Treaty on 24 Apr. 2008, including the necessary parliamentary consent; see Act No 321 of 30 April 2008. 72 Christensen et al, Statsret (n 32) 193. 73 Implied by the fact that the provision distinguishes between the conclusion and fulfilment of international obligations; as pointed out by Max Sørensen, Statsforfatningsret [Constitutional Law], ed P Germer (Copenhagen, Djøf, 1973) 276. 74 See above on the rules of interpretation, assumption and instruction. 75 S 41 DKGL regulates the ordinary legislation procedure. 76 Sørensen, ‘Responsum’ (n 66) 3731 f.

100  Marlene Wind principle of dualism and section 20 has to be understood in light of this principle.77 The wording of the provisions is the following: 1) Powers vested in the authorities of the Realm under this Constitutional Act may, by statute to a certain specified extent be delegated to international authorities set up by mutual agreement with other states for the promotion of international rules of law and co-operation.78 2) For the enactment of a Bill dealing with the above, a majority of five-sixths of the members of the Folketing shall be required. If this majority is not obtained, whereas the majority required for the passing of ordinary Bills is obtained, and if the Government maintains it, the Bill shall be submitted to the electorate for approval or rejection in accordance with the rules for referenda laid down in Section 42.79

In other words, powers that are vested in the legislative, executive and judicial branches may be delegated to a supranational authority provided that the procedure in section 20(2) is followed. Accordingly, a competence transferred to the EU, or another supranational organisation, is then no longer subject to the dualist principle. On the contrary, the supranational organisation may then enact laws directly binding Danish citizens.80 The Danish Supreme Court in its ruling on the constitutionality of the implementation of the Maastricht Treaty of 1998 (hereafter the Maastricht decision) delineated the scope of section 20 along these lines: The application of the qualified procedure in Section 20 of the Constitution is required to the extent that an international organisation is entrusted with the exercise of legislative, administrative or judicial authority with direct effect in this Kingdom, or the exercise of other powers which, according to the Constitution, are vested in the authorities of the Realm, including the power to enter into treaties with other states.81

The passage, in particular, contributes to the interpretation of the formula ‘powers vested in the authorities of the Realm’ in section 20(1). It has been pointed out in legal theory, with reference to the travaux préparatoires, that the reference to ‘powers’ is a reference to section 3 of the Constitution, which embodies the separation of powers.82 This point is seemingly endorsed by the Court’s reference to legislative, administrative and judicial authority with direct effect in

77 Peter Biering and Susanne Lehrer, ‘To Hold a Referendum or Not?’ (2015) 21 European Public Law 169, 172. Henrik Zahle, former Supreme Court judge and professor in constitutional law, referred to s 20 DKGL as a ‘child of dualism’, see Henrik Zahle, EU og den danske grundlov [The EU and the Danish Constitution] (Copenhagen, Christian Ejlers, 1998) 13. 78 Translation of s 20 DKGL taken from the judgment in the so-called Lissabon-case: DKSC 20.02.2013, Case No 199/2012 (Lissabon), U.2013.1451H. This ‘unauthorized, preliminary’ English translation is available at the DKSC website at www.supremecourt.dk/supremecourt/nyheder/ovrigenyheder/Documents/199-12engelsk.pdf. The crucial passage of s 20 DKGL reads: ‘Beføjelser, som efter denne grundlov tilkommer rigets myndigheder, kan ved lov i nærmere bestemt omfang overlades til mellemfolkelige myndigheder’. There are several English translations of this section. Specifically, the difficulties relate to the translation of ‘i et nærmere bestemt omfang’. Rasmussen, ‘Waning’ (n 68) 149, translates it as ‘by statute and to a specific extent’, a translation from the Prime Minister’s office, reads ‘to such extent as provided in a statute’, while Biering/Lehrer, ‘To Hold or Not to Hold?’ (n 77) 172, and Henrik Palmer Olsen, ‘The Danish Supreme Court’s decision on the constitutionality of Denmark’s ratification of the Lisbon Treaty’ (2013) 50 CML Rev 1489, 1491, like the author of this contribution all use the translation ‘by statute to a certain specified extent’, which matches the one used by the DKSC in the aforementioned case. 79 From s 42(6) DKGL it follows that the referendum is corrective: the bill is defeated if a majority of the voters vote against it and this majority represents not less than 30 per cent of all the persons entitled to vote. 80 Biering/Lehrer, ‘To Hold or Not to Hold?’ (n 77) 172. The direct applicability of the supranational act or decision then follows from the supranational aquis itself, not a national incorporation law, see to this extent Zahle, Forfatningsret (n 38) 404. 81 DKSC I 361/1997 (n 69) para 9.2 (unofficial translation, author’s emphasis). 82 Cf Travaux préparatoires (n 67) 3545 f.

Denmark  101 Denmark. As seen in the passage cited above, however, also the other state powers, ‘including’ treaty-making powers (section 19), may be delegated to a supranational organisation.83 Whereas the above concerns what powers can be delegated to the EU (or any other supranational or international organisation) by way of section 20 of the Constitution, other interpretative questions concern the meaning of the wording ‘by statute to a certain specified extent’ in section 20(1). First, this wording raises the question as to what extent powers can be delegated, ie how much power can actually be delegated? According to legal theory, it follows from the wording that a delegation cannot cover all legislative, executive or judicial powers.84 On the other hand, it is established that the formula ‘to a certain specified extent’ does not imply that only insignificant or modest delegations are allowed.85 It also seems that no principle of de minimis should be read into section 20, thereby excluding less important changes from its scope regardless of their form or nature.86 Secondly, the formula ‘to a certain specified extent’ concerns another aspect of central interpretation of section 20: how precise must the description of the delegated power be in the delegating legislation? This is the so-called ‘specification-requirement’. This requirement entails that delegated powers must be enumerated and properly identifiable87 in the delegating act. Seemingly based on the majority view in legal theory,88 the Supreme Court in its Maastricht decision established that The term to a certain specified extent must be interpreted to the effect that a positive delimitation must be made of the powers delegated, partly as regards the fields of responsibility and partly as regards the nature of the powers. The delimitation must enable an assessment to be made of the extent of the delegation of sovereignty. The fields of responsibility may be described in broad categories, and there is no requirement for the extent of the delegation of sovereignty to be stated so precisely that there is no room left for discretion or interpretation. The powers delegated may be indicated by means of reference to a treaty.89

More specifically, the Supreme Court did not find a basis to set aside as unconstitutional the legal arrangement in the Act of Accession to the EU,90 which indicates the powers delegated to the EU by means of reference to the EU treaties.91 Nor did it find the flexibility clause or dynamic and 83 Other powers include the Government’s right to issue money, cf s 26 DKGL. Also the power to impose, repeal or alter taxes enshrined in s 43 DKGL can be delegated, despite the fact that the legislative branch cannot delegate that power to the executive branch in the internal constitutional order, cf Zahle, Forfatningsret (n 38) 412 and Sørensen, Statsforfatningsret (n 73) 310, but differently Ross, Statsforfatningsret I (n 54) 409. 84 Max Sørensen, ‘Responsum fra januar 1971 til Folketingets Markedsudvalg’ (1971) Juristen 434, 436. 85 See DKSC I 361/1997 (n 69) para 9.2. It seems that the Court hereby implicitly referred to the analysis of Sørensen, ‘Folketingets Markedsudvalg’ (n 84) 436. 86 A de minimis limit is, however, seen in the constitutional practice in other countries, for example Norway. Some scholars have attempted to argue in favour of introducing a de minimis standard into s 20 DKGL, see Claus Gulmann, ‘Nogle forfatningsretlige problemer i forbindelse med den danske deltagelse i internationalt samarbejde’ [‘Some Constitutional Problems Related to Danish Participation in International Co-operation’] (1979) 5 Juristen og Økonomen 83–85. However such a view is neither supported in constitutional practice nor in the case law. See on this point Biering/Lehrer, ‘To Hold or Not to Hold?’ (n 77) 176, 190. 87 Olsen, ‘DKSC Lisbon’ (n 78) 1493. 88 Christensen et al, Statsret (n 32) 211 f. It should in this connection be noted that the Danish judges generally do not refer directly to legal doctrine in specific points of its reasoning. 89 DKSC I 361/1997 (n 69) para 9.2, author’s emphasis. 90 Ibid, para 9.3. 91 Cf Accession Act (n 65) with subsequent amendments. S 2 provides that the powers vested in the authorities of the Realm under Constitution can be exercised by the EU institutions to the extent as provided in the treaties enumerated in s 4. The treaties mentioned in s 4 include all the EU treaties, including the treaty amendments, ratified by Denmark. Together ss 2 and 4 thus concern the delegation of powers to the EU in accordance with the s 20 DKGL. Ss 3 and 5 contain rules on the implementation of EU law into Danish law. The former provision incorporates into Danish law EU rules with direct effect which were applicable in 1973 when Denmark acceded to the EC.

102  Marlene Wind rule-making case law of the CJEU in contradiction to the specification requirement.92 Thus, the Maastricht decision can be described as another good example of judicial self-restraint. This is because the court did not go as far as it perhaps could have done. Nevertheless, the judgment at the same time contains elements that have made some compare the decision to the Maastricht decision of the German Constitutional Court which was very critical of the German transfer of new competences to the EU.93 The rhetoric of the Danish Supreme Court was very similar and its combined criticism and self-restraint illustrates its schizophrenic approach to EU law. The principal guarantee against too extensive an application of the provision lies in the formal rules in section 20(2) of the Constitution on the adoption of acts delegating power to an international organisation.94 The section 20 procedure requires – as mentioned above – either a five-sixths majority in Parliament or – if such a qualified majority is not obtainable – approval by referendum. When a treaty, including an EU amendment treaty, requires a delegation of new powers – or extension of already delegated powers – the procedure in section 20(2) has to be used. Otherwise, the procedure in section 19 would suffice and the treaty can be implemented into Danish law by way of the ordinary legislative procedure. The identification of the correct legal basis falls, to begin with, under the competence of the Danish Government.95 In practice, it will first ask the Minister of Justice to give his/her opinion on whether one or more of the provisions of the new EU treaty involve a delegation of powers. The Ministry’s expert lawyers will then analyse the text of the new treaty and publish their conclusions in a report.96 If the report identifies a delegation of new powers to the EU, the Government then proceeds to submit to Parliament a bill of accession delegating powers according to Section 20. So far, neither the Parliament, nor the judiciary has set aside the legal assessments of the Ministry of Justice in this regard which tells us something about the Ministry’s substantial authority and influence in European matters. The section 20 procedure has only been applied in connection with the following treaty changes in the EU: • • • •

Denmark’s accession to the EC in 1972; Denmark’s implementation of the 1992 Maastricht Treaty;97 Denmark’s implementation of the 1997 Amsterdam Treaty; the proposed accession to the euro in 2000 (which was never realised due to the negative outcome of the section 20 referendum in September 2000);98 • the referendum to opt in to the Area of Freedom, Security and Justice (AFSJ) on 3 December 2015; • in the evaluation from 2015 by the Ministry of Justice on whether a future Danish participation in the Banking Union would require a referendum (due to section 20 of the Constitution).99 92 Cf the reasoning in DKSC I 361/1997 (n 69) paras 9.4–9.7. The case is further analysed below. 93 BVerfG 12.10.1993, 2 BvR 2134 (Maastricht) BVerfGE 89, 155; see Rasmussen, Folkestyre (n 23). 94 Cf DKSC I 361/1997 (n 69) para 9.2. 95 Rasmussen, ‘Waning’ (n 68) 150. 96 See, for example, Ministry of Justice’s report of 04.12.2007 on the constitutional issues relating to Denmark’s ratification of the Treaty of Lisbon; and ‘Justitsministeriets redegørelse af Juli 1972 for visse statsretlige spørgsmål i forbindelse med en Dansk tiltrædelse af de Europæiske Fællesskaber’ [‘Ministry of Justice’s Report of July 1972 on the Constitutional Issues relating to Denmark’s Accession to the European Communities’], 41 Nordisk Tidsskrift for International Ret 1971, 67. 97 Technically speaking, the s 20 procedure was applied twice as a result of the negative outcome of the first Maastricht referendum on 2 June 1992. 98 The Constitutional Treaty would have been a fifth occasion. A s 20 referendum was scheduled for September 2005 but was later cancelled due to the negative outcome of the French and Dutch referenda in the spring of 2005. The s 20 procedure was applied without a referendum but with a five-sixths majority in the Folketing for Denmark’s accession to the European Patent Convention in 1989 and the amendments to the Convention in 2005. However, a referendum was held on 25 May 2014 regarding the unitary EU patent, which the Danish electorate approved. 99 Ministry of Justice, 29 April 2015, Notat om hvorvidt dansk deltagelse i det styrkede banksamarbejde forudsætter anvendelse af proceduren i grundlovens § 20 [‘Note on Whether Danish Participation in the Strengthened Banking

Denmark  103 It follows from an express statement in the explanatory memorandum to the draft amendment of the Constitution of 1953 that a section 20 procedure is only required for the adaptation, not the revocation, of a section 20 act.100 Hence, powers delegated to the EU can – at least from a national legal perspective – be withdrawn by an ordinary act adopted by a simple majority in Parliament.

V.  Constitutional Limits to EU Integration A.  Section 20 The constitutional limits to EU integration must be derived from section 20. Further transfer of powers than what is allowed there will require an amendment of the Constitution in accordance with section 88. There are in principle no limits to the delegation of powers as long as the correct procedure is followed, since the Danish Constitution, unlike, for instance the German Basic Law, contains no unamendable core values enshrined in eternity clauses.101 Compared to the constitutions of many other EU Member States, fundamental rights protection in the Danish Constitution is not very comprehensive and reflects the idea of fundamental rights as they were understood at the time of the 1953 constitutional reform, or prior to that. This means that there are very few limits on the legislature. Formally, the ECHR, which was incorporated into Danish law in 1992, plays a more significant role in Danish constitutional practice, but also here new research has shown in particular the Danish Supreme Court to be hesitant.102 Regarding the procedural requirements for normal legislation, legislation pursuant to section 20 and constitutional amendments under section 88, the more intensive the international cooperation, the more involvement of Parliament and the electorate is required.103 It should come as no surprise that the Danish electorate can be rather sceptical towards further EU integration, as for instance illustrated by the Danes’ infamous ‘no’ in the first referendum on the Maastricht Treaty in 1992, but also the more recent referendum on opting into the AFSJ in 2015.104 From a political perspective, it therefore matters a lot whether an EU treaty amendment falls within or outside the legal boundaries of section 20. Making that estimation is, as mentioned above, not left to a constitutional council as in France or a constitutional court as in Germany, but to the political institutions and in particular the anonymous lawyers in the Ministry of Justice. It has, however, become common for concerned and EU-sceptical citizens (with free legal aid from the state) to bring proceedings before the courts to test the constitutionality of EU treaties.105 The legal limits have to some extent been clarified in two important judgments on Denmark’s constitutional position regarding treaty amendments in the EU. In the Maastricht case of 1998,106 Co-operation Presupposes the Application of the Procedure in Section 20 of the Constitution’], www.justitsministeriet.dk/sites/default/files/media/Arbejdsomraader/Vaaben/Notat%20om%20visse%20statsretlige%20 sp%C3%B8rgsm%C3%A5l.pdf. 100 Cf Travaux préparatoires (n 67) 3545 f. For this general view in legal theory see, eg, Christensen et al, Statsret (n 32) 217. 101 Art 79(3) GG. 102 However, this catalogue of rights does not have constitutional status in Danish law, and it is generally assumed that the legislator may choose to depart from the ECHR, cf Jens Elo Rytter, Individets Grundlæggende Rettigheder [The Fundamental Rights of the Individual] (Copenhagen, Djøf, 2013) 50. See also Rytter/Wind, ‘Need of Juristocracy’ (n 1); and Wind, ‘Do Scandinavians’ (n 4). 103 Krunke, ‘Identity’ (n 14) 32. Compare text at nn 72 and 97. 104 Resulting in four opt-outs. 105 This happened in the Maastricht and Lisbon cases. In none of the cases, however, did the Danish Supreme Court rule the new power transfers unconstitutional. 106 DKSC I 361/1997 (n 69).

104  Marlene Wind the Supreme Court delimitated the scope of section 20 ‘up’ to issues requiring amendment of the Constitution pursuant to section 88, whereas the Lisbon case107 from 2013 concerned the delimitation of the legal boundary of section 20 ‘down’ to section 19 and the ordinary legislation procedure. We will go a bit more into the details of these two cases below, as they actually test the constitutionality of the majority’s decision. In that way they engage in some kind of judicial review which is otherwise absent from Danish constitutional culture. However, so far such review has proceeded without setting aside the majority’s decision to transfer new powers to the EU.

1.  The Procedural Door – A Question of Admissibility In order to admit these two cases, the Supreme Court had to make a procedural leap by deviating from the rather strict requirement of locus standi in Danish civil law requiring a specific, individual and current interest of legal nature. In rulings spanning from 1996 all the way to the 1998 Maastricht decision, the Supreme Court accepted – in contrast to its previous jurisprudence – that citizens may have locus standi in cases concerning delegation of powers within areas of general importance to Danish society, thus having a substantive impact on the general population.108 On similar grounds, the Supreme Court admitted the Lisbon case.109 These recent developments mean that transfers of powers under section 20 of the Constitution are no longer left completely to the political decision-making process and legal science. Citizens can test them and the Supreme Court has exercised a form of judicial review by taking on these cases.

2.  The Upper Limit of Section 20 – The Maastricht Case In 1998, the Danish Supreme Court handed down its first landmark decision on Denmark’s relationship with the EU in the Maastricht case.110 The legal question was basically whether Denmark’s implementation of the 1992 Maastricht Treaty should have been done in pursuance of the complicated procedure for constitutional amendment in section 88 instead of the section 20 procedure. The judgment contains a number of relevant and interesting statements. First of all, the judgment confirms the traditional interpretation of section 20(1), according to which the reference to ‘powers vested in the authorities’ implies that an international organisation cannot be granted the power to adopt legislation that contravenes the contents of the Constitution. This is particularly so for the Constitution’s provisions on fundamental rights, as the Constitution does not grant such powers to the Danish constitutional institutions.111 It was additionally settled by the Maastricht decision that it follows from the requirement of section 20(1) that the extent of the delegation must be specified in a legislative act, and that authority cannot be delegated to an international organisation to decide its own powers.112 Hence, the

107 DKSC 199/2012 (n 78). 108 DKSC 12.08.1996, Case I 272/1994 (Maastrichtsagen. Formaliteten), UfR 1996, 1300 H. Hereby, the DKSC overturned its earlier case law (DKSC 04.12.1972, Case I 398/1972 [Aggergren v Queen and PM], UfR 1972, 903 H; DKSC 28.06.1973, Case I 321/1972 [Grønberg v Prime Minister], UfR 1973, 694 H). 109 DKSC 11.01.2011, Case 336/2009 (Hausgaard et al v Prime Minister), UfR 2011, 984 H. 110 DKSC I 361/1997 (n 69). 111 Ibid, para 9.2. The rationale behind it is the following: the content of the constitution can only be altered by way of the amendment procedure in s 88 DKGL. This amendment power, however, cannot be delegated to an international organisation by way of a s 20 procedure, because it is not a power conferred in the legislative, executive or judicial branch. This amendment power in s 88 falls to the electorate (together with the legislator), and the electorate is not one of the ‘authorities of the realm’, cf s 20(1) DKGL; see further, eg, Christensen et al, Statsret (n 32) 211. 112 DKSC I 361/1997 (n 69) para 9.2.

Denmark  105 EU may not be given the competence to define its own competences (no Kompetenz-Kompetenz for the EU). In this connection, the Supreme Court noted that the (then) EC Treaty is based on the principle of conferral and the institutions of the EC may only act within the limits provided by the Treaty.113 The Court accordingly observed: ‘The principle of conferred powers thus implies a restriction on the institutions which is in keeping with the demand for specification in Section 20 of the Constitution’. The plaintiffs had argued that the flexibility clause in Article 235 TEEC in force at that time (now Article 352 TFEU) was applied in such a broad manner that it violated the specification requirement in section 20 of the Constitution and the delegating Act of Accession. The Court rejected this argument by stating that the flexibility clause only authorises adoption of acts without specific authority in areas that lie within the framework created by the other provisions of the Treaty, to attain one of the objectives therein. Thus, the flexibility clause remains within the conferred powers of the EU, as delegated by Denmark, in accordance with section 20 of the Constitution and the Act of Accession. Two other possible substantive limits to the transfer of further powers to the EU through treaty amendments were addressed in the Maastricht ruling: an independent state and a democratic form of government. As regard the first limit, the criterion of an ‘independent state’ is not expressly enumerated in the Constitution, but according to the Supreme Court it is a constitutional precondition.114 However, the Court further stated that the precise scope of this substantive limit relies mostly on considerations of political nature and so far – at the time of the Maastricht Treaty – this precondition had not been violated.115 The reference to the ‘political nature’ of the question is very illustrative of the reluctant review carried out by Danish courts, which are generally more comfortable leaving the interpretation and application of the Constitution with the political actors.116 After dealing with the ‘independent state’ limitation, the Danish Supreme Court turned to what seemingly is a limitation relating to Denmark having a democratic form of government.117 The plaintiffs had claimed that the extent of sovereignty delegated under Denmark’s Act of Accession to the EC Treaty violated the principle of democracy. The Danish Constitution – unlike many other European constitutions – makes no explicit reference to a specific principle of democracy, but according to the plaintiffs such a principle is implied in the Constitution. The Supreme Court seemed to accept the existence of such a constitutional precondition of democracy. In the view of the Court, however, this precondition had not (yet) been violated. The Court observed that transfer of legislative power will always imply a certain encroachment in the Danish democratic system of government, but it emphasised that this was taken into consideration when the extensive procedures in section 20(2) of the Constitution were designed. The Court furthermore stated that it must be for the Danish Parliament to judge whether it is necessary to make the Government’s participation in the EU subject to additional national democratic control.118 113 According to the Court, this interpretation conforms with the note of the Danish Government of 21 January 1997, and ECJ 28.03.1996, Opinion 2/94 (ECHR Accession) ECLI:EU:C:1996:140; see DKSC I 361/1997 (n 69), para 9.4. This interpretation was reaffirmed in DKSC 199/2012 (n 78) at 1519 f. 114 DKSC I 361/1997 (n 69) para 9.8. 115 On the question of ‘independent state’ in relation to the Danish Maastricht case, see further, Ole Spiermann, ‘Hvad kommer efter tyve? En analyse af Højesteretsdommen i “Grundlovssagen”’ [‘What comes after twenty? An analysis of the Supreme Court ruling in the “Constitutional Case”’] (1998) UfR B 325 and for general discussion of the question of ‘independent state’ in relation to s 20 DKGL in the Constitution, see Zahle, EU og grundlov (n 77) 61–66. 116 Also pointed out by Helle Krunke, ‘The Danish Lisbon Judgment – Danish Supreme Court, Case 199/2012, Judgment of 20 February 2013’ (2014) 10 European Constitutional Law Review 542, 565. 117 DKSC I 361/1997 (n 69) para 9.9. 118 See further on this question of democracy, Zahle, EU og grundlov (n 77) 64 f; and Jens Teilberg Søndergaard, ‘Grundlovens forudsætning om demokrati’ [‘The Constitution’s Democratic Precondition’], (2001) UfR B 248. Søndergaard understands the ‘democratic precondition’ more broadly than only confined to the interpretation of s 20.

106  Marlene Wind In legal science it has been debated whether the Court hereby implicitly refused to read into section 20 of the Constitution a so-called ‘structural convergence’ criterion.119 Such a criterion is neither mentioned in the wording of section 20 of the Constitution, nor in the travaux préparatoires. However, whether it is possible to transfer powers in pursuance of section 20 to an international organisation that does not reflect the decision-making procedures and the democratic structures of the Danish constitutional system has been subject to constitutional debate.120 This has been coined as the ‘structural convergence’ requirement. In the aforementioned passage on the democratic precondition, the Supreme Court seemed to distance itself from such a ‘structural convergence’ criterion. Whereas the Court emphatically pointed out that a transfer of legislative power inevitably intervenes in Danish sovereignty and, thus, democracy, it was apparently not concerned with ‘structural convergence’ and the EU institutions to which the powers are transferred.121

3.  The Lower Limit of Section 20 – The Lisbon Case Similarly, the Supreme Court did not seem receptive of a ‘structural convergence’ argument in the Lisbon case. The legal question of this case was basically whether the delegation of powers to the EU in 1973 and 1993 was to the institutions as they existed in 1973 and 1993, so that the total sum of institutional changes to the EU since 1973 would require a new delegation of powers to the revised EU institutions pursuant to a section 20(2) procedure. The Lisbon Treaty was ratified by the ordinary section 19 procedure and implemented into Danish law by the normal legislative procedure.122 The plaintiffs had, inter alia, put forward an argument related to ‘structural convergence’. They claimed that a section 20 procedure was required because the Lisbon Treaty weakened the Danish Parliament’s control of the transferred powers as a result of the increased use of majority voting in the Council and the increased powers of the European Parliament. The Supreme Court responded that the extent to which parliamentary control of powers delegated to the EU is maintained is not decisive for the application of section 20. The focus of the provision is what has been delegated to whom, and, according to this provision, there is nothing to prevent delegating powers to an international organisation as such without specifying the structure and design of that organisation.123 Therefore, neither in the Maastricht nor in the Lisbon decisions did the Supreme Court seem to accept ‘structural convergence’ arguments. Accordingly, the structural convergence criterion has not been transposed from constitutional theory to practice.124 As regards the delimitation of the ‘lower’ boundary of section 20, the Supreme Court held in its Lisbon decision that treaty amendments resulting in institutional changes in the EU, such as the transition from unanimity to qualified majority voting in the Council, do not entail 119 See for an overview of this debate Krunke, ‘Lisbon Judgment’ (n 116) 555 ff with references. 120 Ibid. 121 For this view, Per Lachmann, ‘Grundlovens § 20 og traktater, der ændrer EU’s institutioner’ [‘Section 20 of the Constitution and Treaties that Change the EU Institutions’] (2012) 94 Juristen 259, 262 f and Spiermann, ‘Hvad kommer?’ (n 115) 329. 122 Cf Law No 321 of 30 April 2008. 123 DKSC 199/2012 (n 78) 1516. 124 Cf Lachmann, ‘§ 20 og traktater’ (n 121) 263. However, still after the Maastricht ruling, Zahle advocated for a limited structural convergence criterion, for instance with regard to specific powers of the courts and procedural rules related to these powers that are intended to protect the individual, cf Zahle, Forfatningsret (n 38) 416. As Danish courts are impartial and independent of the legislature and the executive, it is in light of Zahle’s view questionable whether judicial powers can be transferred to an international institution that is not also impartial and independent. This argument was originally put forward by Sørensen, Statsforfatningsret (n 73) 312 f.

Denmark  107 transfer of new powers to the EU in this regard which requires a section 20 procedure.125 The Court, however, established two exceptions to the rule that institutional changes do not require a section 20 procedure. First, the section 20 procedure is required if a treaty amendment results in such fundamental institutional changes within the EU that the EU’s identity is altered, which in the view of the Court, was not the case in concreto.126 Secondly, section 20 must be applied in case a treaty amendment changes the exercise of previously delegated powers contrary to the specifications in the Act of Parliament under section 20 by which the transfer was authorised. The Court held that there was no such specification in the 1972 Accession Act, as subsequently amended.127 The Supreme Court further held that it did not find any basis for setting aside the Government’s and the Parliament’s constitutional assessment that the Lisbon Treaty did not delegate ‘new’ powers to the EU.128 Thus, it was upheld that Denmark’s ratification of the Lisbon Treaty was done correctly within the scope of section 19, followed by an implementing bill approved by simple majority in Parliament. Similarly, treaty amendments, which are merely a codification of powers already conferred upon the EU institutions, whether on the basis of the case law of the CJEU or as a part of the EU institutions’ powers under the so-called Article 352 TFEU, can be done within the scope of section 19 and the ordinary legislation process, with no reference to section 20.129 In sum, in the absence of the transfer of new powers to the EU, the extension of already delegated powers does not activate the procedural hurdle in section 20(2). This means that EU integration in the form of institutional changes, including substantial changes in the democratic structures such as the transition from unanimity to qualified majority voting in the Council, can be ratified and implemented into Danish law by simple majority.130

4.  Danish Opt-outs In Denmark’s case, further EU integration is limited by the country’s four opt-outs.131 These are: 1) EU citizenship,132 2) the Economic and Monetary Union (third phase), ie the euro opt-out,133 3) defence policy and 4) the AFSJ. 125 According to the Supreme Court, s 20 DKGL has consistently been interpreted and applied this way by successive Parliaments and Governments in relation to EU treaty amendments. Olsen, ‘DKSC Lisbon’ (n 78) 1494, points out that it remains unclear whether the Court hereby claimed the existence of customary constitutional law in support of its conclusion or merely noted a political practice for accepting a given interpretation. Either way, the DKSC did not provide any documentation in support of its claim. For a thorough analysis of institutional aspects of section 20 and the Lisbon ruling, see Biering/Lehrer, ‘To Hold or Not to Hold?’ (n 77) 174 f; Krunke, ‘Lisbon Judgment’ (n 116); and Olsen, ‘DKSC Lisbon’ (n 78); see Zahle, Forfatningsret (n 38) 418–22. 126 DKSC 199/2012 (n 78) 1516 f. Such a fundamental identity would, according to the Court ‘be comparable to delegating powers to another international authority’. Biering/Lehrer, ‘To Hold or Not to Hold?’ (n 77) 190 f point out that the ‘new identity’ criterion is ‘hardly clear’. They argue that it would not even be self-evident that the ‘new identity criterion’ would be met, even if the member states chose to dissolve the CJEU as a EU institution. 127 DKSC 199/2012 (n 78) 1517. 128 DKSC 199/2012 (n 78) 1521. For case comments on the Supreme Court’s ruling, see Biering/Lehrer, ‘To Hold or Not to Hold?’ (n 77); Olsen, ‘DKSC Lisbon’ (n 78); and Krunke, ‘Lisbon Judgment’ (n 116). It falls outside the scope of this contribution to go into further details with the specifics of the ruling. 129 DKSC 199/2012 (n 78) 1516. For example, the Court held that explicit adoption of the principle of primacy of EU law with declaration 17 to the Lisbon Treaty was just a codification of the principle, which is firmly established in the case law of the CJEU. Hence a s 20 procedure was not required for that reason. 130 Cf s 19(1) DKGL and the ordinary legislation procedure in Sec 41 DKGL. 131 See further on the Edinburgh agreement and legal aspect of Danish opt-outs, Helle Krunke, ‘From Maastricht to Edinburgh: The Danish Solution’ (2005) 1 European Constitutional Law Review 338. 132 The citizenship opt-out stated that EU citizenship did not replace national citizenship; this opt-out was rendered meaningless when the Amsterdam Treaty adopted the same wording for all members. 133 The abolition of the Euro opt-out was put to a referendum in 2000 and was rejected.

108  Marlene Wind They were secured under the Edinburgh Agreement in 1992 after a majority of Danish voters in a referendum in 1992 initially rejected ratification of the Maastricht Treaty. The four exceptions were the product of a political process trying to come up with a package of measures to accommodate the concerns raised under the first referendum. The Danes ‘bought’ the package, as they, in a referendum in 1993, accepting the Maastricht Treaty with these four opt-outs. Rescinding the Danish opt-outs regarding the euro and the AFSJ on the basis of Protocols 16, 19 and 22 would require delegation of new powers to the EU under section 20 of the Constitution.134 With Protocol 22 to the Lisbon Treaty, Denmark was granted the possibility to exchange the opt-out regarding the AFSJ with an opt-in model inspired by the model applying to UK and Ireland. Accession to this opt-in model requires a section 20 procedure in Denmark, and a referendum was held on 3 December 2015. However, the electorate did not approve the opt-in solution, and the Government together with its yes parties’ coalition lost the referendum with 46.9 per cent yes and 51.1 per cent no.135

B.  Ultra Vires Even though the Danish Supreme Court has only once – as mentioned earlier – held a legislative act unconstitutional (in 1999), it was nevertheless inspired by the decision of the German Bundesverfassungsgericht in the Brunner case to take up principal cases concerning Denmark’s relationship with the EU.136 Five years after the Brunner case it introduced its own ultra vires doctrine in Danish law in its 1998 Maastricht decision. Like the Bundesverfassungsgericht, the Danish Supreme Court raised a constitutional reservation to the supremacy of EU law. It stated that it follows from the requirement ‘to a certain specified extent’ in section 20(1), taken together – interestingly – with the Danish courts’ power to review the constitutionality of legislative acts, that the national courts cannot be deprived of their right to ultimately examine whether a particular

134 See also Krunke, ‘Maastricht to Edinburgh’ (n 131) 348. Abolition of the defence opt-out would probably not require a s 20 procedure, as it does not seem that the EU has the competence to enact decisions and regulations with direct effect for citizens and legal persons in the Member States within the sphere of the CFSP, including the CSDP, cf TEU Ch V.2 and in particular Art 24(2). This is at least the analysis of the Danish Ministry of Justice in relation to its testing of s 20 DKGL against the Lisbon rules on the CFSP and CSDP, cf Ministry of Justice’s report of 4 December 2007 on the constitutional issues relating to Denmark’s ratification of the Treaty of Lisbon, 89–94. Be that as it may, abolition of the defence optout will – politically – require approval in a referendum, since the Danish population has been promised a referendum if any of opt-outs are to be abolished later on, cf the political agreement ‘national compromise’ of 1992. This political promise was reiterated in the political agreement on EU affairs of 21 February 2008, Europapolitisk Aftale [European Political Agreement], supported by the Government and the vast majority of political parties in the Parliament, www. ft.dk/samling/20072/almdel/euu/bilag/150/529685.pdf. 135 Cf the political agreement of December 2014, Statsministeriet, Aftale om Danmark i Europol [Agreement on Denmark in Europol], www.stm.dk/_p_14101.html. The question has been raised in the constitutional debate whether such a model can be allowed within scope of s 20 DKGL. Specifically, the debate regards whether it is possible to delegate sovereignty in advance but before ‘activating’ it. Parliament and Government would in that case be left to decide on a case-by-case basis whether Denmark should take an active part in a specific proposed directive or regulation within the AFSJ, thereby first then ‘activating’ the delegated sovereignty; see the discussion – answered in the affirmative – in the Ministry of Justice’s report of 4 December 2007 on the constitutional issues relating to Denmark’s ratification of the Treaty of Lisbon, 21–28 and 86–89. It is essentially a question of when a s 20 procedure can be carried out, and it was first problematised by H Zahle, ‘Suverænitets-overladelse: Tidspunktet for overladelsen’ [‘Transfer of Sovereignty: The Time of Transfer’] (1992) 74 Juristen 243. 136 BVerfGE 89, 155 (n 93). Koch, ‘Dansk Forfatningsret’ (n 28) 220, argues that it is not possible to fully understand the Danish Maastricht ruling of 1998 without reading it in light of the German Brunner decision; and similarly, Sten Harck and Henrik Palmer Olsen, ‘Decision Concerning the Maastricht Treaty’ (1999) 93 AJIL 209, 213. However, Spiermann, ‘Hvad kommer?’ (n 115) 327, at fn 10, finds no clear ‘kinship’ between the Danish and the German rulings.

Denmark  109 EC legal act exceeds the limits of the transfer of sovereignty determined by a given section 20 act. Consequently, the Court held that Danish courts must rule that an EC act is inapplicable in Denmark if the extraordinary situation should arise that with the required certainty it can be established that an EC act which has been upheld by the EC Court of Justice is based on an application of the Treaty which lies beyond the surrender of sovereignty according to the Accession Act. Similarly, this applies with regard to community-law rules and legal principles which are based on practice of the EC Court of Justice.137

Thus, the Danish Supreme Court stressed its right and obligation to police the constitutional frontier vis-à-vis EU law, reserving the final say as to the constitutionality, and the validity in Denmark, of EU law to Danish courts. The statement included a reference to EU law created by the CJEU. The otherwise hesitant Danish Supreme Court hereby sent a warning to Luxembourg that there are constitutional limits to what Danish courts will accept when it comes to expanding EU powers through judge-made law from the more activist CJEU. However, the devil is as always in the detail and the whole picture sits within a frame. Firstly, it is important to note the reservations which the Danish Supreme Court inserted in its version of the ultra vires doctrine. An ultra vires review is conditional upon a prior assessment by the CJEU.138 Furthermore, the Danish courts can only rule that an EU act is inapplicable if competence is exceeded with the ‘required certainty’. And this, the Supreme Court argues, would be an ‘extraordinary situation’. When these details of reservations are viewed within the frame of the traditional judicial restraint of the Danish courts in cases on the constitutional review of legislation, most were doubtful that the Danish Supreme Court would rule against the CJEU.139 However, as we are to see, this happened recently in the controversial Ajos case.140 The Danish Supreme Court revisited its ultra vires doctrine from its Maastricht ruling 15 years later in its Lisbon ruling. Whereas the Bundesverfassungsgericht in the meantime with its Honeywell case had further narrowed the scope of its ultra vires doctrine, as compared with the Brunner case,141 the Danish Supreme Court simply reaffirmed its position from the Maastricht case in its Lisbon case.142 In particular, the Danish Supreme Court, while still requiring some level of certainty that an EU act or judicial decision has transgressed the delegated competences, it does not require that such transgression is structurally significant, as demanded by the Bundesverfassungsgericht.143 It is also notable that the Danish Supreme Court, unlike its German counterpart in Honeywell, does not apply the language of cooperation with the EU. The Danish Supreme Court thus seems to opt for a more firm (and EU-sceptic) ultra vires doctrine than the Bundesverfassungsgericht. It is, thus, interesting that when it comes to the EU and international cooperation, the Danish Supreme Court seems much more willing to exert judicial review than when it comes to national politics where such threats and such language do not exist. 137 DKSC I 361/1997 (n 69) para 9.6, author’s emphasis. 138 A similar reservation was later introduced by the BVerG 06.07.2010 2 BvR 2661/06 (Honeywell) ECLI:DE:BVerfG:201 0:rs20100706.2bvr266106, para 60; see Mattias Wendel, ‘Lisbon Before the Courts: Comparative Perspectives’ (2011) 7 European Constitutional Law Review 96, 129. 139 This seems to be the majority view in Danish constitutional theory, see eg Olsen, ‘DKSC Lisbon’ (n 78) 1502 f; and Krunke, ‘Lisbon Judgment’ (n 116) 561 ff. So far the DKSC has not acted, but some other national courts have in fact taken steps to claim the invalidity of CJEU rulings. See Robert Zbíral, ‘Czech Constitutional Court, judgment of 31 January 2012, Pl. ÚS 5/12. A Legal revolution or negligible episode? Court of Justice decision proclaimed ultra vires’ (2012) 49 CML Rev 1475, commenting on the Czech Landtová case. 140 DKSC 06.12.2016, Case 15/2014 (Ajos). An unofficial and informal translation is available at www.supremecourt.dk/. 141 Cf Wendel, ‘Before the Courts’ (n 138) 128–31. 142 Compare DKSC I 361/1997 (n 69) para 9.6 to DKSC 199/2012 (n 78) 1520. 143 BVerfG, BvR 2661/06 (n 138) para 61; see Olsen, ‘DKSC Lisbon’ (n 78) 1502.

110  Marlene Wind Moreover, the Danish Supreme Court in Lisbon expanded the scope of the ultra vires control from the judicial to the administrative and political branch. The Court stressed that the Danish authorities are obliged to ensure, in its participation in the EU, that delegated powers are not overstretched.144 Biering and Lehrer, who acted as lawyers on behalf of the Government in the Lisbon case, however, rightly notes that it remains ‘unclear how far this new ultra vires-obligation of the Government extends and whether the Lisbon-decision is merely to be viewed as a polite warning to the Government and the Parliament or rather as an indication of a less forgiving judicature in the future.’145 The Supreme Court repeatedly stressed throughout its Lisbon ruling the possibility of bringing cases before the courts if a specific EU act or CJEU decision, which has a concrete and real impact on Danish citizens, raises doubts as to whether it is based on an application of the treaties that lies beyond Denmark’s previous surrender of sovereignty to the EU.146 Interestingly, the Court, thus, seems to invite cases that can challenge future EU legislation and CJEU case law. Despite its hesitancy to be activist at the national level, the Danish Supreme Court indicates that this principle does not apply to the European legal order. It thus hooks up with other EU-sceptic European supreme and constitutional courts when it comes to EU law and legislation. However, while it was long believed that the Danish courts would not in practice take a tougher stance to ultra vires review of EU acts and CJEU decisions, given the aforementioned traditional judicial restraint of the Danish courts, a recent case indicates a much more confrontational style. In 2016 the Supreme Court ruled against the CJEU in the Ajos case147 which it itself had referred to the CJEU.148 The case concerned the scope of the principle of non-discrimination on grounds of age against the principles of legal certainty and the protection of legitimate expectations.149 Here the Supreme Court sided with Danish law (with one dissenting voice) instead of enforcing EU law, accommodating a judgment which would have been both easy and more obvious. In essence, the CJEU ruled that the general principle prohibiting discrimination on grounds of age precludes national legislation, such as the Danish at issue, which deprives an employee of the right to a severance allowance where the employee is entitled to claim an old-age pension from the employer under a pension scheme which the employee joined before reaching the age of 50, regardless of whether the employee chooses to remain on the employment market or take his retirement.150 The CJEU applied a specific, not entirely novel interpretation of the relation between Directive 2000/78/EC establishing a general framework for equal treatment in employment and occupation151 and the general principle prohibiting discrimination. It found that the directive is merely a specific expression of that principle, and does not go beyond the scope of protection of the principle itself. As a consequence, the CJEU found that the national court must, as a result of the direct applicability of the principle prohibiting discrimination, disapply the discriminatory national provision, even in a dispute between private parties where the 144 DKSC 199/2012 (n 78) 1520: ‘As mentioned above, Denmark’s implementation of the Lisbon Treaty was based on a constitutional assessment that it will not imply delegation of powers requiring application of the s. 20-procedure and the Danish authorities are obliged to ensure that this is observed’ (my emphasis). 145 Biering/Lehrer, ‘To Hold or Not to Hold?’ (n 77) 186. 146 DKSC 199/2012 (n 78), eg 1520 and 1521. 147 DKSC 15/2014 (n 140). 148 CJEU 19.04.2016 C-441/14 (Dansk Industri [DI]) ECLI:EU:C:2016:278. 149 For a very precise analysis see Mikael Rask Madsen, Henrik Palmer Olsen and Urška Šadl, ‘Competing Supremacies and Clashing Institutional Rationalities: the Danish Supreme Court’s Decision in the Ajos Case and the National Limits of Judicial Cooperation’ (2017) 23 European Law Journal 140. 150 CJEU C-441/14 (n 148) paras 21 ff. 151 Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation [2000] OJ L 303/16.

Denmark  111 above-mentioned directive could not be directly applied.152 Furthermore, it denied any option for the national court to avoid such result by balancing it against the principles of legal certainty and the protection of legitimate expectations (of the employer).153 In Ajos, the Danish Supreme Court started by bluntly stating that the CJEU had developed an ‘unwritten principle which applies at treaty level’, while nothing would indicate ‘that there is a specific treaty provision providing the basis for the principle’. A situation such as this, in which a principle at treaty level under EU law is to have direct effect (thereby creating obligations) and be allowed to take precedence over conflicting Danish law in a dispute between individuals, without the principle having any basis in a specific treaty provision, is not foreseen in the Law on accession.154

While the CJEU, the Supreme Court continued, is entitled to develop and establish general principles, such principles ‘are not, however, directly applicable in Denmark by virtue of the Law on accession, and thus cannot be relied on in disputes between individuals’.155 Consequently, ‘the Law on accession does not provide the legal basis to allow the unwritten principle prohibiting discrimination on grounds of age to take precedence over … the Law on salaried employees in so far as the provision is contrary to the prohibition’. The reasoning goes on by affirming: ‘The Supreme Court would be acting outside the scope of its powers as a judicial authority if it were to dis-apply the provision in this situation.’156 This is nothing less than an open rejection of the reasoning and the results of the CJEU’s judgment. It is based on the limits which the Supreme Court takes from its – rather narrow – interpretation of the law of accession as well of the EU Treaties. In this reading, the CJEU would not be entitled to adjudicate on general principles of EU law which would be directly applicable and take precedence over conflicting national legislation. The Supreme Court equally denies that such effect could result from the EU Charter of Fundamental Rights, which, since the Treaty of Lisbon, forms part of primary law, and includes, in Article 21, the general principle prohibiting discrimination. In this respect, the Court157 emphasises the fact that the Treaty of Lisbon had not been passed on the basis of section 20 of the Danish Constitution which would have allowed the additional transfer of powers. The Supreme Court concluded that neither the principles, nor the provisions of the Charter on non-discrimination have been made directly applicable in Denmark by the law on accession. The case caused a lot of controversy in Denmark and, more prominently, in Luxembourg and Brussels. It has been described from many sides as an example of the ‘new sovereigntism in Danish law that is at odds with the project of European integration through law’.158 Most critics see it as a reassertion of national sovereignty and a rejection of EU law supremacy, which has been long underway among Danish judges. It is, thus, rather interesting that when it comes to EU law the Danish Supreme Court seems much more willing to exert judicial review and ‘test’ the legitimacy of supranational legal bodies than when it comes to national politics where you find no such threats and a much more ruler-accommodating style.

152 CJEU C-441/14 (n 148) para 37. In CJEU 12.10.2010, C-499/08 (Ingeniørforeningen i Danmark) ECLI:EU:C:2010:600, which involved a body of the Danish state as employer, the CJEU had already arrived at the same result. This time, however, the controversy was on extending this result to disputes between private parties. 153 CJEU C-441/14 (n 148) paras 28 ff. 154 DKSC 15/2014 (n 140) para 45. 155 Ibid, para 46. 156 Ibid, para 48. 157 Ibid, para 46 f. 158 Madsen/Olsen/Šadl, ‘Competing Supremacies’ (n 149) 140.

112  Marlene Wind This development begs the question: what is the yardstick of this ultra vires review? Is it the national constitutional requirements or the yardstick of EU law? The Danish Supreme Court started by giving a hint with its reference to ‘an application of the Treaty which lies beyond surrender of sovereignty according to the Accession Act’.159 The Accession Act delegates power to the EU by reference to EU treaties.160 The yardstick for ultra vires review, thus, via the Act of Accession seems to be powers conferred on the EU, as spelled out in the treaties, that is, EU law. In Ajos, the Supreme Court confirmed this approach, but went an important step further: it interpreted the powers conferred by the treaties through the lens of the wording of the law on accession. It interpreted the latter narrowly, in that it denied any coverage of general principles by the CJEU which are not expressly spelt out in the treaties. It should be recalled, however, that the Danish Supreme Court has held that on the basis of powers transferred under section 20 in the Constitution, the EU does not have the competence to issue secondary law, which is contrary to the Danish Constitution, including its ‘fundamental rights’.161 It arguably follows that the yardstick for an eventual ultra vires review can also be one of national constitutional requirements, in particular, when it comes to the rather few fundamental rights contained in the Danish Constitution.

C.  Limits to European Integration outside the EU Legal Order Since the Danish Constitution does not contain any specific provisions on the relations with the EU, there is, constitutionally speaking, no difference between the limitations to European integration outside and inside the EU legal order. All the limits must be derived from sections 19 and 20 (and ultimately section 88) of the Constitution, as explained above. Four examples of integration outside the EU treaties could be mentioned briefly: the Fiscal Compact, the European Patent Court, the ESM and the Banking Union. Denmark is party to the intergovernmental Fiscal Compact. Since Denmark is not a eurozone member, and given Denmark’s euro opt-out,162 Denmark’s accession to the Fiscal Compact was enabled by virtue of that treaty’s Article 14(5).163 Denmark ratified the treaty in accordance with its ‘constitutional requirement’ (Article 14[1] of the Fiscal Compact), which specifically was adopted by way of the section 19 procedure, ie by simple majority in Parliament. Thus, Denmark’s accession and implementation was based on the constitutional assessment that the ratification of the Fiscal Compact did not imply a delegation of sovereignty within the meaning of section 20(1), requiring the qualified section 20(2) procedure. The lawyers in the Ministry of Justice, who, as described above, author these legal assessments, stressed in their analysis the intergovernmental 159 See quotation above from DKSC I 361/1997 (n 69) para 9.6; also DKSC 199/2012 (n 78) 1520 and 1521. 160 Cf n 91. 161 Cf DKSC I 361/1997 (n 69) para 9.2. 162 See Protocol (no 16) on certain provisions related to Denmark annexed to the European Union Treaties. 163 Art 14(5) reads: ‘This Treaty shall apply to the Contracting Parties with a derogation, as defined in Article 139(1) of the Treaty on the Functioning of the European Union, or with an exemption, as referred to in Protocol (No 16) on certain provisions related to Denmark annexed to the European Union Treaties, which have ratified this Treaty, as from the date when the decision abrogating that derogation or exemption takes effect, unless the Contracting Party concerned declares its intention to be bound at an earlier date by all or part of the provisions in Titles III and IV of this Treaty’. Denmark’s euro opt-out within the EU legal order is translated into this Declaration in relation to Denmark’s ratification of the Treaty: ‘As a consequence of its obligations with regard to the Treaty, Denmark will not be bound by any EU legislative rules which result from subsequent implementation of Titles III and IV of the Treaty and are adopted on the basis of provisions of the Treaties on the European Union which are applicable only to those Member States which have the euro as their currency.’

Denmark  113 character of the Fiscal Compact. The treaty, it was held, only imposes obligations on the contracting states as such, and does not imply a delegation of powers to enact law with direct effect for Danish citizens (which would have required the section 20(2) procedure).164 The situation was different with Denmark’s accession to the European Patent Convention. This required a section 20 procedure, as the Convention meant that powers conferred to Danish authorities by the Constitution would be transferred to the European Patent Court. The new European Patent Court has jurisdiction to make final decisions in proceedings between private parties on patents covered by the Treaty, and to make decisions regarding evidence in such cases.165 Danish accession to the ESM is not a current legal issue, as Denmark (so far) is not a member of the eurozone. As mentioned above, Danish abolition of its euro opt-out would require a section 20 procedure. Most recently, the Ministry of Justice published a report, in which it is held that Denmark can accede to the Banking Union by a section 19 procedure, thus avoiding the more complicated section 20 procedure which requires a referendum.166

VI.  Constitutional Rules and Practice on Implementing EU Law A.  Denmark and the Preliminary Reference Procedure167 The examples given above of the Danish Supreme Court warning against the power curb by the CJEU and siding with the Bundesverfassungsgericht is also interesting in light of the Danish courts’ somewhat reluctant approach to the preliminary reference procedure. In Ajos, however, the Supreme Court took a step which is unprecedented even for the Bundesverfassungsgericht by openly denying, in contrast to the CJEU, direct effect of the fundamental right to non-discrimination on reasons of age in disputes between individuals. Due to the more or less explicit Danish adherence to a Dworkinian majoritarian democracyconception with the lawmaker at the centre, Danish courts – and the rather powerful Ministry of Justice managing daily EU law business – have not been proactive in forwarding cases to the CJEU for interpretation. The preliminary ruling procedure which was introduced into the Union with the Treaty of Rome in 1957 (then Article 177 TEEC, now Article 267 TFEU) introduces a judicial review mechanism foreign to Danish legal and political tradition.168 As mentioned in my previous research on this issue, political and legal cultures with no tradition of judicial review clearly must regard it as counter-intuitive to ask a supranational court to evaluate and judge on 164 See ‘Notat om visse forfatningsretlige spørgsmål i forbindelse med Danmarks ratifikation af traktaten om stabilitet, samordning og styring i Den Økonomiske og Monetære Union (den såkaldte finanspagt’) [‘Memorandum on Constitutional Issues Relating to Denmark’s accession to Treaty on Stability, Coordination and Governance in the Economic and Monetary Union (the so-called Fiscal Compact)’], Ministry of Justice, 22. February 2012. 165 ‘Notat om dansk tilslutning til aftale om en fælles europæisk domstol/grundlovens § 20’ [‘Memorandum on the Danish Accession to an Agreement on a Common European Court/s 20 DKGL’], Ministry of Justice, 7 May 2013. 166 ‘Notat banksamarbejde’ (n 99). 167 This section draws on Marlene Wind, Who is Afraid of European Constitutionalism? The Nordic Distress with Judicial Review and Constitutional Democracy, iCourts Working Paper Series, No 12 (2014); and Dorte Sindbjerg Martinsen and Marlene Wind, ‘When National Courts go to Europe: Reluctant or active players in the integrationprocess?’, in H Koch, J H H Weiler et al (eds), Europe. The New Legal Legal Realism: Essays in Honour of Hjalte Rasmussen (Århus, Djøf, 2010) 479. 168 Wind/Martinsen/Rotger, ‘Legal Push’ (n 64); and Rytter/Wind, ‘Need of Juristocracy’ (n 1); Wind, ‘Foot-dragging Supporters’ (n 33).

114  Marlene Wind what the lawmaker (and administration) has been deciding when implementing EU law. What makes it even more difficult for a dualist and positivist legal system is probably that we are dealing with a supranational court, which does not reveal dissenting votes and which employs a dynamic style of legal interpretation uncommon to the Danish legal tradition. The EU’s preliminary ruling mechanism in this manner not only introduces judicial review into a Danish legal and political system which has been explicitly and consistently rejected for centuries by national courts (politicians and legal/political theorists), but it also introduces collaboration with a constitutional judicial system that celebrates legal activism and confronts the Danish ‘self-restraint philosophy’ head-on. Referring cases to the CJEU is nevertheless a cornerstone of the EU constitutional order and something that all Nordic countries have had to learn to live with – even Iceland and Norway, which are part of the European Economic Area (EEA).169 In order to make sure that the EU’s legal system develops in a harmonious and consistent manner, it has always been essential that national courts engage in an open and deliberate dialogue with the CJEU. In a long-term perspective, national courts – in the EU as a whole – have indeed accepted and taken up this role.170 It should also be emphasised that even though data on this from the first decades reveal some rather hesitant Nordic Member States, things are clearly not static, and Danish courts have recently become less hesitant to use the mechanism. Overall, Article 267 TFEU references have skyrocketed in number between 1961, when the first preliminary reference was forwarded to the Court, and 2008. By the end of 2008, 6,317 preliminary references had been made to the CJEU. From 1993 onwards, more than 200 references were made annually, with a maximum of 288 references in 2008.171 The interplay between national courts and the CJEU is, thus, a growth factor in its own right in the European integration process. It has often been interpreted as an indication of a growing judicial dialogue between judges and as an increased true and unprecedented constitutional ‘transformation of Europe’.172 However, as indicated above, not all national courts have contributed equally to this general trend. The differences in the total number of references per Member State may be explained by various factors. The length of membership seems to play a role, as Belgium, Germany, France, Italy and the Netherlands have had the highest number of references, whereas the more recent EU-12 Member States for the most part have made much fewer. If we take the years of membership into account, heterogeneity is, however, still significant. It has been argued that population size may in part explain the heterogeneity across member states’ reference practice, ie that the larger states will refer more often than smaller states.173 However, when controlling for this factor, the variance across Member States is still remarkable. In my previous work on this matter, I demonstrated that there is no significant causality between population size and preliminary reference practice.174 In political science terminology, this means

169 JHH Weiler, ‘The Transformation of Europe’ (1991) 100 The Yale Law Journal 2403. 170 Karen Alter, ‘The European Union’s Legal System and Domestic Policy: Spillover or Backlash?’ (2000) 54 International Organizations 489; see also Karen Alter, Establishing the Supremacy of European Law. The Making of an International Rule of Law in Europe (Oxford, OUP, 2001). 171 Data from curia.europa.eu. 172 See Joseph HH Weiler, The Constitution of Europe. ‘Do the New Clothes Have an Emperor?’ and Other Essays on European Integration (Cambridge, CUP, 1999). 173 Niels Fenger, ‘Om danske domstoles relative tilbøjelighed til at forelægge præjudicielle spørgsmål for EU-domstolen’ [‘About the Relative Tendency of Danish Courts to Refer Questions to the CJEU’] (2009) 91 Juristen 269. See also Morten Broberg and Niels Fenger, Preliminary references to the European Court of Justice, 2nd edn (Oxford, OUP, 2014). 174 Wind/Martinsen/Rotger, ‘Legal Push’ (n 64) 79.

Denmark  115 that population size cannot explain the different number of preliminary references that come from any individual country. A very good example of this becomes apparent when we compare Austria and Sweden. They entered the EU the same year – in 1995 – and have roughly the same population sizes. However, when looking at the figures, it is clear that Austria has referred more than four times the number of cases that Sweden has.175 To put it differently, differences in population size do not explain why some Member States refer more cases than others. After all, individuals trying to make their case before the CJEU have a long, tiring and troubled way to go before reaching the EU’s judiciary. In other words, the absence of causality between population size and preliminary references substantiates the plausibility of what was previously proposed: namely that ‘type of democracy’ and tradition of judicial review may at least partly explain why the courts of some Member States are hesitant to enter into a systematic and continuing dialogue with the CJEU.

B.  Denmark, the Ministry of Justice and Implementation of EU Law As an EU member since 1973, Denmark has had considerable time to accustom itself to the European legal system. It is also no secret that Denmark is often referred to as ‘the dutiful pupil in the class’ when it comes to formal implementation of EU legislation. This is no doubt a correct description if we are merely interested in formal and not practical implementation.176 Moreover, as Hagel Sørensen, a long-time legal advisor to the Danish Government, has pointed out, the reason for Denmark’s high and fast implementation frequency concerning EU legislative acts can be explained by the fact that these are executed administratively though bekendtgørelser, that is, ‘executive orders’. In other EU Member States, Italy for instance, the delegation from the legislative to the administrative branch is subject to much stricter constitutional procedures, which clearly influence how fast and vividly the implementation of EU law can be processed.177 Regarding sufficient implementation in practice, Denmark does, however, not always live up to its reputation as one of the most obedient compliers.178 Returning to the role of courts, previous research has shown that Denmark is one of the Member States that has made the fewest preliminary references to the CJEU; a total of 122 between 1973 and 2008 or an average of 3.39 cases per year of membership. Earlier studies have demonstrated that the preliminary reference procedure in Denmark is in part conditioned on an extraordinarily close relationship between the powerful executive branch (the Ministries of Justice and Foreign Affairs) and the Danish courts.179 Historically, there has 175 Wind, ‘Nordics’ (n 34); see also Rytter/Wind, ‘Need of Juristocracy’ (n 1). 176 The difference between formal and practical is important. Formal simply deals with objective time deadlines of implementation, ie whether a certain directive is transposed into national law by a certain pre-specified time limit. It is the Ministries themselves who report to the Commission whether this deadline has been met or not. Practical implementation on the other hand deals with the actual and practical transformation of an EU legislative act on the ground in the local administration, vis-a-vis citizens and firms and on the homepages communicating about EU rights and obligations to the broader public. 177 Karsten Hagel-Sørensen, ‘Fællesskabsretten som en del af dansk ret’ [‘Community Law as Part of Danish Law’], in J Rosenløv and K Thorup (eds), Festskrift til Ole Due (Copenhagen, Gads, 1994) 115; see also the report on Italy in this volume. 178 Dorte Sindbjerg Martinsen, ‘The Europeanisation of Welfare – The Domestic Impact of Intra-European Social Security’ (2005) 43 Journal of Common Market Studies 1027; Martinsen, ‘The Europeanisation of gender equality: who controls the scope of non-discrimination?’ (2007) 14 Journal of European Public Policy 544; Martinsen and Karsten Vrangbæk, ‘The Europeanization of Health Care Governance: Implementing the Market Imperatives of Europe’ (2008) 86 Public Administration 169. 179 Peter Pagh-Rasmussen, ‘Præjudicielle forelæggelser og Juridisk Specialudvalg’ [‘Preliminary Rulings and the Judicial Committee’] (2004) UfR 305; Pagh-Rasmussen, ‘Bopælspligt, menneskerettigheder og EU-ret: en kommentar til

116  Marlene Wind always been a close relationship between the Ministry of Justice and the national courts. Until 1999, Danish judges were exclusively recruited from the Ministry of Justice,180 and the loyalty to this executive body remains almost unchallenged (interviews, Danish judges and civil servants, March 2006).181 This in part explains why the so-called Judicial Committee (originating in the Ministry of Justice) plays a cardinal role when it comes to preliminary references but also regarding implementation and Government interventions before the CJEU. Pagh-Rasmussen demonstrates how the Judicial Committee not only advises the Danish courts through the attorney of the Danish state by participating in the selection and drafting of Article 267 preliminary questions, but also plays a strong role as gate-keeping role vis-à-vis EU law in general.182 However, while the judicial advisors just see their interference in the preliminary reference procedure as ‘doing their job’ in terms of serving the Government and, thus, eventually, the Danish state’s interests, it is still a rather peculiar arrangement. This is because it will often be the same people who advise on the implementation of EU law and deal with infringement cases against the Danish state. This may result in situations where there is very little incentive among the executive branch to have their own policy advice concerning the implementation of EU law tested by a supranational court. This problem has been illustrated by Pagh-Rasmussen, who looked at cases that probably should have been referred to the CJEU. From 1986 to 2003 the Judicial Committee recommended to the national courts not to refer a case to the CJEU in 20 out of 26 instances, even though all 26 cases dealt with the interpretation of EU law and at least one of the parties had explicitly requested an interpretation by the CJEU.183 Generally speaking, the Judicial Committee, the author argues, has only recommended Danish courts to make preliminary references in those cases where there is already direct action being taken against Denmark by the Commission.184 Whereas the Judicial Committee may ‘just be doing its job’ (that is, advising one party in a case, the Danish Government), it is perhaps more puzzling that the Danish courts treat this advice as ‘the highest legal expertise’, as one of the judges I interviewed put it.185 However, being brought up in the Ministry of Justice themselves, most judges may feel inclined to take advice from this source very seriously. As there is no tradition of exercising judicial review of legislation in Denmark, the preliminary reference procedure has, as mentioned above, often been regarded as an unnecessary foreign element interfering in matters which should be dealt with by the national court system itself.186 Danish courts have, thus, repeatedly invoked the acte clair doctrine, and thereby – it seems – avoided asking the CJEU to clarify whether national law was in breach with EU law. From empirical legal studies of this phenomenon, it has been shown that there is an enormous unease and unfamiliarity with the European legal system, which probably has been further

U 2007.99 H i lyset af nye EF-domme’ [‘Residence Obligation, Human Rights and EU Law: a Comment on U 2007.99 H in the Light of new EC judgments’], (2007) UfR 126. See also Wind, ‘Nordics’ (n 34). 180 Ole Hammerslev, ‘The Development of the Danish Legal Profession’ (2008) 53 Scandinavian Studies in Law 283. 181 Jens Peter Christensen, Domstolene – den tredje statsmagt [The Courts – The Third Power in the State] (Århus, Magtudredningen, 2003) 75–90. 182 Pagh-Rasmussen, ‘Præjudicielle’ (n 180). 183 Ibid 307. 184 In particular, the infamous ‘can case’, ECJ 11.09.2002, C-246/99 (Commission v Denmark) ECLI:EU:C:2002:477, in which Denmark was charged by the Commission for hindering the implementation of Directive 94/62/EC of 20 December 1994 on packaging and packaging waste, [1994] OJ L 365/10, dealing with the marketing of cans instead of ordinary bottles on the Danish market. 185 See survey among Danish judges in Wind, ‘Nordics’ (n 34). 186 Own research; see also Pagh-Rasmussen, ‘Præjudicielle’ (n 180).

Denmark  117 emphasised by the dominance of legal positivism in Scandinavia and its ‘anti-rights’ tradition.187 In Denmark there is a firm conviction that rights (which are considered normative and, thus, political) should be decided on and protected by the politicians – not the courts. This was also seen when Denmark in the autumn of 2017 chaired the Council of Europe Committee of Ministers. Here the Government – following the British Government’s Brighton Declaration of 2012188 – wanted to ‘Take human rights back home’, curbing the power of the Strasbourg Court.189 The attempt by Danish politicians to let the ECtHR play a less prominent role in the protection of human rights did not, however, go down well with many other governments and ended up with a somewhat watered down Copenhagen Declaration.190 It is nevertheless telling that the majority in the Danish Parliament, including many legal experts, judges and opinion makers (though not all law professors), could not see the problem and fully supported the wish to let national politicians rather than international judges protect human rights. When critics pointed to the irony of having less democratic regimes applaud the original Danish draft declaration, few paid attention. At a time when illiberalism sweeps the continent and more and more political leaders attempt to take control over national courts, limiting the role of international guardians may be devastating for our overall basic rights protection. If and when national courts and politicians refuse (deliberately or by other means) to protect fundamental rights and even suggest to bring fewer cases to the CJEU and the ECtHR or to limit their powers, the losers will always be those most in need of protection.

VII. Conclusion Denmark has often been portrayed as a reluctant European state which nevertheless is good at keeping its commitments. As Joseph Weiler responded in a seminar on the future of Europe at the European University Institute (EUI), when I presented the somewhat peculiar Nordic reluctance to the integration project: ‘I would always prefer a reticent European Member State which sticks to its commitments than a Member State that promises a lot, but never lives up to what it has promised’.191 This may be worth keeping in mind even though also the ‘sticking to promises’ may not always be as clear-cut as we assume. It is in any case interesting that Europe contains so many variations when it comes to constitutional cultures – something that this book also demonstrates. In Denmark where it is almost impossible to amend the constitution, pragmatism and political and executive sensibilities have to a large extend been guiding the transfer of sovereignty and the relationship to the EU as a whole. There is also no doubt that it has mattered to the Danish position that we entered the EU for mainly pragmatic and commercial reasons together with the United Kingdom. We also share with the British an obsession with sovereignty (and the Parliament) even though few – as a new 187 See Wind, ‘Do Scandinavians’ (n 4). See also Joakim Nergelius, ‘North and South: Can the Nordic States and the European Continent Find each other in the Constitutional Area – or are they too different?’, in M Scheinin (ed), The Welfare State and Constitutionalism in the Nordic Countries (Copenhagen, Nordic Council of Ministers, 2001) 79; Rytter, Grundrettigheder (n 30); see also Rytter/Wind, ‘Need of Juristocracy’ (n 1). 188 www.echr.coe.int/Documents/2012_Brighton_FinalDeclaration_ENG.pdf. 189 For a critical comment by Rask Madsen and Christoffersen (2018) see www.ejiltalk.org/the-european-court-ofhuman-rights-view-of-the-draft-copenhagen-declaration/. 190 https://rm.coe.int/copenhagen-declaration/16807b915c. 191 Seminar ‘The Future of Europe’ at the EUI, 18.05.2016. A conversation between Joseph Weiler, EUI and Marlene Wind, moderated by Bruno de Witte, EUI.

118  Marlene Wind study shows192 – know what that means in constitutional terms. For most Danes, sovereignty is a zero-sum game where European decision-making is always seen as a cost where powers are taken away from the national level. While there is little risk that Denmark may follow the British example of leaving the EU – in particular not after the mess that the British have put themselves into – we must also admit that very few Danish politicians over the years have spoken in favour of the European cause. So when citizens express their more or less fuzzy fear of ‘loosing sovereignty’ to EU bureaucrats or to judges in Luxembourg, politicians have rarely seen an interest in explaining that doing things together for instance on the environment, climate or single market regulation, is in fact increasing rather than undermining sovereignty. One so to speak becomes ‘rule maker’ instead of ‘rule taker’ as Norway is and as the British will be once they have left. No matter what arrangement the British will get with Europe they will be fundamentally dependent on European standards, rules and regulations without having any influence on them. These kinds of insights and arguments are, however, in the Danish debate left to a small handful of experts and commentators. This is a shame and may in the long run undermine the Danish people’s trust in the European project.193 In terms of where Denmark will be moving in the future, constitutionally there would be no formal hindrances in moving closer to the centre. Denmark could join the European core though there could of course be areas where the Ministry of Justice might consider that we needed a new referendum. However, the Ministry of Justice has already some time ago concluded that Denmark could join the Banking Union without a referendum as that there are no formal constitutional implications in terms of transferral of sovereignty. The real obstacle for Denmark when it comes to integrating further is really rather the self-inflicted 25-year-old Danish opt-outs which seem to be very hard to get rid of. We had – as mentioned earlier – a referendum on the AFSJ opt-out in 2015 with no success and even though the Danes often even enthusiastically support military actions outside Denmark, we still have a defence opt-out as well. This opt-out may, however, be the easiest to challenge in a referendum as opinion polls suggest. At the moment, at least a majority of Danes would like to skip it. The same does not go for the euro opt-out, which probably is the most important when it comes to closer European integration. We had a referendum on the euro in 2000 and a majority of Danes rejected it. The Danes want to keep the Danish krone even though it is pegged to the euro and despite the fact that the Danish Central Bank raises or lowers its interest rate less than five minutes after it is done in Frankfurt by the ECB. When looking at the national courts, the lower courts have in fact started to engage more with EU law. Both in terms of applying it in court194 but also in terms of referring cases, even though reticence is still the order of the day. As regards the Danish Supreme Court I am less sure. The Ajos case from 2016 suggests a ‘new sovereigntism’ which has been building up for decades and which now seem to be consolidating, as Madsen, Olsen and Šadl put it: Echoing the mid-20th century sovereigntist theories of public international law, the EU is basically treated [by the Supreme Court, MW] as entirely residual to Danish law. The SCDK [Supreme Court], which has not been known for disturbing the operations of EU law, makes a clear statement in favour of Danish law and legal culture.195

192 Reference to Tænketanken EUROPA, Suveræne Danmark – Danskernes fortællinger om EU og suverænitet [Sovereign Denmark – The Danes’ Narrative of the EU and Sovereignty], Copenhagen, September 2017, http://thinkeuropa.dk/pdf/ Rapport_Suveraene_Danmark_final.pdf. 193 See my op-ed ‘Incompetent political leaders and the media are destroying Europe’, EURACTIV, 31.05.2016, www. euractiv.com/section/future-eu/opinion/incompetent-political-leaders-and-the-media-are-destroying-europe/. 194 Wind, ‘Do Scandinavians’ (n 4). 195 Madsen/Olsen/Šadl, ‘Competing Supremacies’ (n 149).

Denmark  119 Similar conclusions were reached by other scholars as well.196 Whether what we see in Ajos is plain nationalism or a revolt against European law coming from judges who were born out of a majoritarian legal and political culture where courts keep a very low profile and normative considerations and dynamic interpretations always considered political, is hard to tell. However, one might ask whether a serious matter like challenging EU law supremacy is not a very prolific political deed that ought to be left to those who are elected?

References K Alter, ‘The European Union’s Legal System and Domestic Policy: Spillover or Backlash?’, (2000) 54 International Organizations 489. K Alter, Establishing the Supremacy of European Law. The Making of an International Rule of Law in Europe (Oxford, OUP, 2001). E Andersen, Forfatning og Sædvane [Constitution and Customs] (Copenhagen, Gads, 1947). P Andersen, Rigsdagen og Domstolene [Parliament and the Judiciary] (= Den Danske Rigsdag 1849–1949 Vol V) (Copenhagen, J. H. Schultz, 1953). AM Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of politics (Indianapolis IN, BobbsMerrill Educational Publishing, 1962). P Biering and S Lehrer, ‘To Hold a Referendum or Not?’ (2015) 21 European Public Law 169. J Bjarup, ‘The Philosophy of Scandinavian Legal Realism’ (2005) 18 Ratio Juris 1. M Broberg and N Fenger, Preliminary references to the European Court of Justice, 2nd edn (Oxford, OUP, 2014). B Christensen, ‘Rettens forhold til regeringen tiden efter 1849’ [‘The Court’s Relationship with the Government after 1849’], in P Bagge, L Frost and B Hjejle (eds), Højesteret 1661–1961, Vol I (Copenhagen, Gad, 1961) 410. JP Christensen, Domstolene – den tredje statsmagt [The Courts – The Third Power in the State] (Århus, Magtudredningen, 2003). JP Christensen, ‘Ross og Statsforfatningsretten’ [‘Ross and the Constitutional Court’], in J Holtermann and J Ryberg, (eds), Alf Ross. Kritiske Gensyn (Copenhagen, Djøf, 2006) 79. JP Christensen, JA Jensen and MH Jensen, Dansk Statsret [Danish Constitutional Law] (Copenhagen, Djøf, 2012). PM Christiansen and L Togeby, ‘Power and Democracy in Denmark: Still a Viable Democracy’ (2006) 29 Scandinavian Political Studies 1. J Christoffersen, ‘Folkerettens virkning i dansk ret’ [‘The Effect of International Law in Danish Law’] (2001) Ugeskrift for Retsvæsen 143. J Christoffersen and M R Madsen, ‘The End of Virtue? Denmark and the Internationalisation of Human Rights’ (2011) 80 Nordic Journal of International Law 257. B Dahl, ‘Dynamiske domstole, retssikkerhed og demokrati: skal menneskerettigheder udvikles af polikere eller dommere?’ [‘Dynamic Courts, Legal Certainty and Democracy: Should Human Rights be Developed by Politicians or Judges?’] (2017) 99 Juristen 145. JH Danielsen, ‘One of Many National Constraints on European Integration: section 20 of the Danish Constitution’ (2010) 16 European Public Law 181. JH Danielsen, Suverænitet [Sovereignty] (Copenhagen, Djøf, 2010). R Dworkin, Taking Rights Seriously (Cambridge MA, Harvard University Press, 1977). R Dworkin, Freedom’s Law: The Moral Reading of the American Constitution (Cambridge MA, Harvard University Press, 1996).

196 Ulla Neergaard and Karsten Engsig Sørensen, ‘Activist Infight among Courts and Breakdown of Mutual trust? The Danish Supreme Court and the Ajos Case’, 36 Yearbook of European Law 2017, 275.

120  Marlene Wind N Fenger, ‘Om danske domstoles relative tilbøjelighed til at forelægge præjudicielle spørgsmål for EU-domstolen’ [‘About the Relative Tendency of Danish Courts to Refer Questions to the CJEU’] (2009) 91 Juristen 269. B Gomard, ‘Et retspolitisk program for dommerskabt ret’ [‘A Legal Policy Programme for Judge-made Law’], Ugeskrift for Retsvæsen. Special Edition Højesteret 1661–1986 (Copenhagen, Gad, 1986) 45. C Gulmann, ‘Nogle forfatningsretlige problemer i forbindelse med den danske deltagelse i internationalt samarbejde’ [‘Some Constitutional Problems Related to Danish Participation in International Cooperation’] (1979) 5 Juristen og Økonomen. K Hagel-Sørensen, ‘Fællesskabsretten som en del af dansk ret’ [‘Community Law as Part of Danish Law’], in J Rosenløv and K Thorup (eds), Festskrift til Ole Due (Copenhagen, Gads, 1994) 115. O Hammerslev, ‘The Development of the Danish Legal Profession’ (2008) 53 Scandinavian Studies in Law 283. S Harck and H P Olsen, ‘Decision Concerning the Maastricht Treaty’ (1999) 93 AJIL 209. T Knudsen, Fra Enevælde til Folkestyre. Dansk demokratihistorie indtil 1973 [From Absolutism to People’s Government. Danish Democracy History until 1973] (Copenhagen, Akademisk Forlag, 2006). H Koch, ‘Dansk Forfatningsret i Transnational Belysning’ [‘Danish Constitutional Law in Transnational Light’] (1999) 81 Juristen 213. H Krunke, ‘From Maastricht to Edinburgh: The Danish Solution’ (2005) 1 European Constitutional Law Review 338. H Krunke, ‘Constitutional Identity – Seen through a Danish Lens’ (2014) 37 Retfærd No 4/147, 24. H Krunke, ‘The Danish Lisbon Judgment – Danish Supreme Court, Case 199/2012, Judgment of 20 February 2013’ (2014) 10 European Constitutional Law Review 542. P Lachmann, ‘Grundlovens § 20 og traktater, der ændrer EU’s institutioner’ [‘Section 20 of the Constitution and Treaties that Change the EU Institutions’] (2012) 94 Juristen 259. MR Madsen, HP Olsen and U Šadl, ‘Competing Supremacies and Clashing Institutional Rationalities: the Danish Supreme Court’s Decision in the Ajos Case and the National Limits of Judicial Cooperation’ (2017) 23 European Law Journal 140. DS Martinsen, ‘The Europeanisation of Welfare – The Domestic Impact of Intra-European Social Security’ (2005) 43 Journal of Common Market Studies 1027. DS Martinsen, ‘The Europeanisation of gender equality: who controls the scope of non-discrimination?’ (2007) 14 Journal of European Public Policy 544. DS Martinsen and K Vrangbæk, ‘The Europeanization of Health Care Governance: Implementing the Market Imperatives of Europe’ (2008) 86 Public Administration 169. DS Martinsen and M Wind, ‘When National Courts go to Europe: Reluctant or active players in the integration process?’, in H Koch, JH Weiler et al (eds), Europe. The New Legal Legal Realism: Essays in Honour of Hjalte Rasmussen (Århus, Djøf, 2010) 479. M Munch, ‘Grænser for domstolenes retsskabende virksomhed i civile sager’ [‘Limits of Judge-made Law in Civil Cases’] (1989) 71 Juristen 43. U Neergaard and KE Sørensen, ‘Activist Infight among Courts and Breakdown of Mutual trust? The Danish Supreme Court and the Ajos Case’ (2017) 36 Yearbook of European Law 275. J Nergelius, ‘North and South: Can the Nordic States and the European Continent Find each other in the Constitutional Area – or are they too different?’, in M Scheinin (ed), The Welfare State and Constitutionalism in the Nordic Countries (Copenhagen, Nordic Council of Ministers, 2001) 79. HP Olsen, Magtfordeling: en analyse af magtfordelingslæren med særligt henblik på den lovgivende magt [Distribution of Power: An Analysis of the Doctrine of Power Distribution with a Special Focus on the Legislative Power] (Copenhagen, Djøf, 2005). HP Olsen, ‘The Danish Supreme Court’s decision on the constitutionality of Denmark’s ratification of the Lisbon Treaty’ (2013) 50 CML Rev 1489. JP Olsen, Politisk organisering [Political Organisation] (Oslo, Universitetsforlaget, 1978). P Pagh-Rasmussen, ‘Præjudicielle forelæggelser og Juridisk Specialudvalg’ [‘Preliminary Rulings and the Judicial Committee’] (2004) Ugeskrift for Retsvæsen 305.

Denmark  121 P Pagh-Rasmussen, ‘Bopælspligt, menneskerettigheder og EU-ret: en kommentar til U 2007.99 H i lyset af nye EF-domme’ [‘Residence Obligation, Human Rights and EU Law: a Comment on U 2007.99 H in the Light of new EC Judgments’] (2007) Ugeskrift for Retsvæsen 126. H Rasmussen, Folkestyre, Grundlov og Højesteret: Grundlovens § 20 på prøve [Government, Constitution and Supreme Court: Section 20 of the Constitution on Test], (Copenhagen, Christian Ejlers, 1996). H Rasmussen, ‘Denmark’s Waning Constitutionalism and Article 20 of the Constitution on Transfer of Sovereignty’, in A Albi and J Ziller (eds), The European Constitutions and National Constitutions: Ratification and Beyond (Alphen aan den Rijn, Kluwer Law International, 2007) 149. A Ross, Dansk Statsforfatningsret Bind I [Danish Constitutional Law Vol I] (Copenhagen, Nyt Nordisk Forlag, 1980). A Ross, Dansk Statsforfatningsret [Danish Constitutional Law], ed O Espersen (Copenhagen, Nyt Nordisk Forlag, 1983). JE Rytter, Grundrettigheder, domstolens fortolkning og kontrol med lovgivningsmagten [Fundamental Rights, the Court’s Interpretation and the Control of Legislative Power] (Copenhagen, Thomson-Gad Jura, 2000). JE Rytter, Individets Grundlæggende Rettigheder [The Fundamental Rights of the Individual] (Copenhagen, Djøf, 2013). JE Rytter and M Wind, ‘In Need of Juristocracy? The Silence of Denmark in the Development of European Legal Norms’ (2011) 9 International Journal of Constitutional Law 470. JT Søndergaard, ‘Grundlovens forudsætning om demokrati’ [‘The Constitution’s Democratic Precondition’] (2001) Ugeskrift for Retsvæsen B 248. M Sørensen, ‘Responsum vedrørende de problemer, der knytter sig til grundlovens § 18’ [‘Opinion Regarding the Problems Related to Section 18 of the Constitution’], Rigsdagstidende 1952–53, Tillæg A [Rigsdag Gazette, Appendix A], 3709–38. M Sørensen, ‘Responsum fra januar 1971 til Folketingets Markedsudvalg’ (1971) Juristen 434. M Sørensen, Statsforfatningsret [Constitutional Law], ed P Germer (Copenhagen, Djøf, 1973). O Spiermann, ‘Hvad kommer efter tyve? En analyse af Højesteretsdommen i “Grundlovssagen”’ [‘What comes after twenty? An analysis of the Supreme Court ruling in the “Constitutional Case”’] (1998) Ugeskrift for Retsvæsen 325. JB Thayer, ‘The Origin and Scope of the American Doctrine of Constitutional Law’ (1893) 7 Harvard Law Review 129. JHH Weiler, ‘The Transformation of Europe’ (1991) 100 The Yale Law Journal 2403. JHH Weiler, The Constitution of Europe. ‘Do the New Clothes Have an Emperor?’ and Other Essays on European Integration (Cambridge, CUP, 1999). M Wendel, ‘Lisbon Before the Courts: Comparative Perspectives’ (2011) 7 European Constitutional Law Review 96. M Wind, ‘When Parliament Comes First – The Danish Concept of Democracy Meets the European Union’ (2009) 27 Nordic Journal of Human Rights 272. M Wind, ‘The Nordics, the EU and the Reluctance Towards Supranational Judicial Review’ (2010) 48 Journal of Common Market Studies 1039. M Wind, Who is Afraid of European Constitutionalism? The Nordic Distress with Judicial Review and Constitutional Democracy, iCourts Working Paper Series, No 12 (2014). M Wind, ‘Do Scandinavians Care About International Law? A Study of Scandinavian Judges’ Citation Practice to International Law and Courts’ (2016) 85 Nordic Journal of International Law 281. M Wind, ‘The Scandinavians: The Foot-dragging Supporters of European Law?’, in M Derlén and J Lindholm (eds), The Court of Justice of the European Union: Multidisciplinary Perspectives (Oxford, Hart, 2018) 191. M Wind, DS Martinsen and G Rotger, ‘The Uneven Legal Push for Europe: Questioning Variation when National Courts go to Europe’ (2009) 10 European Union Politics 63. H Zahle, ‘Suverænitets-overladelse: Tidspunktet for overladelsen’ [‘Transfer of Sovereignty: The Time of Transfer’] (1992) 74 Juristen 243. H Zahle, ‘Grundlovens menneskerettigheder: Sammenstødet mellem legalistiske og dynamiske retstraditioner’ [‘Human rights of the Constitution: The Clash Between Legalistic and Dynamic Legal

122  Marlene Wind Traditions’], in M Kjærum et al (eds), Grundloven og menneskerettigheder – i et dansk og europæisk perspektiv (Copenhagen, Djøf, 1997) 361. H Zahle, EU og den danske grundlov [The EU and the Danish Constitution] (Copenhagen, Christian Ejlers, 1998). H Zahle, Dansk Forfatningsret – Studieudgave [Danish Constitutional Law – Textbook] (Copenhagen, Christian Ejlers, 2006). R Zbíral, ‘Czech Constitutional Court, judgment of 31 January 2012, Pl. ÚS 5/12. A Legal revolution or negligible episode? Court of Justice decision proclaimed ultra vires’ (2012) 49 CML Rev 1475.

5 Germany MATTHIAS JESTAEDT*

I.  Germany’s Special Situation If one asks about the significance of European Union (EU) Member State constitutions for European integration, the German Constitution has a key role to play. Its special status is based on the concurrence and collaboration of a number of factors that essentially reflect two particularities: the particular shape that constitutional jurisprudence has taken through the Federal Constitutional Court (BVerfG) on the one hand, and the particular position that the BVerfG occupies amongst the constitutional courts of the EU Member States vis-à-vis the Court of Justice of the EU (CJEU) on the other. Without claiming exhaustiveness, the following factors can be mentioned in brief: • The Federal Republic of Germany is not only the EU’s most populous and most economically potent Member State, but is also one of the six founding members of the European Economic Community (EEC). As an original Member State, Germany – unlike the 22 acceded Member States – was never confronted with the situation of having to accept the extant ‘acquis communautaire’ without its own contribution. The law of the EEC/EC/EU developed in parallel and in interdependency with the national law of Germany. • In hardly any other Member State are the ‘juridification’, ‘jurisdictionalisation’ and constitutionalisation of political and social relations and conflicts likely to have been taken as far as in Germany. The BVerfG has shaped this conception of constitutionality and is its central actor. Its only rival in terms of its national and in particular its international significance and effect on others is the US Supreme Court. • The BVerfG interprets its task of preserving and actualising the Basic Law in a similarly selfconfident and creative way as the CJEU does with respect to Community and Union law. We cannot understand Germany’s constitutional order without looking at the BVerfG’s jurisprudence, just as we cannot do so for the Union’s legal order without the CJEU’s jurisprudence. • In matters of the protection of the Constitution by courts, particularly in matters of the protection of fundamental rights, the BVerfG simultaneously plays the role of avant-garde and paragon within the EU (and beyond). This is due to two factors: on the one hand, the far-reaching substantive-teleological conception of the Constitution, and, on the other hand, the fact that access to the BVerfG is regulated or at any rate interpreted and administered in * The manuscript was finished in September 2019; only minor changes have been made after this date.

124  Matthias Jestaedt a manner that allows (almost) any constitutional dispute to be subjected to a decision of the Constitutional Court. Put differently: on the one hand, (almost) any legal issue can be reformulated as an issue of constitutional law, and on the other hand, there is an action available that grants access to the BVerfG in (almost) any constellation in order to resolve issues of constitutional law. • Insofar as the relationship of national constitutionality in view of European integration is concerned, Member State (constitutional) courts look towards the BVerfG as well, whether affirmatively or critically. As a court within a constitutionally densely regulated federal state, the BVerfG draws on particular experience in dealing with vertically structured constitutionality. • The CJEU and BVerfG commenced their judicial activities at the same time, in the early 1950s. At the same pace and, thus, in a form of co-evolution, they became the outstanding judicial bodies for their respective orders that they are nowadays. • Among the Member States’ supreme courts the BVerfG is the most likely candidate to adjudicate on a par with the CJEU and to have its judgments read by the Luxembourg Court. Thus, amongst Member States’ courts, the BVerfG can afford best to challenge the self-confident decision-maker that is the CJEU. These factors show why the BVerfG’s jurisprudence represents a better yardstick for the relationship between European integration and the Basic Law than looking merely at the ‘integration clauses’ in constitutional texts – for at least two reasons. First, the BVerfG defines the terms of the constitutional-legal debate in Germany. Constitutional practices bypassing or going against its views do not stand a chance. Second, the Karlsruhe jurisprudence is a highly reliable indicator for the constitutionality vel non of Community or Union law influences. This suggests that aligning oneself with the BVerfG’s jurisprudence in depicting the relationship of the Basic Law to European integration is a good idea.1

1 The BVerfG jurisprudence on European integration has been extensively covered by German constitutional scholarship (Staatsrechtslehre). In the following, scholarly writings are quoted only sporadically, because of the vastly greater relative importance of the Constitutional Court’s jurisprudence and because the space available is limited. Without any claim to completeness, the following is a selection of monographs on the topic from 2010–15: Claudio Franzius, Vom Nationalstaat zum Mitgliedstaat und wieder zurück? Modifikationen ‘offener Staatlichkeit’ durch das Lissabon-Urteil des Bundesverfassungsgerichts, TranState Working Papers 124 (Bremen, 2010); Markus C Kerber, Der Kampf um den Lissabon-Vertrag (Stuttgart, Lucius & Lucius, 2010); Michael Abels, Das Bundesverfassungsgericht und die Integration Europas (Munich, AVM, 2011); Hans Hugo Klein, Europäische Integration und demokratische Legitimation (Baden-Baden, Nomos, 2011); Anna Katharina Mangold, Gemeinschaftsrecht und deutsches Recht. Die Europäisierung der deutschen Rechtsordnung in historisch-empirischer Sicht (Tübingen, Mohr Siebeck, 2011); Robert Ullerich, Rechtsstaat und Rechtsgemeinschaft im Europarecht (Baden-Baden, Nomos, 2011); Mattias Wendel, Permeabilität im europäischen Verfassungsrecht (Tübingen, Mohr Siebeck, 2011); Philipp Cramer, Artikel 146 Grundgesetz zwischen offener Staatlichkeit und Identitätsbewahrung (Berlin, Duncker & Humblot, 2013); Daniel Knop, Völker- und Europarechtsfreundlichkeit als Verfassungsgrundsätze (Tübingen, Mohr Siebeck, 2013); Rembert Graf Kerssenbrock, Die Vereinigten Staaten von Europa. Die Lissabon-Entscheidung und die Notwendigkeit, Volkssouveränität neu zu verstehen (Berlin, epubli, 2013); Stefan Städter, Noch Hüter der Verfassung? Das Bundesverfassungsgericht und die europäische Integration (Marburg, Metropolis, 2013); Matthias Klatt, Die praktische Konkordanz von Kompetenzen (Tübingen, Mohr Siebeck, 2014); Benjamin Maier, Grundrechtsschutz bei der Durchführung von Richtlinien (Baden-Baden, Nomos, 2014); Robert Christian van Ooyen, Die Staatstheorie des Bundesverfassungsgerichts und Europa. Von Solange über Maastricht zu Lissabon und Euro-Rettung, 5th edn (Baden-Baden, Nomos, 2014); Alexander Proelß, Bundesverfassungsgericht und überstaatliche Gerichtsbarkeit (Tübingen, Mohr Siebeck, 2014); Sebastian Recker, Subjektivierung der Staatsstruktur. Schutzmechanismus nationaler Identität in der Europäischen Union (Cologne/Munich, Heymann, 2014); Meike Schönemeyer, Die Pflicht des Bundesverfassungsgerichts zur Vorlage an den Gerichtshof der Europäischen Union gem. Art. 267 Abs. 3 AEUV (Berlin, Duncker & Humblot, 2014); Jean-Philippe Derosier, Les limites constitutionnelles à l’intégration européene. Étude comparée: Allemagne, France, Italie (Paris, LGDJ, 2015). Anthologies on the same subject since 2010: Andreas Fischer-Lescano, Christian Joerges and Arndt Wonka (eds), The German Constitutional Court’s Lisbon Ruling. Legal and Political-Science Perspectives (Bremen, ZERP, 2010); Armin Hatje, Jan Philipp Terhechte and Ulrich Everling (eds), Grundgesetz und

Germany  125

II.  Characteristics of the Constitutional System in Germany The German constitutional system is based on the Basic Law for the Federal Republic of Germany of 23 May 1949 (hereafter Basic Law or GG). It contains all constitutional provisions at the federal level. Further constitutional laws (Nebenverfassungen) do not exist, except for the constitutions of the 16 federal states (Bundesländer). However, they rank below federal law (see Article 31 GG). The Basic Law as a ‘fully-fledged constitution’ (Vollverfassung) contains both provisions on the organisation of the state as well as a catalogue of fundamental rights. The Basic Law commences by listing the fundamental rights (Articles 1–19 GG) that bind all public authority in a manner subject to full judicial review. europäische Integration. Die Europäische Union nach dem Lissabon-Urteil des Bundesverfassungsgerichts (Baden-Baden, Nomos, 2010); Peter M Huber (ed), Der Vertrag von Lissabon und das nationale Verfassungsrecht. Europäische Verwaltungsnetzwerke (Stuttgart, Boorberg, 2013); Roland Lhotta, Helmar Schöne and Jörn Ketelhut (eds), Das Lissabon-Urteil. Staat, Demokratie und europäische Integration im ‘verfassten politischen Primärraum’ (Wiesbaden, Springer VS, 2013); Claudio Franzius, Franz C Mayer and Jürgen Neyer (eds), Grenzen der europäischen Integration. Herausforderung für Recht und Politik (Baden-Baden, Nomos, 2014); Andreas Haratsch (ed), Nationale Verfassungen und Europarecht (Berlin, BWV, 2014). For 2015 alone the following papers can be mentioned (selection): Matthias Bäcker, ‘Das Grundgesetz als Implementationsgarant der Unionsgrundrechte’, (2015) 50 Europarecht 389; Klaus von Beyme, ‘Modell für neue Demokratien? Die Vorbildrolle des Bundesverfassungsgerichts’, in R van Ooyen and M Möllers (eds), Handbuch Bundesverfassungsgericht im politischen System, 2nd edn (Wiesbaden, Springer, 2015) 927; Christian Calliess, ‘Die europarechtliche Ultra-Vires-Kontrolle des Bundesverfassungsgerichts’, in C Calliess (ed), Herausforderungen an Staat und Verfassung: Völkerrecht – Europarecht – Menschenrechte. Liber amicorum für Torsten Stein zum 70. Geburtstag (Baden-Baden, Nomos, 2015) 446; Claudio Franzius, ‘Grundrechtsschutz in Europa’, (2015) 75 Zeitschrift für auländisches öffentliches Recht und Völkerrecht 383; Marcus Höreth, ‘Wenn Richter mitregieren wollen: Selbstautorisierung beim BVerfG und dem EuGH im Vergleich’, in R van Ooyen and M Möllers (eds), Handbuch Bundesverfassungsgericht im politischen System, 2nd edn (Wiesbaden, Springer, 2015) 875; Peter M Huber and Andreas L Paulus, ‘Cooperation of Constitutional Courts in Europe: The Openness of the German Constitution to International, European, and Comparative Constitutional Law’, in M Andenæs and D Fairgrieve (eds), Courts and Comparative Law, (Oxford, OUP, 2015) 281; Shu-Perng Hwang, ‘Der deutsche Verfassungsstaat im europäischen Mehrebenensystem: Überlegungen zur Auseinandersetzung zwischen den integrationsfreundlichen und -skeptischen Ansätzen’, (2015) 50 Europarecht 703; Albert Ingold, ‘Die verfassungsrechtliche Identität der Bundesrepublik Deutschland. Karriere – Konzept – Kritik’, (2015) 140 Archiv des öffentlichen Rechts 1; Roland Ismer and Dominika Wiesner, ‘Die OMT-Vorlage des Bundesverfassungsgerichts. Eine dogmatische Kritik auf Grundlage juristisch-ökonomischer Analyse’, (2015) 68 Die Öffentliche Verwaltung 81; Peter Graf Kielmansegg, ‘Das Bundesverfassungsgericht und Europa’, P Kielmansegg (ed), Wohin des Wegs, Europa? Beiträge zu einer überfälligen Debatte (Baden-Baden, Nomos, 2015) 114; Koen Lenaerts, ‘Kooperation und Spannung im Verhältnis von EuGH und nationalen Verfassungsgerichten’, (2015) 50 Europarecht 3; Oliver Lepsius, ‘Souveränität und Identität als Frage des Institutionen-Settings’, (2015) 63 Jahrbuch des öffentlichen Rechts der Gegenwart N.F. 62; Roland Lhotta and Jörn Ketelhut, ‘Bundesverfassungsgericht und europäische Integration’, in R van Ooyen and M Möllers (eds), Handbuch Bundesverfassungsgericht im politischen System, 2nd edn (Wiesbaden, Springer, 2015) 845; Eva Julia Lohse, ‘The German Constitutional Court and Preliminary References – Still a Match not Made in Heaven?’ (2014) 16 German Law Journal 1491; Antonio López Castillo, Über die Gemeinschaftsrechtsprechung des deutschen Bundesverfassungsgerichts: Feder oder Schwert der europäischen Integration?, in C Calliess (ed), Herausforderungen an Staat und Verfassung: Völkerrecht – Europarecht – Menschenrechte. Liber amicorum für Torsten Stein zum 70. Geburtstag (Baden-Baden, Nomos, 2015) 718; Markus Ludwigs, ‘Der Ultra-vires-Vorbehalt des BVerfG – Judikative Kompetenzanmaßung oder legitimes Korrektiv?’, (2015) 34 Neue Zeitschrift für Verwaltungsrecht 537; Johannes Masing, ‘Einheit und Vielfalt des Europäischen Grundrechtsschutzes’, (2015) 70 JuristenZeitung 477; Robert Christian van Ooyen, ‘Eine “europafeindliche” Kontinuität? Zum Politikverständnis der Lissabon-Entscheidung des Bundesverfassungsgerichts’, in R van Ooyen, Bundesverfassungsgericht und politische Theorie (Wiesbaden, Springer VS, 2015) 129; Heiko Sauer, ‘Doubtful it Stood …: Competence and Power in European Monetary and Constitutional Law in the Aftermath of the CJEU’s OMT Judgment’, (2015) 16 German Law Journal 971; Christoph Schönberger, ‘Identitäterä. Verfassungsidentität zwischen Widerstandsformel und Musealisierung des Grundgesetzes’, (2015) 63 Jahrbuch des öffentlichen Rechts der Gegenwart N.F. 41; Angela Schwerdtfeger, ‘Europäisches Unionsrecht in der Rechtsprechung des Bundesverfassungsgerichts – Grundrechts-, ultra-vires- und Identitätskontrolle im gewaltenteiligen Mehrebenensystem’, (2015) 50 Europarecht 290; Stefan Städter, ‘Das OMT-Verfahren in Luxemburg und Karlsruhe – ein wesentlicher Schritt der europäischen Krisenbewältigung?’, (2015) 51 Recht und Politik 20; Christian Walter and Markus Vordermayer, ‘Verfassungsidentität als Instrument richterlicher Selbstbeschränkung in transnationalen Integrationsprozessen. Vergleichende Überlegungen anhand der Rechtsprechung von EuGH und EGMR’, (2015) 63 Jahrbuch des öffentlichen Rechts der Gegenwart N.F. 129; Thomas Wischmeyer, ‘Nationale Identität und Verfassungsidentität. Schutzgehalte, Instrumente, Perspektiven’, (2015) 140 Archiv des öffentlichen Rechts 415.

126  Matthias Jestaedt The Basic Law defines the Federal Republic of Germany as a republic, as a parliamentary democracy (with a strong Federal Chancellor) and with a strong conception of the rule of law (Rechtsstaat), as a federal state and as a ‘social state’ (welfare state – Sozialstaat) (see Article  20 GG). Apart from the Federal Government that is presided over by the Federal Chancellor, the state is led at the federal level by the Parliament (Bundestag), which elects and dismisses the Chancellor. Together with the chamber of federal states (Bundesrat), the Bundestag constitutes the federal legislature (see Articles 76 et seq GG). In keeping with the anti-plebiscitary foundations of the Basic Law, the Federal President is not elected by popular vote, but by the Federal Assembly (Bundesversammlung). The President’s powers as head of state are essentially those of a representative and ‘state notary’ (Staatsnotar) function. The Rechtsstaat-related or horizontal separation of powers between legislature, executive and judiciary is complemented by a federative or vertical separation of powers between the Federation (Bund) and its component states (Länder) (which is further extended by municipal self-government as guaranteed in Article 28(2) GG). The Basic Law that was enacted in the three Western zones of occupation (American, British and French) with the approval of the military governors of the Western Allies saw itself as (West) Germany’s anti-totalitarian renewal after the catastrophe of National Socialism and as a contrast to the totalitarian experiment in the Soviet occupation zone, which was later to become the German Democratic Republic. The ‘inhuman regime [of National Socialism 1933–45], which brought immeasurable suffering, death and suppression to Europe and the world’, has ‘an antithetical significance characterising the identity of the constitutional system of the Federal Republic of Germany’.2 The most evident manifestation of this anti-totalitarian foundation in line with the tradition of the Western constitutional state is the fact that the Basic Law opens with the inviolability of human dignity (Article 1(1) sentence 1 GG). (West) Germany’s wish ‘to promote world peace as an equal partner in a united Europe’ (GG, Preamble) after the atrocities and legal nihilism of National Socialism was expressed in several provisions of the original (not amended) Basic Law of 1949, particularly in the clauses incorporating international law (Articles 25, 59(2)), in the prohibition of wars of aggression (Article 26) and in the authorisation to transfer sovereign powers to international organisations and to join a ‘system of mutual collective security’ (Article 24).3 With Germany’s reunification in 1990, the Basic Law, which had been enacted in 1949 as a makeshift solution for the period of partition, became the Constitution for all of Germany – but it had not, until then, been amended very often or extensively. The Basic Law has been altered 62  times since entering into force. More than three-quarters of these amendments concerned questions of the distribution of competences between the Federation and its federal states. The Basic Law was designed as an actively defensive (wehrhaft)4 Constitution, mainly as a reaction to the – real or alleged – structural weaknesses of the Weimar Constitution of 1919 that provided no effective protection against the opponents of a liberal-democratic constitutional order. The Constitution’s defensive stance is reflected in particular in the fact that the Basic Law 2 BVerfG 04.11.2009, 1 BvR 2150/08 (Wunsiedel) BVerfGE 124, 300 (328) (BVerfG translation MN 42) ECLI:DE: BVerfG:2009:rs20091104.1bvr215008. A selection of (full and partial) translations of recent BVerfG judgments into English is available at www.bundesverfassungsgericht.de/EN/Entscheidungen/Suche/suche_node.html. The Marginal Number (MN) is given as reference for exact quotes from English translations of the Court’s decisions when available. Where no translations by the BVerfG itself were available, the quoted decisions have been translated by the author. 3 See below, section IV.A. 4 The BVerfG rather misleadingly translates the German wehrhaft (which means self-protective and defiant) as ‘­militant’, see BVerfG 17.09.2013, 2 BvR 2436/10 (Beobachtung von Abgeordneten) BVerfGE 134, 141 (BVerfG translation MN 75, 112, 114, 117, 128) ECLI:DE:BVerfG:2013:rs20130917.2bvr243610.

Germany  127 (1) tries to ensure the stability and functionality of the central constitutional bodies,5 (2) holds subjective constitutional rights, first and foremost the fundamental rights to be ‘hard’, that is enforceable and, thus, judiciable rights, and (3) provides special mechanisms for the protection of the Constitution. In addition to the Constitutional Court, the latter include in particular the provisions governing the amendment of the Basic Law. Secondary constitutions and (successful) breaches, namely a hollowing-out of the Constitution (Verfassungsdurchbrechungen), as happened during the Weimar Republic, are precluded by the so-called incorporation requirement (Inkorporationsgebot). This requires that amendments to the Basic Law can only be effected by a law that modifies the wording of the Basic Law itself (Article  79(1) sentence 1 GG). In line with the Basic Law’s anti-plebiscitary foundations, laws amending the Constitution must be passed by the Bundestag and Bundesrat with a two-thirds majority (Article  79(2) GG). The liberal-democratic order as well as the basic principles of the federal organisation are not subject to constitutional revision – the so-called ‘eternity clause’ (Ewigkeitsgarantie) (see Article 79(3) GG). The Basic Law’s claim to supremacy as the highest domestic source of law is guaranteed by a powerful constitutional judiciary (unrivalled in diachronic as well as synchronic comparison): the BVerfG, based in Karlsruhe. It is divided into two co-equal chambers, the so-called senates (Senate) of eight judges each. It is based on the ‘Austrian’ model of a specialised, concentrated and isolated constitutional judiciary. The BVerfG can be tasked to decide almost any constitutional issue and decides authoritatively on any such issue. This is because it has jurisdiction to review the constitutionality of statutes and a monopoly to end the validity of federal and Land statutes that are younger than the Constitution (Normverwerfungsmonopol) (see particularly Articles  93(1) No 2, 100(1) GG), and because of its jurisdiction on individual constitutional complaints (Verfassungsbeschwerde) that may be filed by anybody claiming a violation of fundamental rights after the exhaustion of (ordinary) legal remedies (Article  93(1) No 4a GG). Furthermore, the BVerfG is superior to all other domestic courts as a result of its ability to overrule and annul their judgments in proceedings against an alleged violation of fundamental rights. These constitutional complaints against judicial decisions (Urteilsverfassungsbeschwerde) represent almost 95 per cent of all BVerfG proceedings, presently amounting to approximately 6,500 per year.6 Because it can, in principle, answer all constitutional questions authoritatively and in a ‘judicial’ manner, the BVerfG has become by far the most important actor for the protection, enforcement, interpretation and development of the Basic Law.

III.  Constitutional Culture in Germany Constitutional culture is a particular manifestation of the general legal culture. Germany is a good example of this, since the predominant German constitutional culture developed along two strands. The first one is the general, traditional legal culture in Germany. The second strand consists of the specific notions regarding the role and significance of the Constitution in post-war Germany. 5 Over more than 65 years there have only been 18 legislative periods, of which only three did not last the full fouryear term, namely the 6th Bundestag 1969–72, the 9th Bundestag 1980–83 and the 15th Bundestag 2002–05. The federal governments, always coalition governments, have been presided over by only eight Federal Chancellors: Konrad Adenauer (1949–63), Ludwig Erhard (1963–66), Kurt Georg Kiesinger (1966–69), Willy Brandt (1969–74), Helmut Schmidt (1974–82), Helmut Kohl (1982–98), Gerhard Schröder (1998–2005) and Angela Merkel (2005–2021). 6 See BVerfG, Jahresstatistik 2015 (‘2015 – Trendwende oder Ruhe vor dem Sturm?’), particularly at A.I.1., A.I.4. and A.I.5., www.bundesverfassungsgericht.de/DE/Verfahren/Jahresstatistiken/2015/statistik_2015_node.html.

128  Matthias Jestaedt

A.  The Law’s Role and Significance in General Compared to most other – European and non-European – commonwealths, the understanding of the law in Germany could best be characterised as ‘legalistic’ (which does not mean ‘­positivist’). A strict understanding that rules are rules is coupled with a high level of compliance in society and great confidence in the law as an effective social technique. This pronounced focus on the law is inter alia founded upon faith in a specifically juridical rationality that can be measured and applied with ‘scientific’7 means and precision by legal scholarship (Rechtswissenschaft).8 A ­symbiotic relationship of professionalised, largely judicial legal practice on the one hand and legal scholarship on the other hand is, thus, characteristic of the legal culture of German-speaking countries. The common format of communication between practice and scholarship shapes the central sub-discipline of legal scholarship, traditionally called Rechtsdogmatik, a virtually untranslatable word. Rechtsdogmatik conceives itself essentially as a ‘participating’ branch of scholarship. It is concerned with the description and analysis of law, but also wants to control legislation and the application of law, claiming that this is ‘scientific’ in character. Legal questions can allegedly be answered by using the (legal-)scientific method. The point is that this is produced by both legal scholarship and by (higher and supreme) courts. This not only brings legal scholarship and the courts (and their output) close together in terms of subject matter, methods, socialisation and personnel, but also (among other things) causes judicial decisions – at least by higher and supreme courts – quite often to be written in a discursive, almost legal-scholarly style.9

B.  The Constitution’s Role and Significance in Particular With the creation of the Basic Law, the traditional understanding of the law, which can be traced all the way back to the German Historical School, was confronted with a heretofore unknown role for and significance of the Constitution. Not only does this Constitution ‘penetrate’ the whole legal order and – with the aid of its enforcement by the Constitutional Court – orient it towards its own value-judgments as the highest domestic source of law, but very soon it also became the central point of reference for collective identity in the political sphere. During Germany’s partition, for (West) Germany, the Basic Law largely became an Ersatz for the nation, a concept that was considered thoroughly toxic – at least during the first decades after World War II – because of the racist and totalitarian National Socialism. This is a substitute object for patriotism which in other countries focuses on the nation. The term ‘constitutional patriotism’ (Verfassungspatriotismus)10 has been coined for this particular form.11 As opposed to the Weimar Republic era, criticism vis-à-vis the Basic Law or its ordering of the state in the Bonn (and later Berlin) republic that would need to be taken seriously has not been voiced – neither in the realm of politics nor even

7 The German language does not distinguish between science (as natural science) and other branches of scholarship. All are equally denoted as Wissenschaft. Where the commonality between legal scholarship and science is to be highlighted, this chapter will use ‘science’; otherwise it will use ‘scholarship’. 8 Matthias Jestaedt, ‘Wissenschaft im Recht. Rechtsdogmatik im Wissenschaftsvergleich’, (2014) 69 JuristenZeitung 1. 9 See Christoph Schönberger, ‘Höchstrichterliche Rechtsfindung und Auslegung gerichtlicher Entscheidungen’, in W Höfling (ed), Grundsatzfragen der Rechtsetzung und Rechtsfindung, Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer 71 (Berlin, De Gruyter, 2012) 296. 10 Moulding the discourse: Dolf Sternberger, ‘Verfassungspatriotismus’ (1979), in P Haungs (ed), Dolf Sternberger. Schriften, Vol X (Frankfurt/M, Insel, 1990) 3. 11 Concisely: Josef Isensee, ‘Die Verfassung als Vaterland. Zur Staatsverdrängung der Deutschen’, in A Mohler (ed), Wirklichkeit als Tabu. Anmerkungen zur Lage (Munich, Oldenbourg, 1986) 11.

Germany  129 in the realm of scholarship on constitutional law. Political and legal disputes take place on the common ground of the Basic Law. They concern its interpretation, however, not a fight against the present Constitution. Under the Basic Law, the young Constitution and the traditional Dogmatik have entered into a completely novel, previously unknown alliance.12 The BVerfG as the catalyst of the Basic Law’s constitutional Dogmatik soon assumed a prominent place in legal scholarship as well. Already at an early stage under the new Basic Law, the BVerfG was attributed a hybrid position. It is both a court, that is, an institution of the judicative, and simultaneously a constitutional body, that is, an institution of government. Thus, the ‘guardian of the constitution’ (Hüter der Verfassung) is located at the border between law and politics in a peculiar way. Conscious of the Basic Law’s outstanding role in and for politics, the BVerfG is an actor which can transform (power-)political questions into (constitutional-)legal questions which can be decided using scientific method, namely Dogmatik. This means that nearly all controversial political questions end up at the BVerfG sooner or later and are decided authoritatively, by assigning them a place within the Constitutional Court’s sophisticated constitutional doctrinal matrix. Until now, the authority of Karlsruhe’s decisions has never earnestly and permanently been cast into doubt by influential political actors: Karlsruhe locuta, causa finita (the Court has spoken, the case is closed). The BVerfG benefits in this regard from the traditionally great confidence in legal discursive rationalities in the shape of legal doctrinal arguments in Germany. The style in which Karlsruhe argues and justifies its decisions is ‘systematicist’ and ‘extensive’, typical for Rechtsdogmatik, not infrequently far exceeding what is strictly necessary to decide the case in point. Yet this mechanism not only ‘juridifies’ politics, ie shapes political discourse, but also moulds statutes to the Constitution, ie constitutionalises them. Thus, the three decisive phenomena for German constitutional culture originate from the same source. These phenomena are (1) the ‘juridification’ of politics; (2) the constitutionalisation of the entire legal order – in particular by means of the ‘interpretation in conformity with the Constitution’ (verfassungskonforme Auslegung) and the ‘radiating effect of the Constitution’ (Ausstrahlungswirkung der Verfassung); and (3) the ‘jurisdictionalisation’ of constitutional law, ie the possibility to have virtually all constitutional questions decided by a court, the BVerfG. Moreover, all three phenomena give expression to and are a result of a greater confidence in Germany in the Rechtsstaat (and thus in the courts) than in democracy (and thus in Parliament).

IV.  The Basic Law’s Openness vis-à-vis European Integration A.  The Basic Law’s Global Openness The Basic Law’s accomplishment in transforming a totalitarian regime into a liberal-democratic ‘constitutional state’ (Verfassungsstaat) goes hand-in-hand with a pronounced openness of the domestic legal order vis-à-vis global legal orders, that is, to both international and supranational law.13 The Basic Law prepares the way for this openness by acknowledging the role of ‘­inviolable and inalienable human rights as the basis of every community, of peace and of justice 12 See Matthias Jestaedt, ‘Die deutsche Staatsrechtslehre im europäischen Rechtswissenschaftsdiskurs – Kennzeichen, Kernkompetenzen, Rezeptionshindernisse’, 67 (2012) JuristenZeitung 1. 13 For a representative view, see Christian Hillgruber, ‘Der Nationalstaat in der überstaatlichen Verflechtung’, in J Isensee and P Kirchhof (eds), Handbuch des Staatsrechts der Bundesrepublik Deutschland, Vol II, 3rd edn (Heidelberg, Müller, 2004) § 32 MN 75–6 f, 113–4 f.

130  Matthias Jestaedt in the world’ (Article 1(2) GG) and through the constitutional prohibition of wars of aggression (Article 26 GG). It implements this openness in two ways. The first is that the Federal Republic of Germany ‘may enter into a system of mutual collective security’ like the UN or NATO ‘[w]ith a view to maintaining peace’ (Article 24(2) GG) and that it may transfer sovereign rights to international organisations (Article 24(1) GG). The second involves transforming international law into domestic law and ensuring its domestic enforcement. The domestic incorporation of international law can be achieved in two ways: either in a dynamic way by enabling incorporation directly through the constitution, or in a static way by transforming international law by means of a transformation statute (and mediated by the constitution). The Basic Law took the first route with regard to what it calls ‘general rules of international law’. These rules are proclaimed as an integral part of federal law (Article 25 sentence 1 GG) and are granted a rank below the Constitution, but above statutes (Article 25 sentence 2 GG). The second route was taken for international treaty law (Article 59(2) GG). For treaty law, the Constitution requires transformation into domestic law by way of an Approving Act (Zustimmungsgesetz). This statute constitutes its domestic basis of validity and confers on the treaty provisions the domestic rank of the legislation giving effect to them. With regard to the European Convention on Human Rights (ECHR) (as interpreted by the European Court of Human Rights (ECtHR)), the BVerfG has developed a special regime that is not evident if one merely reads the text of the Constitution.14 Under Article 59(2) GG, the ECHR has the rank of a simple federal statute.15 According to the lex-posterior maxim, it is therefore in principle subject to derogation by later (federal) statutes. This, however, would render the Federal Republic of Germany responsible for a breach of international law, which in turn could be sanctioned by the ECtHR. In order to avoid this, the BVerfG’s regime has the ECHR partake in substance in quasi-constitutional rank and attributes ‘constitutional-legal significance’ (verfassungsrechtliche Bedeutung) to it.16 Based on the principle of the Basic Law’s ‘openness to public international law’ (Völkerrechtsfreundlichkeit) which the court deduces from a contextual reading of constitutional provisions concerning international law (Preamble, Articles  1(2), 23–26, 32, 59(2) GG),17 it has developed three strategies over time. The first strategy is to require taking note of the content and development status of the ECHR when interpreting the Basic Law’s fundamental rights, as long as this does not lead to a limitation or reduction of the level of fundamental rights protection. The ECHR’s guarantees as construed by the ECtHR are an ‘interpreting aid in determining the content and scope of fundamental rights and fundamental constitutional principles of the Basic Law’,18 which one could call the elevation of the ECHR’s ‘interpretative rank’. The second strategy draws upon the principle that all executive and judicial power is bound by ‘statute and law’ (Gesetz und Recht), according to Article 20(3) GG. This principle is taken to include the obligation to take the ECHR and ECtHR judgments into consideration when 14 On this see among many others, Angelika Nußberger, ‘Europäische Menschenrechtskonvention’, in J Isensee and P Kirchhof (eds), Handbuch des Staatsrechts der Bundesrepublik Deutschland, Vol X, 3rd edn (Heidelberg, Müller, 2012) § 209. 15 This is established jurisprudence, see only BVerfG 14.10.2004, 2 BvR 1481/04 (Görgülü) BVerfGE 111, 307 (316–7 f) ECLI:DE:BVerfG:2004:rs20041014.2bvr148104. 16 BVerfG 04.05.2011, 2 BvR 2365/09 (Sicherungsverwahrung II) BVerfGE 128, 326 (367) ECLI:DE:BVerfG:2011: rs20110504.2bvr236509 (translation without MN). 17 See, eg, BVerfG 2 BvR 1481/04 (Görgülü) (n 15) 317–9; BVerfG 2 BvR 2365/09 (Sicherungsverwahrung II) (n 16) 367 f; in extenso BVerfG 15.12.2015, 2 BvL 1/12 (Treaty Override) ECLI:DE:BVerfG:2015:ls20151215.2bvl000112, MN 64–76. This line of case law has been confirmed by BVerfG 12.06.2018, 2 BvR 1738/12, 1395/13, 1068/14, and 646/15 (Beamtenstreikverbot) ECLI:DE:BVerfG:2018:rs20180612.2bvr173812, MN 126–35. 18 BVerfG 26.03.1987, 2 BvR 589/79 (Unschuldsvermutung) BVerfGE 74, 358 (370), since then established jurisprudence.

Germany  131 interpreting statutes, as long as this is consistent with the accepted methods of interpretation.19 ‘A complainant may [then] challenge the disregard of this duty of consideration as a violation of the fundamental right whose area of protection is affected in conjunction with the principle of the rule of law’ (Rechtsstaatsprinzip).20 The third and newest strategy can be read as a reinforcement of the first two strategies. In its second judgment on preventive detention of 2011, the Second Senate affirms that the Constitution itself requires an ECHR-compliant interpretation of the Basic Law’s fundamental rights – not schematically, but taking into account the recognised ‘methods of constitutional interpretation as well as the requirements of the Basic Law’.21 In this respect, the grounds of the judgment are based centrally on the commitment of the German people to ‘inviolable and inalienable human rights’ as expressed in Article 1(2) GG. The Court itself speaks rather cautiously about an ‘interpretation that is open to international law’ (völkerrechtsfreundliche Auslegung) or about an ‘interpretation in a manner open to the Convention’ (konventionsfreundlichen Auslegung) (of the fundamental rights) of the Basic Law.22 It is reasonable to assume that this extensive attribution of ‘constitutional significance’ to the ECHR and ECtHR judgments, based on the Basic Law’s openness to public international law, is probably inspired by the ECHR system of adjudication. For this can lead to the Strasbourg Court reviewing even the BVerfG’s decisions on their conformity with the ECHR (and possibly challenging them).23 The BVerfG has held that Article 59(2) GG assigns other treaty obligations the rank of their federal transformation­ statutes and that they are subject to the lex posterior maxim. In particular, the unconstitutionality of statutes violating international law cannot be found in the unwritten principle of the Basic Law’s openness to international law. Although this principle possesses constitutional rank, the BVerfG has held that this principle does not mean that there is a constitutional obligation to unreservedly comply with all norms of international law.24

B.  Particularities of the Basic Law’s Openness vis-à-vis the EU 1.  The Constitutional Situation until the Treaty of Maastricht Programmatically, the preamble of the Basic Law’s 1949 original version states that the German people adopted this Constitution ‘[i]nspired by the determination to promote world peace as an equal partner in a united Europe’. Even though this expresses a central – perhaps the most important – trend-setting decision of post-war Germany, the original version of the Basic Law did not yet contain specific and exclusive provisions with regard to European integration. At the time, European integration was only vaguely discernible and would lead, via the economy-focused EEC/EC, towards a more generally political EU. Furthermore, the concept of a supranational legal community had not yet been born at that time. 19 BVerfG 2 BvR 1481/04 (Görgülü) (n 15) 315 et seq particularly 323–4 f (BVerfG translation MN 30 et seq esp 47–8 f). 20 Ibid, 328–9 (BVerfG translation MN 60–1 f, quote at 30). 21 The BVerfG talks about a ‘results-oriented’ ECHR consultation as an interpretation aid: 2 BvR 2365/09 (Sicherungsverwahrung II) (n 16) 370. 22 Ibid, 366, 370–1. 23 See also the remark on avoiding condemnations of the Federal Republic of Germany by the ECtHR: BVerfG 2 BvR 2365/09 (Sicherungsverwahrung II) (n 16) 369. Just 18 months earlier (and ironically with respect to preventive ­detention), the ECtHR had held that Germany had violated Art 5(1) and Art 7(1) ECHR (M./Deutschland, Judgment of 17 December 2009, No 19359/04) – in the last instance through a BVerfG decision itself: Sicherungsverwahrung I [2004] BVerfGE 109, 133 recte. 24 BVerfG 2 BvL 1/12 (Treaty Override) (n 17) Headnotes 2–4, MN 33–76.

132  Matthias Jestaedt Special provisions relating to Europe were first integrated into the Basic Law with the 1954 (4th) amendment.25 With it, the requirement in Article  79(1) sentence 1 GG that constitutional amendments must alter the Basic Law’s text itself (Textänderungsgebot) was modified for certain constellations of public international law, to the effect that a clarification in the Basic Law suffices. This allowed in particular for the ratification of the treaty on a European Defence Community (EDC) that had been negotiated. With the same constitutional amendment, a corresponding clarification was incorporated in Article  142a GG. Due to the rejection of the EDC by France’s Assemblée nationale, Article  142a GG never became effective and was abolished in 1968. Nevertheless, Article 79(1) sentence 2 GG, to date inoperable, is still part of the valid constitutional law. The ‘European Communities’ entered the Basic Law for the first time in 1970, in a rather unspectacular manner. Within the framework of the 1969 Financial Reform Act, the constitutional legislator supplemented the Federation’s competence for levies and charges ‘imposed within the framework of the European Communities’ (Articles 106(1) No 7 and 108(1) sentence 1 GG). Apart from this, until after Germany’s reunification, the constitutional viewpoint remained that European integration was a case (albeit in various respects a specific one) of those provisions that effectuated Germany’s openness towards international cooperation (see above, section IV.A). It was above all the authorisation in Article 24(1) GG – unaltered to this very day – that permitted the transfer of sovereign powers to international organisations by the Federal Republic. The BVerfG has declared the development of EC law (or its conception by the increasingly active ECJ) constitutional on the legal basis of Article 24(1) GG time and again since the second half of the 1960s. Sometimes explicitly and sometimes implicitly, the BVerfG approved the features and effects of EC law that bore the marks of supranationality (as a special case of internationality). A (first) climax was reached with the two Solange decisions from 1974 and 1986. These decisions, on the one hand, confirmed the Basic Law’s openness to integration and the possibility for an international layer of law to claim primacy before and direct effect within the domestic law, while, on the other hand, making the reservation that EC law must not bring about an incursion into the Basic Law’s fundamental structures.26

2.  The Constitutional Situation Since the Treaty of Maastricht Germany’s 1990 reunification and the 1992 Maastricht Union Treaty triggered more extensive constitutional amendments concerning Europe. These were partly directed at adjusting the constitutional text that no longer adequately reflected the constitutional situation (Nachführung), and partly at constitutionally hedging the new level of European integration reached by the Treaty of Maastricht.27 The key provision was the newly included Article 23 GG. The original Article 23, dating from 1949, had regulated the accession of further (Eastern) German Federal States to the Basic Law (so-called accession article). After the reunification, it was considered satisfied and abrogated in 1990,28 leaving, as it were, space for a separate ‘Europe’ Article.

25 (4.) ‘Gesetz zur Ergänzung des Grundgesetzes’ of 26 March 1954, BGBl I 45. 26 See BVerfG 07.06.2000, 2 BvL 1/97 (Bananenmarktordnung) BVerfGE 102, 147, 163 f ECLI:DE:BVerfG:200 0:ls20000607.2bvl000197. On the ‘Solange’ criteria see below, section V.C.3. 27 For the view that the Treaty of Maastricht ‘finds an adequate constitutional basis’ in Art 23(1) GG, see, eg, BVerfG 31.03.1998, 2 BvR 1877/97 (Euro) BVerfGE 97, 350, 369 (BVerfG translation MN 95) ECLI:DE:BVerfG:1998:rs19980331. 2bvr187797. 28 Through the (36.) constitutional amendment, ‘Gesetz zu dem Vertrag vom 31. August 1990 zwischen der Bundesrepublik Deutschland und der Deutschen Demokratischen Republik über die Herstellung der Einheit Deutschlands – Einigungsvertragsgesetz – und zu der Vereinbarung vom 18. September’, of 23 September 1990, BGBl II 85.

Germany  133 The Joint Constitutional Commission of Bundestag and Bundesrat (GVK)29 that had been formed in 199230 after the reunification introduced31 Article  23(1) GG. This was a specific integration clause with a threefold regulatory content. Sentence 1 protects fundamental structures (Struktursicherungsklausel), copying the BVerfG’s Solange criteria,32 and determines that Germany participates ‘in the development of the European Union that is committed to democratic, social and federal principles, to the rule of law, and to the principle of subsidiarity, and that guarantees a level of protection of basic rights essentially comparable to that afforded by this Basic Law … [w]ith a view to establishing a united Europe’. Sentence 2 affirms the authorisation contained in Article 24(1) GG to transfer sovereign rights. In continuation of the BVerfG’s jurisprudence relating to Article 24(1), this can be seen as the basis for legislation giving effect to supranational law in the German domestic order. Finally, sentence 3 governs and channels the further development of European integration, functioning as an identity-protection clause (Identitätssicherungsklausel): ‘The establishment of the European Union, as well as changes in its treaty foundations and comparable regulations that amend or supplement this Basic Law, or make such amendments or supplements possible, shall be subject to paragraphs (2) and (3) of Article 79.’ Far-reaching modifications of the EU integration programme that affect Germany’s constitutional situation are treated like constitutional revisions (with the exception of the requirement of textual alteration, Article 79(1) sentence 1 GG). First, the federal legislature must approve them by enacting the necessary domestic legislation to give effect to it. That legislation must be in the form of a statute fulfilling the quora for constitutional amendments according to Article 79(2) GG (two-thirds majorities in Bundestag and Bundesrat). Second, the modification must comply with the substantive limits of any constitutional revision as outlined by the ‘eternity clause’ (Ewigkeitsgarantie) of Article 79(3) GG. The further six (since 2009: seven) paragraphs of Article 23 GG respond to the fact that a transfer of sovereign powers to the European level without domestic adaptations leads to a shift in influence at the expense of the legislative powers and for the benefit of the Government. With Article  23(2–7, or 1a–7 now), the position of the Bundestag and Bundesrat in EU matters is strengthened. Paragraph 2 states the right (and obligation) of the Bundestag and Bundesrat to participate as well as the Federal Government’s obligation to a ‘comprehensive’ and ‘earliest possible’ information of the Federation’s legislative bodies in general terms. The following paragraphs specify the Bundestag’s (3) and the Bundesrat’s (4–7) rights to co-operation in a very detailed manner. Statutes (Ausführungsgesetze) regulating ‘details’ (Article  23(3) sentence 3, (7)) have been passed with regard to both the Bundestag’s33 and the Bundesrat’s34 participation. Finally, when the Treaty of Lisbon entered into force, paragraph 1a was introduced and now allows the Bundestag and Bundesrat to bring a subsidiarity action before the CJEU. The remaining 1992 constitutional amendments prompted by the GVK are related to: • implementing the right to vote for Union citizens in local government elections (ie, to county and municipal councils) by Article 28(1) sentence 3 GG;35 29 See Bericht der Gemeinsamen Verfassungskommission, BTDrucks 12/6000 of 5 November 1993. 30 Through the (38.) ‘Gesetz zur Änderung des Grundgesetzes’ of 21 December 1992, BGBl I 2086. 31 To be precise, it should be added that although the GVK has discussed and shaped the constitutional amendments mentioned below, they have been implemented in the regular procedure of constitutional amendment under Art 79 GG. 32 See in detail below, section V.C.3. 33 Act on Co-operation between the Federal Government and the German Federal Parliament in Matters of the European Union (EUZBBG) of 4 July 2013, BGBl 2013 I 2170. 34 ‘Act on Co-operation of the Federation and the Federal States in Matters of the European Union’ (EUZBLG) of 19 March 1993, BGBl 1993 I 313. 35 This is the constitutional legislator’s reaction to the 1990 BVerfG decision discarding the statutory introduction of a right to vote in municipal elections for (all) foreigners as unconstitutional: BVerfG 31.10.1990, 2 BvF 2, 6/89

134  Matthias Jestaedt • forming a Europe Committee of the Bundestag vested with special powers by Article 45 GG and a corresponding Europe Chamber of the Bundesrat by Article 52(3a) GG; • expanding the Bundesrat’s functions to include its participation in EU matters by Article 50 GG; • authorising the transfer of functions of the Federal Bank (Bundesbank) to the European Central Bank within the framework of the EU by Article  88 sentence 2 GG,36 which had become necessary for the realisation and constitutional hedging of the monetary Union laid out in the Treaty of Maastricht;37 • clarifying the powers of the Joint Committee of the Bundestag and Bundesrat that is under specific conditions responsible for the legislation during the state of defence on the basis of Article 115e(2) GG. Within the framework of the 1993 asylum amendment,38 also spurred by the GVK, the prohibition on extraditing Germans was relaxed with regard to EU Member States by Article 16(2) sentence 2 GG on the one hand, and the German (fundamental) right to asylum was placed into the frame of the evolving EU asylum system by Article 16a(2) and (5) GG on the other hand. The bulk of the amendments initiated by the GVK was decided upon in 199439 and also included the handling of draft legislation on transferring sovereign rights to the EU (Article 76(2) sentence 5 and (3) sentence 5). The 2006 First Reform of Federalism (Föderalismusreform I)40 addressed Germany’s liabilities originating from EU law and placed their distribution between Federation and federal states on an explicit constitutional footing. Article 104a(6) GG and Article 109(2) and (5) sentence 1 impose budgetary duties on both levels in case of EU obligations or sanction measures. The latest relevant constitutional amendment (entered into force 2009) is the authorisation contained in Article 87d(1) sentence 2 GG to have air navigation services provided by foreign air navigation service providers ‘which are authorised in accordance with European Community law’. Until the end of 1992, the Basic Law did not explicitly acknowledge European i­ntegration – with the exception of two references in the financial constitution known only to specialists (Articles  106(1) No 7, 108(1) sentence 1 GG). By contrast, 15 constitutional provisions today contain an explicit reference to the European Communities or the European Union.41

3.  The Weak Indicative Effect of the Constitutional Text The multitude of explicit references to the European integration process indicates that the Basic Law and EU law are standing in a veritable regulatory network. In that framework the Basic Law processed the EU law impacts in various ways and it may be classified as a Europeanised (Kommunalwahlrecht für Ausländer in Schleswig-Holstein) BVerfGE 83, 37; BVerfG 31.10.1990, 2 BvF 3/89 (Ausländerwahlrecht zu hamburgischen Bezirksversammlungen) BVerfGE 83, 60. 36 For the significance of Article 88 sentence 2 GG, see BVerfG 2 BvR 1877/97 (Euro) (n 27) 372. 37 Ibid: ‘When the Act amending the Basic Law of 21 December 1992 (BGBl I 2086) laid an explicit constitutional foundation for European integration in Art. 23 Basic Law, the constitutional amending legislator was faced, as the next step in integration, with the decision for the Monetary Union (BTDrucks 12/3338, 5).’ 38 (39.) ‘Gesetz zur Änderung des Grundgesetzes (Artikel 16 und 18)’ of 28 June 1993, BGBl I 1002. 39 (42.) ‘Gesetz zur Änderung des Grundgesetzes (Art. 3, 20a, 28, 29, 72, 74, 75, 76, 77, 80, 87, 93, 118a und 125a)’ of 27 October 1994, BGBl I 3146. 40 (52.) ‘Gesetz zur Änderung des Grundgesetzes (Artikel 22, 23, 33, 52, 72, 73, 74, 74a, 75, 84, 85, 87c, 91a, 91b, 93, 98, 104a, 104b, 105, 107, 109, 125a, 125b, 125c, 143c)’ of 28 August 2006, BGBl I 2034. 41 Art. 16(2) sentence 2, 16a(2) and (5), 23(1-7), 28(1) sentence 3, 45, 50, 52(3a), 76(2) and (3), 87d(1), 88 sentence 2, 104a(6), 106(1), 108(1), 109(2) and (5), 115e(2) GG.

Germany  135 constitution of a nation state. However, the textual finding is not a reliable indicator as to which questions domestic (German) constitutional and supranational law chime together and as to which tensions prevail. It does not even reliably indicate where to look for the constitutional centrepiece in the matter of European integration. What is more, central issues such as the primacy of application, the direct effect or the manner of EU law implementation are not even addressed in the Basic Law. This, along with the virtue of law-abidingness that is esteemed highly in Germany, may be a reason for the fact that no general, structural or permanent obstacles to EU law enforcement can be detected in the legislative, administrative, and particularly judicial everyday legal life. However, to conclude from the relative smoothness of the day-to-day application of the law that the encounter of German constitutional law and EC or EU law respectively is free from tensions would go amiss. In Germany, problems and tensions with EC or EU law arise in the borderline areas of law42 – concerning the ‘standard of fundamental rights’ (Grundrechtsstandard)43 and the ‘last instance’ (letzte Wort),44 concerning the question of principle for the ‘primary political area’ (politischen Primärraum),45 the ‘masters of the Treaties’ (Herren der Verträge)46 and the existential question of the ‘(constitutional) identity’ ((Verfassungs-)Identität).47 Solutions to these problems, however, are difficult to find by simply consulting the constitutional text. In some cases, they are even difficult to develop from it. At any rate, they relate to the integration clause of Article 23(1) GG. He who asks about the constitutional situation in Germany on this point will find reliable answers in the decisions of the BVerfG only – and will find plenty there.

V.  European Integration and the Federal Constitutional Court A.  Co-evolution of the CJEU’s and the BVerfG’s Jurisprudence It is probably no exaggeration to attest to the BVerfG a special role amongst those EU Member State courts that are specifically entrusted with the protection of the national constitutions. This is because a number of EU Member States do not follow the Austrian model of an indepen­dent constitutional jurisprudence. Also, with regard to the position, function, significance and selfperception, the BVerfG is peerless – not only – in the EU. Among the founding EC Member States, for instance, only Italy possesses an at least roughly comparable institution in the form of its Corte Constituzionale.48 Among the other members, the oldest constitutional court still 42 The BVerfG’s president speaks about the Court thinking ‘very often from a “worst case scenario”’ as the point of departure, Andreas Voßkuhle, ‘Integration durch Recht – Der Beitrag des Bundesverfassungsgerichts’, (2016) 71 JuristenZeitung 161, 167. 43 BVerfG 22.10.1986, 2 BvR 197/83 (Solange II) BVerfGE 73, 339, 378 ECLI:DE:BVerfG:1986:rs19861022.2bvr019783; BVerfG 12.10.1993, 2 BvR 2134 (Maastricht) BVerfGE 89, 155, 175 ECLI:DE:BVerfG:1993:rs19931012.2bvr213492. 44 BVerfG 30.06.2009, 2 BvE 2/08 (Lissabon) BVerfGE 123, 267, 400 (BVerfG translation MN 340) ECLI:DE:BVerfG: 2009:es20090630.2bve000208. 45 Ibid, 382 (BVerfG translation MN 301). 46 See BVerfG 08.04.1987, 2 BvR 687/85 (Kloppenburg) BVerfGE 75, 223, 242 ECLI:DE:BVerfG:1987:rs19870408.2 bvr068785; BVerfG 2 BvR 2134 (Maastricht) (n 43) 190; BVerfG 2 BvE 2/08 (Lissabon) (n 44) 348–9, 381 et seq; BVerfG 06.07.2010, 2 BvR 2661/06 (Honeywell) BVerfGE 126, 286, 302–3 f ECLI:DE:BVerfG:2010:rs20100706.2bvr266106; BVerfG 14.01.2014, 2 BvR 2728/13 (OMT (Vorlage an EuGH)) BVerfGE 134, 366, MN 26 ECLI:DE:BVerfG:2014:rs20140114. 2bvr272813; BVerfG 15.12.2015, 2 BvR 2735/14 (Europäischer Haftbefehl II) ECLI:DE:BVerfG:2015:rs20151215. 2bvr273514 MN 44. 47 Pars pro toto: BVerfG 2 BvE 2/08 (Lissabon) (n 44) 268 (Headnote 4) – since then established jurisprudence. 48 The French Conseil constitutionnel was not founded until the 1958 5th Republic constitution; only over the course of decades has it assumed a role similar to those that had been characteristic for the BVerfG and the Corte Constituzionale

136  Matthias Jestaedt existing in Europe, the Austrian Constitutional Tribunal (VfGH), stands out. Nevertheless, even the VfGH, like the constitutional courts of the younger EU Member States, is more or less oriented towards the BVerfG in exercising the jurisdiction it is entrusted with.49 When it comes to the significance of EU Member States’ constitutions vis-à-vis European integration and assessing the BVerfG’s integration-related jurisprudence, another fact must be taken into account. It is not only that the BVerfG and the ECJ/CJEU originate in the same era of an aggravating Cold War and that they simultaneously – the BVerfG in 1951 and the ECJ in 195250 – started their jurisdictional activity. A number of striking parallels may be observed with respect to the increase in importance taking place for both judicial bodies from an evolutionary perspective. The more or less synchronic ascent of the BVerfG and the ECJ took place in a temporally parallel development of juxtaposition and mutual dialogue, which may be characterised at times as a joint venture or as a struggle against each other. Spectacular examples of a ‘dialogue’ between the Luxembourg and the Karlsruhe judges have for decades been questions of fundamental rights protection. Without the Solange I (1974)51 and the Solange II decisions (1986),52 the unfolding of the fundamental rights protection in EC law from the early 1970s onwards cannot be recounted in full. With the Second Senate’s Honeywell order (2010),53 the BVerfG took a stand on the ECJ’s decision in the Mangold case (2005).54 The Court of Justice’s way to its decision in Digital Rights Ireland (2014) had been paved by the BVerfG’s decision in Vorratsdatenspeicherung (2011).55 The same-date decisions in the Åkerberg Fransson (2013)56 and Melloni (2013)57 cases present the background for Karlsruhe’s reactions in the decisions in Antiterrordatei (First Senate, 2013)58 and Europäischer Haftbefehl II (Second Senate, 2015).59 With its reference for a preliminary ruling in the OMT case (2014),60 the BVerfG requested the CJEU’s ruling on the Gauweiler61 case directly, and shared the verdict of the CJEU, not without stressing the conditions of legality the CJEU had formulated.62 The BVerfG (Second from the start; for more detail Olivier Jouanjan, ‘Verfassungsrechtsprechung in Frankreich’, in A  v  Bogdandy, C Grabenwarter and P Huber (eds), Handbuch Ius Publicum Europaeum, Vol VI (Heidelberg, Müller, 2016) § 99 MN 22 et seq, particularly 22–25 (‘Geschichte eines Verfassungswunders’); and the report on France in this volume. The Belgian Cour d’arbitrage, that was designated as ‘Grondwettelijk Hof ’/’Cour constitutionnelle’/‘Verfassungsgerichtshof ’ in 2007, has only been founded in 1980/1983 and did not adjudicate until 1985; see Christian Behrendt, ‘Der belgische Verfassungsgerichtshof ’, in A v Bogdandy, C Grabenwarter and P Huber (eds), Handbuch Ius Publicum Europaeum, Vol VI (Heidelberg, Müller, 2016) § 96 MN 18–9 f; and the report on Belgium in this volume. 49 See for details Armin von Bogdandy, Christoph Grabenwarter and Peter M Huber (eds), Handbuch Ius Publicum Europaeum, Vol VI (Heidelberg, Müller, 2016). 50 The first BVerfG Senate decision dates back to 9 September 1951 (BVerfG 09.09.1951, 2 BvQ 1/51 (Südweststaat) BVerfGE 1, 1 ECLI:DE:BVerfG:1951:qs19510909.2bvq000151), the first ECJ decisions to the 21 December 1954 (ECJ 21.12.1954 Case 1/54 (France v High Authority) ECLI:EU:C:1954:7; ECJ 21.12.1954 Case 2/54 (Italy v High Authority) ECLI:EU:C:1954:8); the first case pending (ECJ, Case 1/53 Verband Deutscher Reeder v High Authority) was settled by way of a withdrawal of application. 51 BVerfG 29.05.1974, 2 BvL 52/71 (Solange I) BVerfGE 37, 271 ECLI:DE:BVerfG:1974:ls19740529.2bvl005271. 52 BVerfG 2 BvR 197/83 (Solange II) (n 43). 53 BVerfG 2 BvR 2661/06 (Honeywell) (n 46). 54 ECJ 22.11.2005 C-144/04 (Mangold) ECLI:EU:C:2005:709. 55 BVerfG 02.03.2010, 1 BvR 256/08 (Vorratsdatenspeicherung) BVerfGE 125, 260 ECLI:DE:BVerfG:2010:rs20100302. 1bvr025608. 56 CJEU 07.05.2013 C-671/10 (Åkerberg Fransson) ECLI:EU:C:2013:105. 57 CJEU 26.02.2013 C-399/11 (Melloni) ECLI:EU:C:2013:107. 58 BVerfG 24.04.2013, 1 BvR 1215/07 (Antiterrordatei) BVerfGE 133, 277 ECLI:DE:BVerfG:2013:rs20130424. 1bvr121507. 59 BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46). 60 BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46). 61 CJEU 16.06.2015 C-62/14 (Peter Gauweiler and Others) ECLI:EU:C:2015:400. 62 BVerfG 21.06.2016, 2 BvR 2728/13 (OMT (Hauptsacheverfahren)) ECLI:DE:BVerfG:2016:rs20160621.2bvr272813, paras 174 et seq.

Germany  137 Senate, 2017)63 went on to request a preliminary ruling on the ECB’s quantitative easing policy, mainly the Public Sector Purchase Programme (PSPP), thereby causing the answer of the CJEU in the Weiss case.64 The co-evolution of ECJ/CJEU and BVerfG can be related to at least six aspects that characterise a structurally analogous development. First, both courts are established in the beginning of the 1950s as novel juridical guardians of unprecedented highest-ranking sources of law – since these claim primacy as well as direct application before all other law. In both cases it was initially unpredictable whether and how the new judicial bodies will take root among and partly at the expense of well-established institutions.65 Second, both courts were equipped with extensive jurisdictional competences from the start, without it being predictable or probably even imaginable how extensively both courts would interpret and exercise their powers. Third, both courts justified their own significance essentially through decisions that claim a particular quality for the respective laws with which they were entrusted. For the BVerfG, the 1958 Lüth decision66 deserves mention, which introduced the ‘radiating effect’ (Ausstrahlungswirkung) of fundamental rights. For the ECJ, it is the 1964 Costa/ENEL case,67 establishing the primacy of EC law over all national law of the Member States. Fourth, both courts tend to practice an expansive interpretation of their respective law. Such expansive interpretation has two specific consequences. One consequence is that these laws’ area of application becomes ever more extensive. The other consequence is that the legal requirements applied by courts in exercising judicial review become ever denser. Fifth, both courts became key actors in their respective legal order of reference that cannot adequately, much less exclusively, be characterised as mere bodies of the judiciary. This is for instance expressed in the fact that the BVerfG proclaimed itself a ‘constitutional body’ (Verfassungsorgan),68 whereas the CJEU is referred to as the ‘engine of integration’.69 Sixth, the interferences or even confrontations of both courts have increased only (but then considerably) when the EC were no longer meant to be exclusively economic, whose key rights are constituted by the market freedoms. That is, the confrontation intensified when the EC began to strive for an ‘ever closer union’ under the umbrella of the EU wanting to be a fundamental rights union, a monetary union, a social union and a political union – one might term this a constitutional union.

B.  The Karlsruhe Integration Jurisprudence – A Statistical Approximation One may also speak of a co-evolution with regard to the fact that the BVerfG has accompanied EC and EU law development – and with it, the ECJ’s/CJEU’s jurisprudence as well – almost from 63 BVerfG 18.07.2017, 2 BvR 859/15 (PSPP) ECLI:DE:BVerfG:2017:rs20170718.2bvr085915. 64 CJEU 11.12.2018 C-493/17 (Weiss and Others) ECLI:EU:C:2018:1000. The final judgment of the BVerfG is still pending at the time of writing. 65 For the BVerfG’s development during the last 60 years, see notably Christoph Schönberger, ‘Anmerkungen zu Karlsruhe’, in M Jestaedt, O Lepsius, C Möllers and C Schönberger, Das entgrenzte Gericht (Frankfurt/M, Suhrkamp, 2011) 9; Matthias Jestaedt, ‘Phänomen Bundesverfassungsgericht: Was das Gericht zu dem macht, was es ist’, ibid, 77; Oliver Lepsius, ‘Die maßstabsetzende Gewalt’, ibid, 159; Christoph Möllers, ‘Legalität, Legitimität und Legitimation des Bundesverfassungsgerichts’, ibid, 281. 66 BVerfG 15.01.1958, 1 BvR 400/51 (Lüth) BVerfGE 7, 198 ECLI:DE:BVerfG:1951:rs19580115.1bvr040051. 67 ECJ 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66. 68 Such in the ‘Denkschrift des Bundesverfassungsgerichts’ (1957) Jahrbuch des öffentlichen Rechts der Gegenwart N.F. 144; see also § 1(1) BVerfGG (already in the original version of 12 March 1951, BGBl I 243): ‘The Federal Constitutional Court shall be a federal court of justice, unaffiliated with, and independent of, all other constitutional organs.’ 69 See Paul Kirchhof, ‘Der deutsche Staat im Prozeß der europäischen Integration’, in J Isensee and P Kirchhof (eds), Handbuch des Staatsrechts der Bundesrepublik Deutschland, Vol X, 3rd edn (Heidelberg, Müller, 2012) § 214 MN 55, with references.

138  Matthias Jestaedt the start with its own decisions. Role, type and significance of the Karlsruhe accompaniment have changed in reaction to the developing European integration. It changed not only from the point of view of the other Member States and the EC/EU, but also and in particular in the domestic discussion over the European unification venture. The turning points were the Treaties of Maastricht (1992) and Lisbon (2007). This is not least due to the fact that the dominant political class of the Federal Republic of Germany has a basic consensus on European policy, which largely prevents the legitimacy of the European Union project from being questioned in principle and subjected to criticism in the established parties with governmental responsibility at federal or state level. Since the fundamental questions about the ‘Why’ and the ‘Where to’ of the EU, about its ‘How’ and ‘How much’, do not enter the grand political arena, they seek another forum in constitutional law and the BVerfG. Thus, the BVerfG is often disproportionally called upon to answer questions of European integration by EU critics, whose points of view will not be heard by the relevant political actors. In this way, proceedings before the BVerfG become a substitute to the political arena, where an open dispute about the EU is lacking. Whoever wishes to gain a reliable impression of the BVerfG’s jurisprudence on European matters should start by getting an overview of the variety, temporal sequence and subject matter of the relevant decisions. In the nearly 50 years between 1967 and 2015, more than two dozen Senate decisions can be counted that take a more-than-marginal stand on questions of European integration from the Basic Law’s point of view. Specifically, we are talking about the following decisions (in chronological order): Table 5.1  Statistics of the BVerfG Jurisprudence on Questions of European Integration70 #

Date

File reference71

Subject matter/ case name

Senate Dissent Length Reference

1.

18.10.1967 1 BvR 248/63, 216/67

Gemeinsame Marktorganisation für Obst und Gemüse

I

–

7 pp

Fn 78

2.

09.06.1971 2 BvR 225/69

Umsatzausgleichsteuer auf eingeführtes Milchpulver

II

–

36 pp

Fn 78

3.

29.05.1974 2 BvL 52/71

Solange I

II

yes

35 pp

Fn 43

4.

25.07.1979 2 BvL 6/77

Absatzfondsgesetz

II

–

17 pp

Fn 105

5.

22.10.1986 2 BvR 197/83

Solange II

II

–

50 pp

Fn 43

6.

08.04.1987 2 BvR 687/85

Kloppenburg

II

–

24 pp

Fn 46

7.

12.10.1993 2 BvR 2134, 2159/92

Maastricht

II

–

59 pp

Fn 43 (continued)

70 Only Senate decisions are listed; chamber decisions, comprising in fact an average of 99.5 per cent of all BVerfG decisions but in principle following the path sketched out in the Senate decisions, were not taken into consideration. Likewise, Senate decisions dealing solely with procedural questions without a specific relation to Union law were not taken into consideration (eg BVerfGE 134, 357; BVerfGE 136, 119; BVerfGE 136, 121; BVerfGE 136, 125; BVerfGE 137, 103, 137, 345). 71 Legend: BvC = Election complaint (Wahlbeschwerde), Art  41(2) GG; BvE = Proceedings between governmental bodies (Organstreit), Art 93(1) no 1 GG; BvF = ‘abstract’ review of statutes (abstrakte Normenkontrolle), Art 93(1) no 2 GG; BvL = ‘concrete’ review of statutes (konkrete Normenkontrolle), Art 100(1) GG; BvR = Verfassungsbeschwerde (constitutional complaint), Art 93(1) no 4a GG.

Germany  139 Table 5.1  (Continued) #

Date

File reference

Subject matter/ case name

Senate Dissent Length Reference

8.

31.03.1998 2 BvR 1877/97, 50/98

Euro

II

–

29 pp

Fn 27

9.

07.06.2000 2 BvL 1/97

Bananenmarktordnung

II

–

20 pp

Fn 26

10. 18.07.2005 2 BvR 2236/04

Europäischer Haftbefehl I

II

yes

76 pp

Fn 162

11. 13.03.2007 1 BvF 1/05

TreibhausgasEmissionsberechtigung

I

–

33 pp

Fn 107

12. 30.06.2009 2 BvE 2, 5/08, 2 BvR 1010, 1022, 1259/08, 182 /09

Lissabon

II

–

171 pp

Fn 44

13. 06.07.2010 2 BvR 2661/06

Honeywell

II

yes

46 pp

Fn 46

14. 19.07.2011 1 BvR 1916/09

Le Corbusier

I

–

30 pp

Fn 82

15. 07.09.2011 2 BvR 987, 1485, 1099/10

Euro-Rettungsschirm

II

–

63 pp

Fn 123

16. 04.10.2011 1 BvL 3/08

Investitionszulage

I

–

23 pp

Fn 106

17. 09.11.2011 2 BvC 4, 6, 8/10

5%-Sperrklausel bei EP-Wahlen

II

yes

56 pp

Fn 7272

18. 28.02.2012 2 BvE 8/11

EFSF-BT-Sondergremium

II

–

50 pp

Fn 190

19. 19.06.2012 2 BvE 4/11

EFSF

II

–

79 pp

Fn 72

ESM – Aussetzung der 20. 12.09.2012 2 BvR 1390, Ratifikation I 1421, 1438, 1439, 1449/12, 2 BvE 6/12

II

–

93 pp

Fn 145

21. 12.09.2012 2 BvR 1824/12

ESM – Aussetzung der Ratifikation II

II

–

8 pp

Fn 72

22. 24.04.2013 1 BvR 1215/07

Antiterrordatei

I

–

101 pp

Fn 58

23. 14.01.2014 2 BvR 2728, 2729, 2730, 2731/13, 2 BvE 13/13

OMT (Reference to ECJ)

II

yes

73 pp

Fn 46

24. 26.02.2014 2 BvE 2, 5–10, 12/13, 2 BvR 2220, 2221, 2238/13

3%-Sperrklausel bei EP-Wahlen

II

yes

64 pp

Fn 92

(continued) 72 BVerfG 09.11.2011, 2 BvC 4/10 (5 %-Sperrklausel bei EP-Wahlen) BVerfGE 129, 300 ECLI:DE:BVerfG:2011: cs20111109.2bvc000410; BVerfG 19.06.2012, 2 BvE 4/11 (EFSF) BVerfGE 131, 152 ECLI:DE:BVerfG:2012:es20120619. 2bve000411; BVerfG 12.09.2012, 2 BvR 1824/12 (ESM – Aussetzung der Ratifikation II) BVerfGE 132, 287 ECLI:DE:BVerfG:20 12:rs20120912.2bvr182412; BVerfG 13.10.2016, 2 BvR 1368/16 (CETA – Preliminary Injunction I) BVerfGE 143, 65 DE: BVerfG:2016:rs20161013.2bvr136816; BVerfG 07.12.2016, 2 BvR 1444/16 (CETA – Preliminary Injunction II) BVerfGE 143, 65 ECLI:DE:BVerfG:2016:rs20161207.2bvr144416.

140  Matthias Jestaedt Table 5.1  (Continued) #

Date

File reference

Subject matter/ case name

Senate Dissent Length Reference

25. 18.03.2014 2 BvR 1390, 1421, 1438, 1439, 1440, 1824/12, 2 BvE 6/12

ESM – Hauptsacheverfahren

II

–

117 pp

Fn 145

26. 15.12.2015 2 BvR 2735/14

Europäischer Haftbefehl II

II

–

60 pp

Fn 46

27. 21.06.2016 2 BvR 2728, 2729, 2730, 2731/13, 2 BvE 13/13

OMT – Hauptsacheverfahren

II

–

112 pp

Fn 62

28. 13.10.2016 2 BvR 1368/16, 1444/16, 1482/16, 3/16

CETA – Preliminary Injunction I

II

–

37 pp

Fn 72

29. 07.12.2016 2 BvR 1444/16, 1482/16, 1823/16, 3/16

CETA – Preliminary Injunction II

17 pp

Fn 72

30. 18.07.2017 2 BvR 859/15, 1651/15, 2006/15, 980/16

PSPP (Reference to ECJ)

79 pp

Fn 63

II

–

It must be noted that although the Maastricht and particularly the Lisbon decisions73 both attracted the attention especially of foreign observers in a particular way, the BVerfG had already set key points of the agenda in matters of European integration even a quarter of a century before the ground-breaking Maastricht decision. It is, however, also correct that the frequency of decisions of principle has significantly increased since the Treaty of Lisbon entered into force in 2009. While 12 decisions were passed during the first 40 years (1967–2007), 18 – that are, on average, considerably longer – can already be counted during the 10 years since the Lisbon decision (2009–18). Of the 30 decisions filling almost 1,700 printed pages in the official reports (the Lisbon decision alone comprises 171 pages), 25 originate from the Second Senate (primarily responsible for constitutional cases relating to government organisation) and five from the First Senate (which is primarily responsible for fundamental rights). It is also revealing in which type of procedure the decisions came about. According to the type of procedure conceivable from the file reference,74 the 30 proceedings (cut across in different ways) divide into 23 individual constitutional complaints, 10 proceedings between governmental bodies (Organstreitverfahren), four applications by inferior courts for ‘concrete’ reviews of statutes (konkrete Normenkontrolle), one application by a state government for ‘abstract’ review of statutes (abstrakte Normenkontrolle) and one electoral complaint (Wahlbeschwerde). Dissenting opinions were added to six out of 30 decisions, corresponding to 20 per cent of the decisions. This is a comparatively high quota, considering the general dissenting opinion quota for Senate



73 BVerfG 74 See

2 BvR 2134 (Maastricht) (n 43); BVerfG 2 BvE 2/08 (Lissabon) (n 44). n 71.

Germany  141 decisions of 7.8 per cent.75 This indicates that the reasoning in the decisions is a result of struggles on the bench. The more recent decisions in particular – above all, the Lisbon decision that laid the ground for the current direction of the court – have been met with strong reservations from parts of the academic constitutional law community. The Court, however, as far as can be seen, has been little impressed by this. As for their subject matter, the earlier decisions dealt with questions of fundamental rights as well as questions concerning the possibility of judicial review of secondary EC law. Since the founding of the EU with the Treaty of Maastricht, questions of democratic legitimisation, parliamentary participation and budgetary sovereignty moved to the fore. This was even though questions of fundamental rights are to be decided again on the EU level due to the strengthening of the fundamental rights protection by the Charter of Fundamental Rights.

C.  Arguments Promoting Integration Evidently, the temporal and spatial extension, the dogmatic complexity and the sophistication of the Karlsruhe jurisprudence cannot be addressed adequately within the limits of this chapter. In an attempt at bringing the Karlsruhe jurisprudence down to a common denominator, ignoring for a moment how varied and heterogeneous the decisions have been over the course of almost half a century, one might say: both senates are loyal to the CJEU’s jurisprudence in a critical way (as they are bound by the Basic Law), striving at the same time to promote European integration and to protect German identity. Accordingly, on the one hand, they do not give in to all EC and EU law claims, disregarding their mandate to protect the Basic Law. On the other hand, they try to avoid conflicts with EU law and with the CJEU wherever feasible. This structural ­ambivalence – or dialectics, as it were – of the Karlsruhe jurisprudence may, thus, serve as a leitmotif for an analysis that can be achieved within the given parameters here. This analysis will distinguish, among the concepts and forms of argument that appear in the case law, between those that (tend to) serve the Basic Law’s openness for the European integration and those that rather limit and canalise it.76 The former will be called arguments promoting integration (see below, section V.C.1–5) and the latter, arguments limiting integration (see below, section V.D.). Note, however, that hardly any concept may be exclusively interpreted in either the one or the other sense and that legal solutions most often result from a combination of arguments promoting and limiting integration. This needs to be emphasised expressly in order to avoid misconceptions.

1.  Acknowledging the Supranational Character of European Union Law (Primacy, Direct Effect) The BVerfG has never questioned the notion of supranational law, encompassing the elements of primacy (of application) and the direct effect, which the CJEU pronounced as the particularity of EC or EU law. On the contrary, it has seen it as compatible with the Basic Law from the very start. 75 The general dissenting opinion quota of 7.8  per  cent is calculated as follows: of a total of 2,172 Senate decisions over the time period from 1971 until 2015 (BVerfGE 30 to BVerfGE 137) 158 have been passed including a dissenting opinion (see BVerfG, Jahresstatistik 2015, www.bundesverfassungsgericht.de/DE/Verfahren/Jahresstatistiken/2015/ statistik_2015_node.html (at A.I.7.)). 76 This approach takes account of the fact that the BVerfG’s jurisprudence operates in a pronounced term- and conceptled way; this facilitates and encourages a doctrinally expansive system thinking (Systemdenken). – See for a similar depiction also Voßkuhle, ‘Integration’ (n 42) 161; see further Andreas Voßkuhle, ‘Die europäische Integration und das Bundesverfassungsgericht’, in Juristische Studiengesellschaft Karlsruhe, Jahresband 2014 (Heidelberg, Müller, 2015) 207.

142  Matthias Jestaedt Already in its first relevant decision of 1967, the order concerning Regulation (EEC) No 159/66 on the Common Market Organisation for Fruit and Vegetables, the BVerfG recognised that the EEC, although not a state, was a ‘community sui generis in the process of progressing integration’ and exercised an autonomous public authority that Member States neither need affirm nor may repeal. The secondary law formed ‘a distinct legal system, whose norms are neither international law nor national law of the Member States’. In this context, the BVerfG affirmatively cites the ECJ’s phrase, coined inter alia in the Costa/ENEL case,77 that EC law and the domestic law of the Member States are ‘two separate and distinct legal orders’, the secondary EC law running from an ‘independent source of law’.78 Accordingly, the BVerfG does not express constitutional concerns against the ECJ’s jurisprudence on the direct effect of (primary and) secondary legal instruments in general, or on the (vertical) direct effect of directives in particular.79 The primacy of application of EC law instruments over Member State statute law is said to be based on unwritten norms of primary EC law.80 As the BVerfG argues, quoting Costa/ENEL, the EU could not exist as a legal community (Rechtsgemeinschaft) if the uniform applicability of EU law in the Member States was not guaranteed.81 In its more recent decisions, the BVerfG has confirmed that the primacy of application of supranational (primary and secondary) law also generally applies at the expense of the Basic Law itself: ‘The primacy of application of Union law over national law in principle also applies to conflicting national constitutional law and leads to its inapplicability in case of collision.’82 Since supranational law and national law are two independent legal orders, the primacy of application of EC or EU law needs a domestic legal basis (Rechtsanwendungsbefehl) so that it can have effect within the legal order erected by the Basic Law. This domestic legislation giving effect to it is seen in the Approving Acts (Zustimmungsgesetze) for the Treaties, based on Articles 24(1) (old version) and 23(1) GG (new version) respectively, and Article  59(2) GG.83 Thus, from the German constitutional point of view, the primacy of application for the benefit of European primary and secondary law over (almost) all German law is neither original nor unconditional but rather derivative and qualified. First, it is derived from the Approving Acts as its domestic legal anchor based on Article 24(1) GG (old version) (until 1993) and Article 23(1) GG (since 1993).84 Second, it is conditional upon compliance with the constitutional limits that the Basic Law sets for the transfer of sovereign powers to the European level in its Article 24(1) GG (old version) (until 1993) and Article 23(1) GG (since 1993).85 To sharpen the argument: even EU law’s general primacy of application over constitutional law 77 ECJ 6/64 (Costa v E.N.E.L.) (n 67). 78 All quotations: BVerfG 17.10.1967, 1 BvR 248/63 (Gemeinsame Marktorganisation für Obst und Gemüse) BVerfGE 22, 293, 296 ECLI:DE:BVerfG:1967:rs19671018.1bvr024863; confirmed by BVerfG 09.06.1971, 2 BvR 225/69 (Umsatzausgleichsteuer auf eingeführtes Milchpulver) BVerfGE 31, 145, 173–4 f ECLI:DE:BVerfG:1971:rs19710609. 2bvr022569; BVerfG 2 BvL 52/71 (Solange I) (n 51) 277–8. 79 Leading case: BVerfG 2 BvR 687/85 (Kloppenburg) (n 46) 234–44, particularly 240–4. 80 See ibid, 244 referring to BVerfG 2 BvR 225/69 (Umsatzausgleichsteuer auf eingeführtes Milchpulver) (n 78) 173 et seq. 81 BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 302. 82 Most recently, BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 38 referring to BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 301; BVerfG 19.07.2011, 1 BvR 1916/09 (Le Corbusier) BVerfGE 129, 78, 100 ECLI:DE:BVerfG: 2011:rs20110719.1 BvR 191609. 83 Established jurisprudence, see BVerfG 2 BvR 197/83 (Solange II) (n 43) 375. 84 See BVerfG 2 BvE 2/08 (Lissabon) (n 44) 400–1. 85 See BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 26 referring to BVerfG 2 BvR 197/83 (Solange II) (n 43) 375; BVerfG 2 BvE 2/08 (Lissabon) (n 44) 398; BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 302; the judgment asserts but does not justify the following thesis: ‘Unlike the primacy of application of federal law in a federal state, the precedence of Union law, which is based on national legislation giving effect to it, cannot be comprehensive’. Whether the primacy is originary or derivative is not decided by the question whether the primacy applies comprehensively or limitedly. That the Union law primacy applies only conditionally and thus limitedly in Germany is not tied to the EU legal nature as a non-federal state, but plainly depends upon the integration opening clause of Art 23(1) GG.

Germany  143 is a ‘primacy by virtue of constitutional empowerment’ (Vorrang kraft verfassungsrechtlicher Ermächtigung)86 that cannot reach further than the underlying authorisation.87 Nevertheless, the integration clause of Article 23(1) GG contains a ‘promise of effectiveness and implementation’88 as a last basis for the domestic effect of the primacy of application of primary and secondary EU law in its favour. The way in which the primacy of application may work at the expense of national constitutional provisions is documented by the First Senate’s line of argument in Le Corbusier (2011). In that case, the BVerfG affirms an ‘expansion of applicability’ (Anwendungserweiterung)89 of Article  19(3) GG.90 While this provision by itself is valid only for ‘domestic legal entities’, the BVerfG expands its scope of application to legal entities from other EU Member States. It does so by way of an ‘interpretation of national law in conformity with Union law’ (unionsrechtskonformer Auslegung des nationalen Rechts) – even though the Court does not consider this ‘expansion of applicability’ to be justifiable by the provision’s wording:91 The fundamental freedoms and the general ban on discrimination [ie, Article 18 TFEU] prohibit the unequal treatment of domestic and foreign enterprises from the European Union in the sphere of application of Union law, and in this regard override the limitation of protection of fundamental rights to domestic legal entities provided for in Article 19.3 of the Basic Law.92

To justify its reasoning, the Senate argues: The expansion of the application of Article  19.3 of the Basic Law to cover legal entities from other Member States of the European Union is a reaction to developments in the European Treaties and Union legislation, and avoids a collision with Union law … The provisions of European Union law do not suppress Article 19.3 of the Basic Law, but merely prompt an extension of the protection of fundamental rights to further legal subjects of the Single Market.93

2.  Acknowledging the CJEU’s Power to Develop the Law The BVerfG insists that EC and EU law is applicable in Germany (1) only by virtue of domestic legislation adopted to give effect to it (Rechtsanwendungsbefehl), ie the Approving Acts, (2) only within the scope of the competences transferred, and (3) only by virtue of an act of transfer that observes the ‘requirements of parliamentary responsibility’94 (ie the responsibility 86 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 397 (BVerfG translation MN 332). 87 Representatively BVerfG 2 BvR 197/83 (Solange II) (n 43) 374–5; BVerfG 2 BvR 2134 (Maastricht) (n 43) 190; BVerfG 2 BvE 2/08 (Lissabon) (n 44) 402; BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 302; BVerfG 1 BvR 1916/09 (Le Corbusier) (n 82) 99; BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 26; BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 34. 88 BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 37 referring to BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 302. 89 BVerfG 1 BvR 1916/09 (Le Corbusier) (n 82) 96–7. 90 Art 19(3) GG reads as follows: ‘The fundamental rights shall also apply to domestic legal entities to the extent that the nature of such rights permits.’ 91 BVerfG 1 BvR 1916/09 (Le Corbusier) (n 82) 99 (BVerfG translation MN 80, 68). The First Senate considers it essential that the 1948/49 Parliamentary Council did not formulate the limitation on ‘domestic’ legal entities against the background of a ‘highly integrated “association of states” (BVerfGE 123, 267 [348] [= BVerfG 2 BvE 2/08 (Lissabon) (n 44)])’. 92 BVerfG 1 BvR 1916/09 (Le Corbusier) (n 82) 97 (BVerfG translation MN 75). Irrespectively, the First Senate assumed in the ‘Le Corbusier’ order that a reference to the CJEU was superfluous according to the ‘acte clair’ doctrine of the CILFIT jurisprudence (ECJ 06.10.1982 Case 283/81 (CILFIT v Ministero della Sanità) ECLI:EU:C:1982:335): BVerfG 1 BvR 1916/09 (Le Corbusier) (n 82) 99; similarly in the results BVerfG 26.02.2014, 2 BvE 2/13 (3%-Sperrklausel bei EP-Wahlen) BVerfGE 135, 259, MN 42 ECLI:DE:BVerfG:2014:es20140226.2bve000213. 93 BVerfG 1 BvR 1916/09 (Le Corbusier) (n 82) 99–100 (BVerfG translation MN 81). 94 BVerfG 2 BvR 2134 (Maastricht) (n 43) 200.

144  Matthias Jestaedt of the German  Bundestag), which defines the transfer’s extent and scope. Nonetheless, the BVerfG does not consider the fact that it was only the CJEU’s ‘further development of the law’ (Rechtsfortbildung) that has created core effects of EC and EU law to be a departure from or violation of the constitutional authorisation for European integration in Article 23(1) GG new version (Article 24(1) GG old version). The BVerfG’s established jurisprudence emphasises that the CJEU is ‘not precluded from refining the law by means of methodically bound case law’.95 In the Kloppenburg judgment of 1987, which was ground-breaking in this respect, the BVerfG held: ‘[i]t is compatible with Article 24 (1) GG [old version] to assign to the Court of Justice, an international institution, an authority of this kind to develop the law in the sphere of the jurisdiction of that institution’ especially since this method of ascertaining and applying the law has been ‘developed over centuries of common European legal tradition and refinement of law’. In the OMT judgment, the BVerfG confirmed that this is the case ‘even when the Court of Justice adopts a view against which weighty arguments could be made’, and that this ‘applies both in the context of the identity review and the ultra vires review’.96 Of course, recognising the CJEU’s authority to develop the law is not supposed to mean that it has ‘been given adjudicative power to extend its jurisdiction without limit’.97 Yet some examples should demonstrate the extent to which the BVerfG acknowledges the CJEU’s alleged power to further develop the law. The BVerfG did not raise any objections in principle against (1) the application of the implied powers doctrine, (2) the interpretation and application of EC or EU law in light of the effet utile, (3) the development by the Luxembourg Court of principles of primacy and direct applicability of EC or EU law, or (4) deriving general principles of EC and EU law.98 Regarding the Mangold case,99 the BVerfG did not even object to the very extensive interpretation of the prohibition of age discrimination and the advance effect of directives.100

3.  Effectuating the Primacy of the CJEU in the Basic Law Those who reproach the BVerfG for having referred only two cases to the CJEU for a preliminary ruling so far101 – in OMT (2014) and in PSPP (2017) – fail to see that German courts are 95 BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 305 (BVerfG translation MN 62) referring to BVerfG 2 BvR 687/85 (Kloppenburg) (n 46) 242–3; BVerfG 2 BvE 2/08 (Lissabon) (n 44) 351–2 (BVerfG translation MN 237–8). 96 BVerfG 2 BvR 2728/13 (OMT (Hauptsacheverfahren)) (n 62) para 161. 97 BVerfG 2 BvR 687/85 (Kloppenburg) (n 46) 242–3. It is not quite clear how this relates to the Court’s statement that what is decisive in the end is ‘that the membership of the Federal Republic of Germany and the ensuing rights and o ­ bligations – especially the legally binding direct acts of the European Communities within the national legal sphere – have been defined in the Treaty so as to be predictable for the legislature, and have been enacted by it with sufficient certainty in the act of assent’ (BVerfG 2 BvR 2134 (Maastricht) (n 43) 187–8), referring to BVerfG 23.06.1987, 2 BvR 1107/79 (Eurocontrol) BVerfGE 58, 1, 37 ECLI:DE:BVerfG:1987:rs19870623:2bvr110779; BVerfG 18.12.1984, 2 BvE 13/83 (Pershing II) BVerfGE 68, 1, 98–9 ECLI:DE:BVerfG:1984:es:19841218:2bve001383; quoted verbatim in BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 21. It is hardly plausible that the CJEU’s interpretation of Union law, be it characterised as judicial law-making or as extensive interpretation, is always ‘predictable for the legislature, and [has] been enacted by it with sufficient certainty in the act of assent’. 98 See, eg, BVerfG 2 BvE 2/08 (Lissabon) (n 44) 351. On the creation of Community law fundamental rights protection since the 1970s, see, expressly, BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 306. 99 ECJ C-144/04 (Mangold) (n 54). 100 BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 308–13, in this case, of course, applying the standard of the ultra vires review (see below, section V.D.1). The Court held (ibid, 308, BVerfG translation MN 68): ‘It is irrelevant whether the outcome found in the Mangold ruling can still be gained by recognised legal interpretation methods and whether any existing shortcomings would be evident. At any rate, it does not constitute a transgression of the sovereign powers assigned to the European Union by an Approving Act [Zustimmungsgesetz], thus violating the principle of conferral in a manifest and structurally effective manner.’ 101 Scarce as this specific action might be, it is not unusual when compared to the practice of other constitutional courts: for example, until the end of 2014 the Spanish Tribunal Constitucional and the French Conseil constitutionnel also

Germany  145 both nominally and relatively among those most inclined to refer cases.102 They also fail to see that the BVerfG (currently) holds that the German courts’ obligation to refer (Article  267(3) TFEU) is an obligation under constitutional law. According to the BVerfG, the CJEU is a ‘lawful judge’ (­ gesetzlicher Richter) within the meaning of the Basic Law. Therefore, if a German court disregards its obligation, as a Member State’s court of last instance, to initiate preliminary ruling proceedings, this is at the same time a violation of the (procedural) fundamental right to a lawful judge, enshrined in Article 101(1) sentence 2 GG. The alleged violation of this right can be used as a basis for an individual constitutional complaint before the BVerfG (Article 93(1) No 4a GG), after the exhaustion of all other legal remedies.103 However, the BVerfG criticises ‘the interpretation and application of rules on competences only if, on a sensible interpretation of the concepts determining the Basic Law, they no longer appear to be comprehensible and are manifestly untenable’, ie, arbitrary.104 This is how the BVerfG helps to ensure that German courts observe the CJEU’s primacy with regard to the interpretation as well as the annulment of EU law, as laid down in Article 267 TFEU. From the very beginning, Karlsruhe’s jurisprudence has been shaped by the idea that, under Community law, ‘the Court of Justice, in relation to the courts of the Member States, has the final power of decision on the interpretation of the Treaty and on the validity and interpretation of the acts of secondary Community law referred to therein [sc in Article 177 EEC Treaty, today Article 267 TFEU]’ – just as, conversely, under the Basic Law the final power of decision on the interpretation of the Basic Law and on the validity of the laws adopted on its basis is vested in the BVerfG.105 The BVerfG also contributes to securing the CJEU’s authority: In accordance with its Solange jurisprudence (see section V.C.4 below), the BVerfG holds that the submission of a (German) statute by a judge for possible annulment (konkrete Normenkontrolle, Article 100(1) GG) is inadmissible for lack of relevance to the referring court’s decision, if and as long as that court ‘has not clarified whether the law which it considers to be unconstitutional was enacted in implementation of a discretion remaining to the national legislature under Union law’.106 If necessary, the referring court first has to seek a preliminary ruling by the CJEU (Article 267 TFEU). Where the national legislator has no discretion in the transposition of directives, the BVerfG and the other German courts, in accordance with the Solange jurisprudence, refrain from reviewing the act of transposition in terms of the fundamental rights of the Basic Law. In these referred but one case each by way of Article 267 TFEU (or previous versions of this article), the Italian Corte Costituzionale referred two cases, the Austrian Verfassungsgerichtshof five and some constitutional courts have referred none, like the Polish and Czech. The Belgian Cour constitutionnelle is the leader with 30 referrals. (Figures taken from: Court of Justice of the European Union, Annual Report 2014 (Luxembourg, Publications Office of the European Union, 2015) 117–9). 102 From 1952 to 2014, there were 8,710 references for preliminary ruling, whereof 2,137, or 24.54%, originated from courts in Germany. For comparison, the other original Member States – Italy, the Netherlands, France, Belgium and Luxembourg – account for 1,279, 909, 906, 762 and 83 references, respectively. The number of references for the United Kingdom (since 1974) is 573, for Spain (since 1986) is 354 and for Austria (since 1995) is 447 (see Court of Justice of the European Union, Annual Report 2014 (Luxembourg, Publications Office of the European Union, 2015) 115–6). The percentage of references that come from the highest state courts is 33.5% in Germany, which is comparatively high. It is the highest among the most populous Member States (Italy 18.67%; Netherlands 65.9%; France 22.3%; Belgium 25.2%; Luxembourg 38.55%; United Kingdom 20.77%; Spain 15.25%; Austria 42.28%) (ibid, 117–9). 103 Landmark case: BVerfG 2 BvR 197/83 (Solange II) (n 43) 366–9. See also Gabriele Britz, ‘Verfassungsrechtliche Effektuierung des Vorabentscheidungsverfahrens’, (2012) Neue Juristische Wochenschrift 1313. 104 BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 286 (Headnote 3), 315 et seq, affirming BVerfG 2 BvR 687/85 (Kloppenburg) (n 46) 233–4; BVerfG 1 BvR 1916/09 (Le Corbusier) (n 82) 105–6. 105 Quote: BVerfG 25.07.1979, 2 BvL 6/77 (Absatzfondsgesetz) BVerfGE 52, 187, 200 ECLI:DE:BVerfG:1979:ls19790725. 2bvl000677; BVerfG 2 BvR 687/85 (Kloppenburg) (n 46) 234. 106 BVerfG 04.10.2011, 1 BvL 3/08 (Investitionszulage) BVerfGE 129, 186, 186 (Headnote 1), see also 199 et seq ECLI:DE: BVerfG:2011:ls20111004.1bvl000308.

146  Matthias Jestaedt cases, the BVerfG considers that under the guarantee of legal protection (Rechtsschutzgarantie, Article 19(4) GG) ordinary German courts are obliged to review the EC/EU measure in question as a possible ­violation of the fundamental rights of EC/EU law. If necessary, they have to initiate preliminary ruling proceedings.107 Submissions by a court to the BVerfG without them first having reviewed the measure itself are inadmissible. In particular, such a submission is not inadmissible if it constituted an abus de droit of the obligation to submit, eg regarding the right to a lawful judge (Recht auf den gesetzlichen Richter, Article 101(1) sentence 2 GG).108

4. The Solange Jurisprudence In 1974 the BVerfG rendered its much-discussed Solange I decision, by five votes to three and with a dissenting opinion. Here, the BVerfG first developed a concept of identity, according to which the transfer of sovereign power (Hoheitsrechte) to the EEC/EC/EU under Article 24(1) GG (old version) can only take place with due regard for ‘the basic structure of the Constitution, which forms the basis of its identity’. It counts ‘the part of the Basic Law dealing with fundamental rights’ to be among the Constitution’s essential features.109 On this basis, the first Solange formula identified reservations regarding the applicability (and therefore the primacy) of EC law as long as the integration process has not progressed so far that Community law also receives a catalogue of fundamental rights decided by a parliament and of settled validity, which is adequate in comparison with the catalogue of fundamental rights contained in the Basic Law, a reference by a court in the Federal Republic of Germany to the Federal Constitutional Court in judicial review proceedings [Article 100 (1) GG], following the obtaining of a ruling of the European Court under Article 177 of the Treaty [now Article 267 TFEU], is admissible and necessary if the German court regards the rule of Community law which is relevant to its decision as inapplicable in the interpretation given by the European Court, because and in so far as it conflicts with one of the fundamental rights in the Basic Law.110

However, the Court was keen to point out that this is, ‘therefore, a legal difficulty arising exclusively from the Community’s continuing integration process, which is still in flux and which will end with the present transitional phase.’111 107 BVerfG 13.03.2007, 1 BvF 1/05 (Treibhausgas-Emissionsberechtigung) BVerfGE 118, 79, 95 et seq ECLI:DE:BVerfG: 2007:fs20070313.1bvf000105; see also BVerfG 1 BvL 3/08 (Investitionszulage) (n 106) 186, 199. 108 See BVerfG 1 BvL 3/08 (Investitionszulage) (n 106) 204. 109 BVerfG 2 BvL 52/71 (Solange I) (n 51) 279–80. The BVerfG established this reading of Art 24(1) GG not only with regard to European integration but as a general rule for the application of Art  24(1) GG, see BVerfG 2 BvR 1107/79 (Eurocontrol) (n 97) 40–1; academia mostly approved of this. On a side note, the concept of identity at first did not expressly refer to the ‘eternity clause’ enshrined in Art 79(3) GG. 110 BVerfG 2 BvL 52/71 (Solange I) (n 51) 271 (Headnote), 285. Justices Rupp, Hirsch and Wand argued in their dissenting opinion (which is well worth reading), that secondary Community law cannot be reviewed for its compatibility with the fundamental rights of the Basic Law. They also contested the procedural possibility to challenge an Act of secondary Community law by way of a specific judicial review of statutes under Art 100(1) GG (ibid, 291 et seq, 300 et seq). The decision is all the more remarkable because even the majority of the Senate held that in the case at hand Regulation No 120/67/ EEC and Regulation No 473/67/EEC did not infringe fundamental rights guaranteed under the Basic Law: ibid, 288–91). In other words, since there was no conflict with the Basic Law’s fundamental rights, the Senate did not have to address these fundamental questions as it did. In an order of 1979, the Second Senate incorporated the dissenting opinion of 1974 when it clarified that the jurisdiction of the BVerfG for the specific judicial review of statutes (Art 100(1) GG) extends only to the domestic act approving the treaty, not to the Act of primary or secondary Community law itself: BVerfG 25.07.1979, 2 BvL 6/77 (Absatzfondsgesetz) (n 104). This, by the way, is the same judgment in which the BVerfG leaves the question undecided, whether the Solange I jurisprudence is upheld, ‘having regard to political and legal developments which might have taken place in the meantime in European Community matters’ (ibid, 202–3). 111 BVerfG 2 BvL 52/71 (Solange I) (n 51) 280–1. As early as 1974, the BVerfG characterised the decisions of the ECJ as ‘admittedly favourable to fundamental rights so far’ (ibid, 280) but held that this fact alone did not mean that there was sufficient ‘legal certainty’ for the permanent adequacy of the Community law standard of protection of fundamental rights with that of the Basic Law.

Germany  147 Approximately 12 years later, in 1986, the Second Senate rendered an order, dubbed Solange II, unanimously this time, and affirmed the reservations with regard to constitutional identity, repeating the relevant passages of Solange I.112 However, by then it could take into account the ECJ’s emerging jurisprudence on fundamental rights – the ECJ’s judgment in the Nold case of 1974 is regarded as a breakthrough in this respect113 – and its reference to the ECHR with its additional protocols. The BVerfG held that ‘in the meantime’, a measure of protection of fundamental rights had been established under EC law, which ‘in its conception, substance and manner of implementation is essentially comparable with the standards of fundamental rights provided for in the Basic Law’.114 Consequently, it did not deem the reservation expressed in 1974 in Solange I to be well-founded any longer. With great confidence in the protection of fundamental rights on the EC level, the BVerfG turned the Solange formula on its head: As long as the European Communities, and in particular the case law of the European Court, generally ensure an effective protection of fundamental rights as against the sovereign powers of the Communities which is to be regarded as substantially similar to the protection of fundamental rights required unconditionally by the Basic Law … the Federal Constitutional Court will no longer exercise its jurisdiction to decide on the applicability of secondary Community legislation … and it will no longer review such legislation by the standard of the fundamental rights contained in the Basic Law; references to the Court under Article 100(1) GG for that purpose are therefore inadmissible.115

In hindsight, it seems fair to see in Solange I one of the factors that promoted the development of fundamental rights protection in EC law, energetically established by the ECJ/CJEU in the aftermath of the BVerfG’s judgment. Therefore, the BVerfG has indirectly had a share in raising the standard of fundamental rights protection under EC law. In the BVerfG’s Maastricht judgment of 1993, the Second Senate emphasised, on the one hand, that ‘[a]cts of the particular public power of a supranational organisation which is separate from the State power of the Member States may also affect those persons protected by fundamental rights in Germany.’ On the other hand, the duties of the Constitutional Court are not only the protection of fundamental rights vis-à-vis German governmental institutions, but – ­explicitly diverging from the Eurocontrol judgment of 1981116 – they also comprise the ‘protection of fundamental rights in Germany’.117 Some commentators regarded this as a departure from, or at least a relativisation of, the Solange line of jurisprudence.118 However, nearly seven years later, in Banana Market Organisation (2000), the BVerfG clarified that there are no contradictions between the Solange II and Maastricht decisions,119 thus returning to the Solange line of jurisprudence. In the Greenhouse Gas Emissions Allowance order of 2007, the First Senate followed suit and extended this jurisprudence, previously only relevant for regulations, to apply to directives

112 BVerfG 2 BvR 197/83 (Solange II) (n 43) 375 et seq. 113 ECJ 14.05.1974, 4/73 (Nold KG v Commission) ECLI:EU:C:1974:51. Ironically, this judgment was not rendered after, but 15 days before the Solange I order of 29 May 1974. 114 BVerfG 2 BvR 197/83 (Solange II) (n 43) 378–81, quote at 378. 115 Ibid 340 (Headnote 2), 387; since then a constant line of jurisprudence, see, eg, BVerfG 2 BvR 2134 (Maastricht) (n 43) 174–5; BVerfG 2 BvL 1/97 (Bananenmarktordnung) (n 26) 167; BVerfG 2 BvE 2/08 (Lissabon) (n 44) 335. 116 BVerfG 2 BvR 1107/79 (Eurocontrol) (n 97) 27. 117 BVerfG 2 BvR 2134 (Maastricht) (n 43) 175. 118 This does not seem far-fetched, given that the BVerfG has emphasised, in a different context, that it cannot in principle decide on the ‘validity of Community law, since it does not present an act of German state authority’ (BVerfG 1 BvF 1/05 (Treibhausgas-Emissionsberechtigung) (n 107) 95), referring to BVerfG 1 BvR 248/63 (Gemeinsame Marktorganisation für Obst und Gemüse) (n 78) 295 et seq; BVerfG 2 BvL 52/71 (Solange I) (n 51) 281–2. 119 BVerfG 2 BvL 1/97 (Bananenmarktordnung) (n 26) 163–5); in this judgment, the Second Senate retraces the ­development of its own jurisprudence, starting with Solange I (ibid, 161–4).

148  Matthias Jestaedt as well. The Senate went on to emphasise that, apart from secondary EC law, the national implementation measure for a directive which leaves no latitude for Member States is also exempt from review (according to the Solange II rule) on the standard of the fundamental rights of the Basic Law, both by the BVerfG and the regular courts.120 This is the current state of the BVerfG’s jurisprudence on this issue – with one exception. The so-called ‘ultra vires review’ (ultra viresKontrolle) and ‘review of constitutional identity’ (Identititätskontrolle) (see below, section V.D.1, 2) are not subject to the restrictions of the Solange line of jurisprudence.121

5.  Cooperative Relationship – Openness Towards European Law – Responsibility for Integration The BVerfG’s current jurisprudence is influenced deeply by the Second Senate’s landmark Maastricht and Lisbon judgments. There, it developed three concepts of ‘relationship’. The first concept is the ‘cooperative relationship’ (Kooperationsverhältnis) between the BVerfG and the ECJ. The second is the ‘openness to European law’ (Europarechtsfreundlichkeit) of both the interpretation and application of the Basic Law. The third is the ‘responsibility for integration’ (Integrationsverantwortung) of German constitutional organs as well as of other Member States and of the EU itself. These three concepts are similarly deduced, have a similar legal nature and have a similar goal, but their scopes of application also overlap. First, they are not expressly established in the constitutional text, but are revealed by way of a synoptic interpretation (Zusammenschau)122 of the relevant constitutional provisions, ie, by way of a teleological extrapolation on a medium level of abstraction. Second, despite this, the BVerfG has extracted a number of constitutional obligations and prohibitions from these which are considered binding for the central actors. Third, all three of these concepts address questions of the relationship and division of responsibility between the German (constitutional) and the EU legal orders and between (national) identity and (supranational) integration. A central idea throughout is the operability of both parts, as well as of the resultant whole.123 Put in terms of the theory of the division of federal competences (bundesstaatliche Kompetenzlehre), none of the three concepts is a basis for competences for one of the two sides (federation or component state, respectively Union or Member State). Instead, presupposing that the unit in question already has the relevant competence, they merely regulate its exercise for the benefit of the regime as a whole, across the separate spheres of competences. They, thus, ‘regulate the exertion of competences’ (Kompetenzausübungsregulative) and as such they are similar to a principle with a long tradition

120 See BVerfG 1 BvF 1/05 (Treibhausgas-Emissionsberechtigung) (n 107) 79 (Headnote 1a), 95 et seq; see also BVerfG 1 BvR 1916/09 (Le Corbusier) (n 82) 90; BVerfG 1 BvL 3/08 (Investitionszulage) (n 106) 199. 121 Unambiguously in BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 34: ‘If a complainant invokes a violation of the guarantee of human dignity, the Federal Constitutional Court – notwithstanding the [Solange] jurisprudence to date, concerning the inadmissibility of constitutional complaints and submissions by courts which asserted that fundamental rights guaranteed in the Basic Law have been infringed by secondary European Community or Union Law (see BVerfGE 73, 339 [378 et seq]; 102, 147 [161 et seq]) – reviews a grave violation of fundamental rights of this kind by way of identity review (see BVerfGE 113, 273 [295 et seq]; 123, 267 [344, 353–4]; 126, 286 [302–3]; 129, 78 [100]; 134, 366 [384–5 MN 27] …).’ 122 Expressly to the same effect, with regard to the principle of openness toward European law, see Voßkuhle, ‘Integration’ (n 42) 165. 123 For ‘the constitutionally protected viability of the European Union’s legal order’, see BVerfG 07.09.2011, 2 BvR 987/10 (Euro-Rettungsschirm) BVerfGE 129, 124, 172 (BVerfG translation MN 109) ECLI:DE:BVerfG:2011:rs20110907. 2bvr098710, referring to BVerfG 2 BvL 52/71 (Solange I) (n 51) 284; BVerfG 2 BvR 197/83 (Solange II) (n 43) 387; BVerfG 2 BvL 1/97 (Bananenmarktordnung) (n 26) 162–3; BVerfG 2 BvE 2/08 (Lissabon) (n 44) 399.

Germany  149 in BVerfG jurisprudence: ‘federal fealty’ (Bundestreue), also called principle of ‘friendly conduct within the federation’ (bundesfreundliches Verhalten).124 (1) The notorious idea of a ‘cooperative relationship’ between the BVerfG and the ECJ/CJEU first appeared in the Maastricht judgment. The BVerfG uses it to refer to the way it exercises its jurisdiction concerning the application of secondary EC law in Germany, in line with Solange II. Because, and to the extent that, the ECJ ‘guarantees the protection of fundamental rights in each individual case for the entire area of the European Communities’, the Federal Constitutional Court can ‘limit itself to a general guarantee of mandatory standards regarding fundamental rights’.125 Nearly two decades earlier, in the Solange I order, the Second Senate had already emphasised that the ECJ and the BVerfG have a ‘duty to concern themselves in their decisions with the concordance of the two systems of law’, ie, of EC and national law.126

[T]he binding of the Federal Republic of Germany (and of all Member States) by the Treaty is not, according to the meaning and spirit of the Treaties, one-sided, but also binds the Community which they establish to carry out its part in order to resolve the conflict here assumed, that is, to seek a system which is compatible with an entrenched precept of the constitutional law of the Federal Republic of Germany.127

In the Maastricht decision, the Senate does not explicitly refer to this passage. In its earlier decisions, however, the BVerfG placed great importance on the argument that, while the BVerfG and the ECJ have different legal orders as their yardsticks and are concerned with law on different hierarchical positions, they still provide justice in a symbiotic relationship.128 It also shows that the BVerfG had in mind a rule of mutual consideration.129 (2) In the Lisbon judgment, the BVerfG ‘created’130 the ‘principle of openness towards European law’ (Europarechtsfreundlichkeit). On the one hand, this is more extensive than the concept of a cooperative relationship (Kooperationsverhältnis), which controls the ‘European association of constitutional courts’ (europäischer Verfassungsgerichtsverbund),131 since it governs the interpretation and application of the Basic Law by all ‘state authority’ (Staatsgewalt) subject to it. On the other hand, it is also a more concise and ‘forceful’ instrument of control, since specific obligations and prohibitions are derived from it. The Basic Law’s Preamble and Article 23(1) GG constitute a constitutional mandate to realise a united Europe. This is the counterpart to the BVerfG’s longstanding principle of ‘openness to international law’

124 See, eg, Matthias Jestaedt, ‘Bundesstaat als Verfassungsprinzip’, in J Isensee and P Kirchhof (eds), Handbuch des Staatsrechts der Bundesrepublik Deutschland, Vol II, 3rd edn (Heidelberg, Müller, 2004) § 29 MN 73 et seq. Of course, the comparison with federal fealty is misleading insofar as it is an element of the Basic Law in its capacity as ­constitution of the entire state (Gesamtstaat), consisting of both the federal institutions (Zentralstaat) and the Member States (Gliedstaaten; in this case the Länder). By contrast, with regard to the EU, the Basic Law is only a Member State ­(‘gliedstaatliche’) ­constitution, and there is no comprehensive (‘gesamtstaatliche’) constitution. 125 BVerfG 2 BvR 2134 (Maastricht) (n 43) 175. 126 BVerfG 2 BvL 52/71 (Solange I) (n 51) 278. 127 Ibid, 279. 128 See BVerfG 2 BvR 225/69 (Umsatzausgleichsteuer auf eingeführtes Milchpulver) (n 78) 173 et seq; BVerfG 2 BvL 52/71 (Solange I) (n 51) 277–8; BVerfG 2 BvR 197/83 (Solange II) (n 43) 366–7. 129 Particularly apparent in BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 303 (BVerfG translation MN 57): ‘The tensions, which are basically unavoidable according to this construction, are to be harmonised cooperatively in accordance with the European integration idea and relaxed through mutual consideration.’ 130 A recent characterisation by the president of the BVerfG: ‘aus der Taufe gehoben’, Voßkuhle, ‘Integration’ (n 42) 165. 131 Andreas Voßkuhle, ‘Der europäische Verfassungsgerichtsverbund’, 29 Neue Zeitschrift für Verwaltungsrecht (2010) 1.

150  Matthias Jestaedt



(Völkerrechtsfreundlichkeit).132 On a general level, it leads to the obligation ‘for the German constitutional bodies’133 (including the BVerfG), and for ‘German agencies … when they act functionally for the European Union within its institutional organisation’,134 to participate in European integration in an active and constructive manner.135 This includes the – ­constitutional – obligation ‘to observe European Union law’.136 So far, the BVerfG has created specific obligations out of the ‘principle of openness towards European Law’ with regard to two types of situation that, tellingly, concern the Court itself. Both the ultra vires review as well as the review of constitutional identity (for both see below, section V.D.1, 2) ‘may only be exercised in a manner which is open towards European law’.137 On a first level, this means that the legal order of the Basic Law tasks the BVerfG with both types of review to the exclusion of all other German courts.138 For the BVerfG, this means that it has to involve the CJEU ‘as necessary’, which, in turn, is solely competent to interpret EU law with binding effect by way of preliminary-ruling proceedings (Article 267 TFEU). Only then may the BVerfG apply its constitutional standard to EU law instruments in the course of either the ultra vires or the identity review.139 With regard to ultra vires review, the ‘principle of openness towards European law’ also controls the standard of review. The Court can find an act ultra vires only if the breach of competences is sufficiently qualified, namely if

the act of the authority of the European Union [is] manifestly in violation of competences and … highly significant in the structure of competences between the Member States and the Union with regard to the principle of conferral and to the binding nature of the statute under the rule of law.140

(3) The Lisbon judgment is also where the concept of ‘responsibility for integration’ was first mentioned.141 This phrase seems to be less precise and rather ambivalent as to its effect. It is, however, clear that this concept is intended to serve the programme of integration laid down in the ‘openness to integration’ clause of Article 23(1) GG. The president of the BVerfG sums it up as follows:

‘Responsibility for integration’ … signifies the permanent and sustainable acceptance of responsibility in the course of European Integration. This primarily concerns the transfer of sovereign powers and the shape of European decision-making procedures; subsequently,

132 See below, section V.E.1. 133 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 346 (BVerfG translation MN 225). 134 BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 172 (BVerfG translation MN 109). 135 Landmark case BVerfG 2 BvE 2/08 (Lissabon) (n 44) 346–7, quote at 346 (BVerfG translation MN 225). 136 BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 172. 137 BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 303 (BVerfG translation MN 58) referring to BVerfG 2 BvE 2/08 (Lissabon) (n 44) 354. 138 See BVerfG 2 BvE 2/08 (Lissabon) (n 44) 354 (BVerfG translation MN 241); BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 43; BVerfG 2 BvR 2728/13 (OMT (Hauptsacheverfahren)) (n 62) para 155. 139 See BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 303–4; BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 24; BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 46 (the phrase ‘as necessary’ appears here). However, there were several instances where the BVerfG considered a referral not to be necessary with regard to the CJEU’s acte clair doctrine (ECJ 283/81 (CILFIT) (n 92) MN 16–7), most recently in BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 125. 140 BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 286 (Headnote 1.b), 304 (BVerfG translation MN 61); BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 24. See below, section V.D.1. 141 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 351 (BVerfG translation MN 236): ‘a special responsibility is incumbent on the legislative bodies, in addition to the Federal Government, within the context of participation which in Germany, has to comply internally with the requirements under Art 23.1 of the Basic Law (responsibility for integration) and which may be invoked in any proceedings before the Federal Constitutional Court.’

Germany  151 it also touches upon the dynamic development as well as the execution of the Treaties. In short, it means making the integration process a legal process.142



The ‘national responsibility for integration’ in Article  23(1) GG obligates the Federal Government (Bundesregierung) and the federal legislature, as well as the BVerfG itself. Moreover, the BVerfG holds that the European institutions are also bound by this ‘responsibility for integration’ (presumably by EU law), but does not give any legal basis for such an obligation.143 The BVerfG holds that the ‘responsibility for integration’ also includes the duty of the German legislature ‘[f]or borderline cases of what is still constitutionally admissible’,144 to take, ‘where necessary, … precautions in its legislation accompanying approval to ensure that the responsibility for integration of the legislative bodies can sufficiently develop’.145 However, the Court’s jurisprudence is not sufficiently clear on the question whether this ‘responsibility for integration’ is more than a mere abbreviation for what the Constitution prescribes, especially whether it goes beyond Article 23(1) GG, and whether it therefore has a regulatory content of its own which exceeds the specific provisions of the Basic Law.

D.  Arguments Limiting Integration, Namely Promoting Constitutional Identity 1.  Ultra Vires Review The BVerfG expresses the conditionality of the ‘responsibility for integration’ enshrined in Article 23(1) GG in a short formula: [I]t is true that the Basic Law grants the legislature powers to engage in a far-reaching transfer of sovereign powers to the European Union. However, the powers are granted under the condition that the sovereign statehood of a constitutional state is maintained on the basis of an integration programme according to the principle of conferral and respecting the Member States’ constitutional identity, and that at the same time the Member States do not lose their ability to politically and socially shape living conditions on their own responsibility.146

There are two central mechanisms which the BVerfG employs in its recent jurisprudence to determine whether EU legal acts remain within the limits of the authorisation for European integration (Integrationsermächtigung) of Article 23(1) GG. These mechanisms are the so-called ultra vires review (ultra vires-Kontrolle) and the so-called review of constitutional identity ((Verfassungs-)Identitätskontrolle) (on the latter, see below, section V.D.2). It may seem that they are ­unprecedented, but that is not quite the case. At least since the mid-1970s, the BVerfG’s jurisprudence was shaped by three fundamental arguments. First, that EC or EU law applies in Germany only by virtue of the domestic Approving Acts to the European Treaties (in conjunction with the authorisation for European integration in 142 Voßkuhle, ‘Integration’ (n 42) 166. 143 See, eg, BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 31. 144 This refers to authorisations with a blanket character that still comply with the constitutional requirements, or to the participation in drafting dynamic (European) treaty provisions. 145 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 353 (BVerfG translation MN 239); BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 180–1; BVerfG 12.09.2012, 2 BvR 1390/12 (ESM (Aussetzung der Ratifikation I)) BVerfGE 132, 195, MN 105 ECLI:DE:BVerfG:2012:rs20120912a.2bvr139012; BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 31; BVerfG 18.03.2019, 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) [2014] BVerfGE 135, 317, MN 159 ECLI:DE:BVerfG: 2014:rs20140318.2bvr139012. 146 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 347 (BVerfG translation MN 226).

152  Matthias Jestaedt Article 24(1) GG (old version) or Article 23(1) GG (new version)). Second, that the constitutional authorisation for European integration does not allow for the European integration to proceed by way of measures that would violate ‘the basic structure of the Constitution, which forms the basis of its identity’. Third, that acts of EC or EU law have to keep within and must not transgress the limits set by the domestic act of transformation.147 However, while in its earlier decisions the BVerfG did not develop a specific programme for legal review from this distinction, ever since the Lisbon judgment it clearly distinguishes between the two concepts of ultra vires review and review of constitutional identity.148 Also, in earlier decisions the constitutional conditions and restrictions for the ‘openness’ towards European integration were founded solely on the authorisation for European integration of Article 24(1) GG (old version). Since the Maastricht judgment, however, the BVerfG also draws on the principle of democratic legitimation, in conjunction with the ‘eternity clause’ of Article 79(3) GG, to which Article  23(1) sentence 3 GG (new version) expressly refers.149 In the OMT order of 2014, the BVerfG held: [T]he actions of the institutions and agencies of the European Union are democratically legitimated – as far as Germany is concerned – in the legislative Acts of Assent to the Treaty establishing the European Union and the Treaty on the Functioning of the European Union, which were enacted on the basis of Art. 23 sec. 1 GG, and in the programme of integration set out therein. An essential element of this programme of integration is the principle of conferral.150

It concludes that acts of EU institutions and agencies are binding only within the scope of these boundaries in Germany, the adherence to which is subject to ultra vires review and review of constitutional identity. However, in order to preserve the viability of the Community legal order, the mandatory ‘application of constitutional law that is open to European law’ requires that it is incumbent on the BVerfG alone – excluding all other German courts – to declare an act to be ultra vires and a violation of constitutional identity.151 Concentrating both reviews within the BVerfG also ensures they are applied in the spirit of the ‘co-operative relationship’ that the BVerfG has with the CJEU. In the course of its ultra vires review, the BVerfG asks ‘whether acts of European institutions and agencies are based on manifest transgressions of powers’.152 A violation is manifest, or sufficiently qualified (which amounts to the same thing), only if the act of the authority of the European Union [is] manifestly in violation of competences and … highly significant in the structure of competences between the Member States and the Union with regard to the principle of conferral and to the binding nature of the statute under the rule of law.153

In its 2016 OMT judgment, the BVerfG concretised the notion of a ‘shift of competences to the detriment of the Member States’: A shift of competences to the detriment of the Member States can only be found to be present if the exceeding of competences carries considerable weight for the principle of democracy and the sovereignty of the people. This is for instance the case if it is capable of altering the fundamental competences of the 147 Landmark case: BVerfG 2 BvL 52/71 (Solange I) (n 51) 279–80. 148 Landmark case: BVerfG 2 BvE 2/08 (Lissabon) (n 44) 268 (Headnote 4), 353–4 (BVerfG translation Headnote 5). 149 On the fact that this line of argument is largely due to issues of constitutional procedural law, see below, section V.D.4. 150 BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 20 (emphasis added); compare also BVerfG 2 BvR 2728/13 (OMT (Hauptsacheverfahren)) (n 62) para 151. 151 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 354 (BVerfG translation MN 241). 152 BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 22. 153 BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 286 (Headnote 1.a), 304 (BVerfG translation MN 61); BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 24.

Germany  153 European Union and thereby of undermining the principle of conferral. Such a case can be assumed if the exercise of the competence by the institution, body, office, or agency of the European Union were to require a treaty amendment in accordance with Art. 48 TEU or making use of an evolutionary clause (Evolutivklausel ), i.e. in Germany, action on the part of the legislature – be it in accordance with Art. 23 sec. 1 sentence 2 GG, or in accordance with the Act Regarding Responsibility for European Integration (Integrationsverantwortungsgesetz; cf. already Art. 235 TECE former version).154

Thus, the violation of competences by the EU must at the same time be evident and cause structural changes. However, under the constitutional principle of openness to European Law (Europarechtsfreundlichkeit; see above, section V.C.5), the BVerfG is not allowed to declare an Act of an EU institution or agency ultra vires without consulting the CJEU by way of a preliminary reference, thus giving the CJEU an opportunity to rule on the validity and interpretation of the act in question.155 In 2014, the OMT case became the first one to be referred by the BVerfG to the CJEU for a preliminary ruling. Ultimately, this means that for the BVerfG to be able to pronounce the verdict of ultra vires and non-applicability, the CJEU’s preliminary ruling itself would have to be an evident violation of competences causing structural changes. As the BVerfG puts it, EU institutions and bodies would ‘manifestly exceed their transferred competences’ only if ‘the competence cannot be justified under any legal standpoint’. Also, the CJEU ‘has a right of tolerance of error’ which ‘reaches its limit only when an interpretation of the Treaties is manifestly utterly incomprehensible and thus objectively arbitrary’.156 Not least because of this consideration the BVerfG, given EU law’s institutional and procedural safeguards, thinks that ultra vires acts will be ‘marginal cases … to be expected rarely’.157

2.  Review of Constitutional Identity As mentioned before, ever since the Solange I order of 1974, the BVerfG has refused to recognise as law applicable in Germany any Act of primary or secondary EC law ‘which would destroy the identity of the present Constitution of the Federal Republic of Germany by encroaching on the structures which make it up’.158 That argument was based on Article 24(1) GG (old version). However, the review of constitutional identity was only gradually, and not until much later, developed in its current form, starting with the Lisbon judgment.159 When Article 23 GG, the new ‘Article on Europe’ notably modelled on the BVerfG’s Solange jurisprudence, was inserted in 1992, the protection of the so-called ‘constitutional identity’ found an explicit buttress in the text of the Basic Law. Article 23(1) sentence 3 GG reads: ‘The establishment of the European Union, as well as changes in its treaty foundations and comparable regulations that amend or supplement this Basic Law, or make such amendments or supplements possible, shall be subject to paragraphs (2) and (3) of Article 79.’ The so-called ‘eternity clause’ in 154 BVerfG 2 BvR 2728/13 (OMT (Hauptsacheverfahren)) (n 62) para 151. 155 BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 303–4 (BVerfG translation MN 58–61) (in his dissenting opinion, Judge Landau argues that in the light of the Lisbon judgment, rendered by the same Senate one year earlier, the additional qualification first mentioned in Honeywell (ultra vires acts have to cause significant changes to the structure of competences) is an unjustified excess of the ‘requirements on the finding of an ultra vires act on the part of the Community or Union bodies by the Federal Constitutional Court’: ibid, 322–4, quote at 318 (BVerfG translation MN 101–4, quote at MN 95)); BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 24. 156 BVerfG 2 BvR 2728/13 (OMT (Hauptsacheverfahren)) (n 62) para 148–9; see also BVerfG 2 BvR 859/15 (PSPP) (n 63) para 53. 157 BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 26; previously: BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 303 (BVerfG translation MN 57). 158 BVerfG 2 BvL 52/71 (Solange I) (n 51) 279. 159 See BVerfG 2 BvE 2/08 (Lissabon) (n 44) 268 (Headnote 4), 353–4.

154  Matthias Jestaedt Article 79(3) GG forms part of the preconditions and limitations to an amendment of the Basic Law. It stipulates that ‘[a]mendments [of the Constitution] affecting the division of the Federation into Länder, their participation on principle in the legislative process, [and] the principles laid down in Articles 1 and 20 shall be inadmissible.’ Extending the scope of Article 79(3) GG to a transfer of German sovereign powers to the EU, which would change the constitutional structure, means that such a transfer cannot go further than an amendment of the Constitution could. The BVerfG, both in its earlier and in its more recent jurisprudence, assigns a central role as gatekeeper for the recognition of EC and EU law in Germany to the concepts of the ‘identity of the present Constitution’ (Identität der geltenden Verfassung) and the ‘constitutional identity’ (Verfassungsidentität). However, these two reservations on identity do not mean the same thing. This is because in the jurisprudence of the 1970s and 1980s, following the spirit of Solange, there is no reference to the – more specific and more concise – provisions of Article 79(3) GG. Instead, the Court develops the concept of a ‘constitutional structure of the Basic Law’160 consisting of inalienable essential features, which it derives from a systematic interpretation in light of the ‘overall context of the whole Constitution’.161 The ‘eternity clause’ might have formed a backdrop, but it was not mentioned eo nomine and was not needed to establish the features of the inalienable constitutional structure.162 The jurisprudence since Lisbon develops a different notion of ‘constitutional identity’. ‘Constitutional identity’ now appears as a mere ‘abstract’ of the inalienable constitutional features under Article 79(3) GG.163 This is accompanied by a twofold modification, or change of focus, which might be seen both as a limit to and as a reinforcement of the concept of constitutional identity. The new meaning of the concept of ‘constitutional identity’ encompasses only the four legal positions mentioned in Article 79(3) GG. These legal positions are (1) the division of the Federation into Länder, (2) the participation of the Länder in the legislative process, (3) the principles laid down in Article 1, and (4) the principles laid down in Article 20. Compared to the idea of the protection of the constitutional structure in general, this seems like a restrictive reading. However, the explicit reference to Article 79(3) GG focuses (and, one might say, strengthens) the concept of the inalienability of constitutional identity to the point that it is withheld from the discretion even of the constitution-amending legislature. At the same time, it makes clear that the four identity features are not to be balanced against other legal interests in any way that would diminish the scope of their protection.164 160 BVerfG 2 BvL 52/71 (Solange I) (n 51) 280. 161 Ibid, 279. 162 In BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46), the Court summarises its jurisprudence concerning the review of identity. However, it is misleading of the BVerfG to identify ‘the area of constitutional identity, which cannot be transferred’ with Art 79(3) in conjunction with Arts 1 and 20 GG and then to cite Kloppenburg and Maastricht, amongst others, for support. For in none of the three passages cited (BVerfG 2 BvR 687/85 (Kloppenburg) (n 46) 235, 242; BVerfG 2 BvR 2134 (Maastricht) (n 43) 188) does the Court explicitly draw on Art 79(3) GG. The citation of European Arrest Warrant I is even less convincing; for even though the Second Senate explicitly refers to Article 79(3) GG in that judgment (BVerfG 18.07.2005, 2 BvR 2236/04 (Europäischer Haftbefehl I) BVerfGE 113, 274, 296 (BVerfG translation MN 72) ECLI:DE:BVerfG:2005:rs20050718.2bvr223604), this does not happen in the context of Art 23(1) GG, but in order to set the standard of review regarding the constitutionality of Art 16(2) sentence 2 GG, which has been in force since 2000 and allows for the extradition of a German citizen to an EU Member State or to an international tribunal, if this happens on the basis of an enabling statute and provided that the rule of law is observed. 163 Particularly clear: BVerfG 2 BvE 2/08 (Lissabon) (n 44) 354 (BVerfG translation MN 240): ‘the inviolable core content of the constitutional identity of the Basic Law pursuant to Article 23.1 third sentence in conjunction with Article 79.3 of the Basic Law’; BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 22, 27: ‘the area of constitutional identity, which cannot be transferred (Art. 79 sec. 3 in conjunction with Art. 1 and Art. 20 GG)’ or ‘the constitutional identity protected by Art. 79 sec. 3 GG’; likewise BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 302 (BVerfG translation MN 55) (with further references): ‘the area of constitutional identity which is not assignable (Article 79.3 in conjunction with Article 1 and Article 20 of the Basic Law)’. 164 See BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 29.

Germany  155 As in the case of ultra vires review, the BVerfG argues commensurately with its model of the separate but co-ordinated jurisdictions of the CJEU and the BVerfG, a model which it has followed throughout its decisions on European integration: In their co-operative relationship, it is for the Court of Justice to interpret the act [of Union Law which gave rise to the control of constitutional identity]. On the other hand, it is for the Federal Constitutional Court to determine the inviolable core content of the constitutional identity, and to review whether the act (in the interpretation determined by the Court of Justice) interferes with this core.165

Consequently, and in light of the principle of openness toward European law,166 the BVerfG has to consult the CJEU and enquire about its interpretation of the EU legal act via the preliminary ruling proceeding (Article 267 TFEU). Only then may the BVerfG itself decide on the matter.167 Therefore, the impact on and relevance for the legal practice of reviewing the constitutional identity based on Article 23(1) sentence 3 GG depends on two factors. First, the BVerfG must have requested a preliminary ruling (Article 267 TFEU) and with it an authoritative interpretation of an Act of primary or secondary EU law by the CJEU. At this point, the CJEU is able to influence Karlsruhe’s subsequent decision. Second, it depends on how extensively the BVerfG interprets the core of constitutional identity in Article  79(3) GG. In the jurisprudence since Lisbon, the Second Senate clearly shows a tendency to interpret the substantive content of the so-called ‘eternity-clause’ in an extensive manner. Without claiming to be exhaustive, certain features of constitutional identity emerge from the BVerfG’s jurisprudence. (1) Following the mainstream reading of Article 79(3) GG, the BVerfG holds that ‘the structural principles of the state laid down in Article 20 of the Basic Law, ie democracy, the rule of law, the principle of the social state, the republic, the federal state, as well as the substance of elementary fundamental rights indispensable for the respect of human dignity are, in any case, not amenable to any amendment because of their fundamental quality.’168 (2) Given that Article  79(3) GG ‘even prevents a constitution-amending legislature from disposing of the identity of the free constitutional order’, the BVerfG comes to the conclusion that the ‘eternity guarantee’ also ensures a ‘sovereign statehood for Germany’.169 The BVerfG further offers the idea that the Basic Law is not founded on ‘a self-serving and selfglorifying concept of sovereign statehood’ but that sovereign statehood ‘stands for a pacified area and the order guaranteed therein on the basis of individual freedom and collective self-determination’.170 165 Ibid, MN 27. 166 See BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 43. 167 In deciding whether Community law complies with the constitutional identity according to Art 23(1) sentence 3 GG in conjunction with Art 79(3) GG, the BVerfG regards preliminary proceedings under Art 267 TFEU not to be necessary if there is no reasonable doubt about the interpretation and application of the Act of Community law under the CILFIT or acte clair doctrine of the ECJ (ECJ 283/81 (CILFIT) (n 92)); BVerfG 1 BvR 1215/07 (Antiterrordatei) (n 58) MN 90–1; BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 125. In its judgment Antiterrordatei, the BVerfG held that it was under no obligation to request a preliminary ruling because of an acte clair. It arrived at this conclusion by way of a clearly restrictive interpretation of the ECJ’s decision in the Åkerberg Fransson case (CJEU C-671/10 (Åkerberg Fransson) (n 56)): ‘As part of a co-operative relationship between the Federal Constitutional Court and the European Court of Justice …, this decision must not be read in a way that would view it as an apparent ultra vires act or as if it endangered the protection and enforcement of the fundamental rights in the Member States (Art. 23 sec. 1 sentence 1 GG) in a way that questioned the identity of the Basic Law’s constitutional order’ (MN 91). 168 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 343 (BVerfG translation MN 217). 169 Ibid, 343 (BVerfG translation MN 216). 170 Both quotes: ibid, 346 (BVerfG translation MN 223, 224); The BVerfG uses the term ‘sovereign statehood of a constitutional state’ (Verfassungsstaatlichkeit) for example ibid, 347 (BVerfG translation MN 226) (emphasis added). In BVerfG 2 BvR 2236/04 (Europäischer Haftbefehl I) (n 162) and with reference to BVerfG 2 BvR 2134 (Maastricht) (n 43) 182–3, the

156  Matthias Jestaedt (3) If, therefore, Article 79(3) GG guarantees and precludes constitutional amendments of the basic federal structure as a whole (not just the two elements expressly mentioned in Article  79(3) GG, ie, the division of the Federation into Länder and the principle of the Länder participating in the legislative process)171 and if it therefore guarantees ­‘sovereign statehood for Germany’, a European federal state is probably incompatible with Article 79(3) GG.172 (4) In the same manner, the Basic Law disallows transferring Kompetenz-Kompetenz, ie, ‘to transfer sovereign powers in such a way that their exercise can independently establish other competences for the European Union’,173 as well as a ‘blanket empowerment for the exercise of public authority’.174 According to the BVerfG, the Constitution is, however, clearly compatible with a form of integration that builds on an acquis communautaire and interprets competences along the lines of the implied-powers doctrine or the principle of effet utile.175 (5) Neither the transfer of sovereign powers, nor the elaboration of the Union authority acting autonomously is allowed to result in ‘the system of democratic rule in Germany being undermined’. In particular, the German Bundestag must retain its ‘own responsibilities and competences of substantial political importance’, and the federal Government, which politically answers to Parliament, must be in a position to exert a decisive influence on European decision-making procedures.176 This is not to mean that the ‘safeguarding of sovereignty, demanded by the principle of democracy in the valid constitutional system prescribed by the Basic Law in a manner that is open to integration and to international law’ necessitated ‘a pre-determined number or certain types of sovereign rights [to] remain in the hands of the state’.177 Yet the BVerfG holds that

European unification on the basis of a treaty union of sovereign states may, however, not be achieved in such a way that not sufficient space is left to the Member States for the political formation of the economic, cultural and social living conditions. This applies in particular to areas which shape the citizens’ living conditions, in particular the private sphere of their own responsibility and of political and social security, protected by fundamental rights, as well as to political decisions that rely especially on cultural, historical and linguistic perceptions and which develop in public discourse in the party political and parliamentary sphere of public politics.178

Second Senate speaks of a potential development, inadmissible under Art 79(3) GG, if ‘the legal system that is established by the Basic Law [were to] los[e] the core elements of statehood [Entstaatlichung]’ [298] [BVerfG translation MN 76]). 171 Art 23(1) sentence 3 GG refers to the ‘eternity clause’ of Art 79(3) GG in a comprehensive fashion, ie, encompassing every one of its guarantees. However, the BVerfG seems to favour a restrictive reading of the constitutional text, as it reduces the ‘constitutional identity protected under Article  79(3) GG’ to ‘the principles laid down in Article  1 and 20 GG’, omitting the first two elements (the division of the Federation into Länder and the participation of the Länder in the legislative process); BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 42. 172 Effectively arguing in this manner: BVerfG 2 BvE 2/08 (Lissabon) (n 44) 347–8, 364–5. 173 Ibid 349 et seq, quote at 349 (BVerfG translation MN 233). See also the leading case BVerfG 2 BvR 2134 (Maastricht) (n 43) 194–5. 174 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 351; see also, eg, BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145) MN 160. 175 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 351. 176 Both quotes: BVerfG 2 BvE 2/08 (Lissabon) (n 44) 356 (BVerfG translation MN 244, 246). See also the leading case BVerfG 2 BvR 2134 (Maastricht) (n 43) 184–5, 207–8. 177 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 357 (BVerfG translation MN 248). 178 Ibid, 358 (BVerfG translation MN 249).

Germany  157

The BVerfG mentions as

[p]articularly sensitive for the ability of a constitutional state to democratically shape itself … decisions on substantive and formal criminal law (1), on the disposition of the monopoly on the use of force by the police within the state and by the military towards the exterior (2), fundamental fiscal decisions on public revenue and public expenditure, the latter being particularly motivated, inter alia, by social policy considerations (3), decisions on the shaping of living conditions in a social state (4) and decisions of particular cultural importance, for example on family law, the school and education system and on dealing with religious communities (5).179

(6) The BVerfG holds that the citizen’s right to free and equal participation in public authority (Article 38(1) sentence 1 GG) is anchored in human dignity (Article 1(1) GG) and belongs to the foundations of German constitutional law that are unchangeable due to the principle of democracy (Article 20(1) and (2) GG with Article 79(3) GG).180 From this it deduces a number of minimum standards for free and equal participation, which are guaranteed under Article 79(3) GG. (7) A necessary condition for the safeguarding of political latitude in the sense of the core of identity of the Constitution (Article 20 (1) and (2), Article 79 (3) of the Basic Law) is that the budget legislature makes its decisions on revenue and expenditure free of other-directedness on the part of the bodies and of other Member States of the European Union and remains permanently ‘the master of its decisions’ … [I]t follows from the democratic basis of budget autonomy that the Bundestag may not consent to an intergovernmentally or supranationally agreed automatic guarantee or performance which is not subject to strict requirements and whose effects are not limited, which – once it has been set in motion – is removed from the Bundestag’s control and influence.181 Moreover, no permanent mechanisms may be created under international treaties which are tantamount to accepting liability for decisions by free will of other states, above all if they entail consequences which are hard to calculate. The Bundestag must individually approve every large-scale federal aid measure on the international or European Union level made in solidarity resulting in expenditure.182 (8) According to the BVerfG ‘the interests protected by the [principle of] constitutional identity enshrined in Article 79(3) GG, which are protected also against encroachment by supranational public authority’, include ‘the principles of Article  1 GG, ie the duty of all state authority to respect and to protect human dignity (Article 1(1) sentence 2 GG), but also the principle that any sanction presupposes guilt, which forms part of the guarantee of human dignity under Article 1(1)GG.’183 Therefore, if somebody were to be extradited in order to carry out a foreign criminal judgment rendered in absentia, even though the person was 179 Ibid, 359 (BVerfG translation MN 252). 180 Ibid, 341 (BVerfG translation MN 211); affirmed, eg, in BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145) MN 159 with further references. 181 BVerfG 2 BvR 1390/12 (ESM (Aussetzung der Ratifikation I)) (n 145) MN 109 (BVerfG translation MN 213), referring to BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 180; see also BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145) MN 122 et seq, 161; BVerfG 2 BvR 2728/13 (OMT (Hauptsacheverfahren)) (n 62) para 138. 182 BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 28, referring BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 177 et seq; BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145) MN 161. See also BVerfG 2 BvR 1390/12 (ESM (Aussetzung der Ratifikation I)) (n 145) 135 et seq, esp 149 (BVerfG translation MN 239 et seq, esp 253) about a specific limit for the total amount of all payment obligations of the Federal Republic of Germany. 183 BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 48 (further developed at MN 53 f) referring to BVerfG 2 BvE 2/08 (Lissabon) (n 44) 413.

158  Matthias Jestaedt not informed about the proceedings and their conclusion and even though that person had no effective means to retroactively obtain a fair hearing and effectively defend himself, this would make the said person a mere object of state proceedings and thus constitute a violation of Article 1(1) GG.184 The BVerfG undoubtedly adopts a very broad interpretation of constitutional identity guaranteed under Article 79(3) GG. Still, in order to do justice to the Karlsruhe jurisprudence, one should consider that in the specific application of that principle, the BVerfG is keen to find compatible, as far as possible, the inalienable guarantee of constitutional identity with EU law (as applied).185 With regard to the Basic Law’s openness toward EU law, the BVerfG also sets ‘strict conditions for activating the review of constitutional identity’. In turn, this means ‘increased admissibility requirements’ for constitutional remedies.186 Only a superficial consideration ignoring that the BVerfG has to satisfy numerous and contradictory claims with its approach would qualify this approach as all-bark-and-no-bite.

3.  Mobilising Parliament Since the Maastricht judgment – and a fortiori since the Lisbon judgment, where the Maastricht arguments were repeated in more forceful language – the BVerfG’s jurisprudence on European integration has been dominated by an attempt to ensure the democratic legitimacy of the integration process. In the EU, which the Court calls a ‘union of democratic states’ (Verbund demokratischer Staaten), it is ‘first and foremost for the national peoples of the Member States to provide democratic legitimation [of the exercise of sovereign powers] via their national parliaments’.187 This makes the German Bundestag the focus of the Court’s attempts to ensure constitutional legality for the process of European integration. Not only must the Bundestag retain substantially important responsibilities and competences,188 but it is also considered the linchpin between the EU and its Member State Germany as far as legitimacy is concerned. This also means that the Bundestag has a certain ‘responsibility for integration’ (Integrationsverantwortung).189 This responsibility enjoins it to ensure that the programme of integration in Article 23(1) GG is executed, ie, that the conditions in that provision of the Constitution are observed. Together with the Bundesrat, the Bundestag is the federal legislature (including the sole power to make constitutional law, Article 79(2) GG). Thus, it is dependent on its decisions as to which sovereign rights are transferred to the EU, how this is done, and which must be the safeguards (Article 23(1) sentences 2–3 GG). This protection against an emptying of the competences and duties of Parliament is essentially a constitutional safeguard against parliamentary self-restraint. The Bundestag is protected inter alia against transferring its own responsibility to other actors through vague authorisations.

184 See BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 60. 185 The extremely detailed argument in BVerfG 2 BvR 1390/12 (ESM (Aussetzung der Ratifikation I)) (n 145) MN 126–215 and BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145) MN 176–245 is paradigmatic in seeking to achieve compatibility between the ESM-Treaty and the TSCG with the provisions of the constitution. 186 Both quotes: BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 50. 187 BVerfG 2 BvR 2134 (Maastricht) (n 43) 155 (Headnote 3a), see also 184, 185, 187–8. 188 See BVerfG 2 BvR 2134 (Maastricht) (n 43) 171–2; BVerfG 2 BvE 2/08 (Lissabon) (n 44) 330–1, 340–1; BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 167 et seq, 177 et seq; BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 18. 189 See above, Section V.C.5.

Germany  159 In particular, the BVerfG has repeatedly highlighted the ‘overall budgetary responsibility’ (haushaltspolitische Gesamtverantwortung)190 of Parliament (protected as it is by the ‘control of constitutional identity’191): A necessary condition for the safeguarding of political latitude in the sense of the core of identity of the Constitution (Article 20(1) and (2), Article 79(3) of the Basic Law) is that the budget legislature makes its decisions on revenue and expenditure free of other-directedness on the part of the bodies and of other Member States of the European Union and remains permanently ‘the master of its decisions’.192

The Second Senate, in its ESM judgment of 2014, shows how far-reaching and yet specific the duties incumbent on the Bundestag can be: [N]o permanent mechanisms may be created under international treaties which are tantamount to accepting liability for decisions of other states, above all if they entail consequences which are hard to calculate. The Bundestag must individually approve every large-scale federal aid measure on the international or European Union level that was made in solidarity and results in expenditure. Insofar as supranational agreements are entered into, which due to their scale may be of structural significance for parliament’s right to decide on the budget, for example by giving guarantees the honouring of which may endanger budget autonomy, or by participating in similar financial safeguarding systems, not only does every individual disposal require the Bundestag’s approval; it must also be ensured that there is sufficient parliamentary influence on the way the funds provided are used.193

It would seem, from this mobilisation of the Bundestag that the BVerfG wishes to push fundamental questions of European (integration) policy towards the political arena, with Parliament as the centre of debate, going against consensus opinion.

4.  Mobilising the Citizen Protecting the Basic Law against legal acts which are either ultra vires or violate the constitutional identity would be impossible, however, if the BVerfG did not have sufficient opportunities to pronounce on these matters. A political force has not yet emerged in Germany (and is unlikely to emerge in the foreseeable future) which is able, strong enough, and willing to initiate ‘abstract’ reviews of statutes (abstrakte Normenkontrolle) against transformation statutes concerning the European treaties, which could be interpreted as German criticism of European integration. Largely because this is so, the BVerfG, beginning with the Maastricht judgment of 1993,194 has mobilised ‘electoral citizens’ (Wahlbürger) in order to enable them to protect the inalienable democratic substance of the Basic Law (unaufgebbare demokratische Substanz des Grundgesetzes).195 The right to vote is enshrined as a quasi-basic right in Article 38(1) sentence 1 GG, actionable by individual constitutional complaint (Verfassungsbeschwerde; Article 93(1) No 4a GG). Here, 190 See BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 177 et seq, esp 179–80 (BVerfG translation MN 124); BVerfG 28.02.2012, 2 BvE 8/11 (EFSF-BT-Sondergremium) BVerfGE 130, 318, 344–5, 345 et seq ECLI:DE:BVerfG:2012:es20120228. 2bve000811; BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145) MN 161 et seq. 191 See above, section V.D.2. 192 BVerfG 2 BvR 1390/12 (ESM (Aussetzung der Ratifikation I)) (n 145) MN 109–10 (BVerfG translation MN 213). 193 BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145) MN 165 (BVerfG translation MN 165), referring to BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 180–1; BVerfG 2 BvR 1390/12 (ESM (Aussetzung der Ratifikation I)) (n 145) MN 110. 194 Landmark case: BVerfG 2 BvR 2134 (Maastricht) (n 43) 171–2, 182–3; the two other central cases on this point are: BVerfG 2 BvE 2/08 (Lissabon) (n 44) 328–9, 440–1; BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 167 et seq, 177 et seq. 195 In the following cases, Art  38(1) sentence 1 GG has enabled review by the BVerfG in the first place (or at least influenced it): BVerfG 2 BvR 2134 (Maastricht) (n 43); BVerfG 2 BvR 1877/97 (Euro) (n 27) 350–1; BVerfG 2 BvE 2/08

160  Matthias Jestaedt it is substantivised and, thus, gives the BVerfG jurisdiction to exercise judicial review. The Court holds that the right to vote in elections to the German Bundestag simultaneously provides citizens with a guarantee against an emptying of the competences and duties of Parliament by way of Article 23 GG which would violate the democratic principle in its ‘inviolable’ core (Article 79(3) in combination with Article 20(1)–(2) GG).196 Beyond the express constitutional right to vote in elections to the Bundestag, this reading of Article 38(1) sentence 1 GG argues that voters also have a right for the parliamentary chamber which they elect, the Bundestag, to have a certain range of powers to take substantive decisions. The Court speaks in this connection of the ‘guarantee of effective popular government’197 and a ‘right to democracy’.198 In the OMT order, the BVerfG further concretised the ‘substantive content of what is guaranteed by the right to vote’ enshrined in Article 38(1) sentence 1 GG – which it had earlier described as ‘essential democratic content of Article 38 GG’.199 The right to vote is violated only, but always so, if this right is in danger of being rendered ineffective in an area that is essential for the political self-determination of the people, i.e. if the democratic self-government of the people – through the German Bundestag – is permanently restricted in such a way that central political decisions can no longer be made independently.200

The ‘core of parliamentary rights in [a] democracy’ protected by Article  38(1) sentence 1 GG includes the budgetary autonomy of Parliament.201 The right to vote in Article 38(1) sentence 1 GG is violated inter alia when ‘the sustained (long-term) budgetary autonomy of the German Bundestag’ is being eroded.202 The crux of the matter, the Court holds, is ‘the ability of a constitutional state to democratically shape itself ’,203 which is protected by Article  20(1)–(2) in conjunction with Article  79(3) GG even vis-à-vis the constitutional legislator.204 The electoral citizen can also demand, as part of complying with the constitutional principle of democracy, ‘respect of the constituent power of the people’. In that context, the electoral citizen can even bring ‘the loss of statehood of the Federal Republic of Germany’ before the Court.205 However,  this (Lissabon) (n 44); BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123); BVerfG 2 BvR 1390/12 (ESM (Aussetzung der Ratifikation I)) (n 145); BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46); BVerfG 2 BvE 2/13(3%-Sperrklausel bei EP-Wahlen) (n 92); BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145). 196 See BVerfG 2 BvR 2134 (Maastricht) (n 43) 171–2; BVerfG 2 BvE 2/08 (Lissabon) (n 44) 330–1, 340–1; BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 167 et seq, 177 et seq; BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 18. The BVerfG argues in an analogous manner for the status of parliamentarians which is protected by the constitution in Art 38(1) sentence 2 GG, see, eg, BVerfG 2 BvE 8/11 (EFSF-BT-Sondergremium) (n 190) 340, 342 et seq. 197 BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 168 (BVerfG translation MN 98). 198 BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145) MN 125. 199 BVerfG 2 BvR 2134 (Maastricht) (n 43) 179. 200 BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 19, referring to: BVerfG 2 BvR 2134 (Maastricht) (n 43) 172; BVerfG 2 BvE 2/08 (Lissabon) (n 44) 330; BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 168. 201 BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 170 (BVerfG translation MN 104). 202 Ibid, 167 (BVerfG translation MN 97). 203 BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145) MN 161, referring to: BVerfG 2 BvE 2/08 (Lissabon) (n 44) 359; BVerfG 2 BvR 1390/12 (ESM (Aussetzung der Ratifikation I)) (n 145) MN 106. In BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 170 (BVerfG translation MN 104), the Second Senate mentions the necessity ‘to preserve the democratic freedom of action for the body politic in the long term.’ 204 See, eg, BVerfG 2 BvR 1390/12 (ESM (Aussetzung der Ratifikation I)) (n 145) MN 104. 205 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 340, 331 (BVerfG translation MN 208, 167). The (correct) holding in Maastricht that the final article of the Basic Law (Art 146 GG) is not a subjective individual right in the sense employed by Art 93(1) No 4a GG (BVerfG 2 BvR 2134 (Maastricht) (n 43) 180) is countered by the Lisbon judgment by the argument that ‘[t]he fact that Article 146 of the Basic Law does not found an independent, individual right, which is claimable and therefore capable of founding a constitutional complaint … does not rule out that a violation of Article 146 of the Basic Law in conjunction with the fundamental rights, and rights equivalent to fundamental rights, listed in Article  93.1 No 4a of the Basic Law – here Article 38.1 first sentence of the Basic Law – may be alleged’ (BVerfG 2 BvE 2/08 (Lissabon) (n 44) 332) (BVerfG translation MN 180). This is unconvincing, seeing that the same senate in the same judgment (Lissabon)

Germany  161 right (as enshrined in Article  38(1) sentence 1 GG in conjunction with Articles  20(1)–(2) and 79(3) GG) also has its limits. In particular it does not go as far as entitling the electoral citizen to have the Court reviewing ‘the lawfulness of democratic majority decisions’,206 ‘[p]arliament’s internal, functional allocation of responsibilities’, the rules on legislative procedure,207 or even the ‘pre-legal … constitutional requirements’ ‘that the decisional [parliamentary procedures] and the policy objectives pursued in each case be generally perceptible and comprehensible’.208

E.  Contextualising the Federal Constitutional Court’s Jurisprudence on the EU Since the Maastricht judgment at the latest, the BVerfG’s decisions on questions of European integration have been keenly observed and met with particular hopes and expectations, if not also with concern and disapproval. This has been so in particular in German domestic debate, by other Member States, and by the European institutions (not least, the CJEU). Since the BVerfG has been reading strict limitations on integration into the Basic Law, haloed as they are by the ‘eternity clause’, the Court is confronted with the question, à propos nearly every judgment, whether this will now be the occasion where a rift will open vis-à-vis the CJEU. This is not the place to give a detailed description of the complex and multi-layered discussion and to take a stand on it. Still, it might be very helpful, if not imperative for a sound evaluation of the Karlsruhe jurisprudence,209 to take two aspects into account. First, there is the question of how the BVerfG’s decisions on the EU relate to the rest of the Court’s jurisprudence (see below, section V.E.1). Second, one might ask whether the BVerfG makes the EU view its own and how it does so (see below, section V.E.2). However, before discussing these two questions, it might be helpful to point to a third aspect which might help to better understand (if not necessarily to agree with) the Court’s jurisprudence. The BVerfG expressly does not discuss the constitutional limits to European integration with regard to most of the CJEU’s tools to reinforce the effectiveness of EC and EU law, be they ‘primacy of application’, effet utile or the interpretation of national law in conformity with EU law. The BVerfG objects to none of these acts of judicial law-making by the ECJ/CJEU, which deeply affect domestic law. Instead, it uses the ‘limits of European integration’ only where the extension of the supranational protection of fundamental rights narrows the scope of domestic fundamental rights (Solange I, Solange II, Banana Market Organisation, Honeywell, Counter-Terrorism Database, European Arrest Warrant II). The Court will also use this tool in cases where restrictions of domestic democracy is alleged – foremost concerning budgetary measures (Maastricht, Euro, Lissabon, Euro Rescue Package, EFSF, ESM) – as well as in cases where the loss of Germany’s statehood is alleged (Maastricht, Lissabon). holds (although based on an erroneous reading of Art 146 GG) that ‘[l]ike Article 38.1 first sentence of the Basic Law, Article 146 of the Basic Law creates a right of participation of the citizen entitled to vote.’ (BVerfG translation MN 179) If Art 146 GG does create an autonomous right beyond Art 38(1) sentence 1 GG, whose alleged violation, however, can undisputedly not be used to base an individual constitutional complaint, then why and how can the jurisdictional basis for such a complaint be founded ‘in conjunction with’ Art 38(1) sentence 1 GG? 206 BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 19. 207 BVerfG 2 BvR 1390/12 (ESM (Hauptsacheverfahren)) (n 145) MN 125 et seq (BVerfG translation MN 127). 208 BVerfG 2 BvR 1877/97 (Euro) (n 27) 369 (BVerfG translation MN 84). 209 It is not ignored here, when the phrase ‘Karlsruhe jurisprudence on EU matters’ is used, that this suggestion of homogeneity is not without problems, given that there are two independent senates with their own approaches, a large number of rulings over a period that spans half a century, and that the composition of these judicial bodies issuing judgments has changed over the years.

162  Matthias Jestaedt The three constellations each reflect, compared to other Member States, a specific angst of loss in Germany. First, in probably no other EU Member State are both the standard of protection of fundamental rights and the importance of the Constitutional Court as high and their (symbolic) relevance as pronounced as they are in Germany. Second, Germany’s fiscal and budgetary soundness, together with the sheer amount of its potential liabilities vis-à-vis the financing requirements (within and) of the EU, are unparalleled among EU Member States. Under both aspects, the Federal Republic has little to gain from ‘More Europe’, but much to lose if the BVerfG ceases to be the guarantor of legality. Third, the Court’s claim that it needs to protect Germany’s statehood is perhaps best understood by recalling that the concept of ‘nation’ is disavowed here and marks a void. Conversely, this might also explain why the majority of EU Member States, which have an intact national identity and hope for a rich domestic dividend from an extension of the CJEU’s protection of fundamental rights and from more communalisation of financial resources, remain calm with respect to precisely those points that cause alarm at the BVerfG.

1.  The Internal Perspective If one compares the BVerfG’s decisions on European integration with its other jurisprudence, it may be noted that the decisions (perhaps with the exception of the Lisbon judgment) do not deviate markedly. The Court neither takes recourse to a special method on questions of European integration, nor does it develop a special doctrine (Sonderdogmatik) in the sense that it operates with alien arguments or with a deviating arrangement of presentation. In principle, the ‘sound’ is the same. Both when it writes arguments promoting integration as well as limiting integration, the BVerfG brings to bear its far-reaching substantive-teleological interpretive approach – wellknown, for example, from its fundamental-rights jurisprudence and praised by many. A critique of the BVerfG’s method and doctrine will, thus, have to be consistent insofar as it needs to avoid one-sidedly singling out the aspects limiting integration. A doctrinal fit (dogmatische Passung) beyond questions of method exists with regard to the Court’s jurisprudence on the relationship between the constitutional order and public international law, particularly the ECHR. Here, too, the BVerfG operates with doctrinal constructions that are not explicitly part of the constitutional text but which nonetheless significantly determine the ECHR’s role and rank in the constitutional order.210 Here, too, the relationship is determined to a significant degree by the Basic Law’s openness to public international law (Völkerrechtsfreundlichkeit), the ‘elder sibling’ of the openness to European law (Europarechtsfreundlichkeit).211 Both serve to avoid collisions between national and international law in a comparable manner by imposing on national constitutional law (and on the Constitutional Court) a duty (albeit limited) to consider non-national (ie, international and European) law. Such ‘consideration’ can mean, if necessary, that even the Basic Law may need to be interpreted in conformity with supranational and international law.212 These two conclusions do not, however, preempt the objection that the Court’s jurisprudence on European integration operates with an arsenal of doctrinal tools that cannot easily be traced 210 See for a recent discussion on the foundation, scope and content of the principle of the Basic Law’s openness to public international law BVerfG 2 BvL 1/12 (Treaty Override) (n 17) MN 64–76, with further references. On the privileged status granted to the ECHR in comparison with other international treaties by the constitutional jurisprudence see above, section IV.A. 211 Eg, BVerfG 2 BvR 2365/09 (Sicherungsverwahrung II) (n 16) 366, 369–70. 212 For the ECHR: BVerfG 2 BvR 2365/09 (Sicherungsverwahrung II) (n 16) 366–7, for the limits see 371–2.

Germany  163 back to the Basic Law. To be perfectly clear: legal arguments limiting integration as well as those promoting integration incur criticism under methodical and doctrinal auspices. Such criticism is, thus, neither an expression of political sympathy, nor of antipathy to the EU. This can be exemplified regarding five central doctrinal concepts. (1) The cooperative relationship between the ECJ/CJEU and the BVerfG, developed in the Maastricht judgment, is unclear. In particular, how is it founded, what is its legal nature, what consequences arise, what duties does it create, what does it authorise? (2) The ‘responsibility for integration’, as developed in the Lisbon judgment, is no less ambiguous (going on meaningless, perhaps?). Can it be deduced from the Constitution and, if so, how can it possibly bind the other EU Member States? If, however, it is inherent in the European treaties, then how is the BVerfG’s reliance on it in matters of German constitutional law justified? Further questions can be added. (3) What about the justification for the BVerfG’s development of the concept of ‘openness to European law’? What is its legal nature? Is it an expression of loyalty to the EU, which is to be found in the Constitution? The reference to the older, ‘multi-use’ principle of the Basic Law’s openness to public international law does not alleviate doubts as to its constitutionality but reinforces them. (4) Authors do not doubt (this is emphasised here) that the much-maligned ‘review of constitutional identity’ exists in light of the clear wording of Article 23(1) sentence 3 GG. Persuasive criticism may, however, be brought against the way in which this test is applied, ie, against the extensive interpretation of the ‘eternity clause’ according to Article 79(3) GG. The pouvoir constituant had included this norm in the Basic Law as a bulwark against totalitarian temptations. In particular, it has been (re)interpreted as a general guarantee of ‘constitutional identity’. Thus, the (limited) idea of identity is expanded and emancipated as a constitutional guarantee.213 (5) Finally, the BVerfG derives a remarkable cascade of constitutional-doctrinal consequences from the electoral citizen’s right to elect the members of the German Bundestag in general, direct, free, equal and secret elections (Article 38(1) sentence 1 GG). The deduced guarantees include a ‘right to democracy’, a right to the maintenance of substantial competences of Parliament, to the preservation of the Bundestag’s ‘overall budgetary responsibility’, to the ‘ability of a constitutional state to democratically shape itself ’, to the ‘respect of the constituent power of the people’, and to the guarantee of Germany’s ‘statehood’ – all of which are actionable by constitutional complaint. The BVerfG’s underlying intention to make fundamental legal questions justiciable is certainly understandable. However, that goal justifies neither the Court’s doctrinal strategy of ‘everything-is-connected-to-everything’, nor the stylisation of the right to vote into a democratic all-inclusive right.

2.  The External Perspective As demonstrated above, the BVerfG’s jurisprudence on the European integration is subject to some well-founded criticism. However, we surely cannot blame it for its persistent, earnest, almost notorious struggle for the establishment of practical concordance between the demands, on the one hand, of national constitutional law, and, on the other, those of EC or EU law (one could even argue that this is its primary mark of distinction). In doing so, it acknowledges the independence 213 For an attempt to justify an extensive conception of constitutional identity going beyond Art 79(3) GG and being founded independently of it, see Wischmeyer, ‘Nationale Identität’ (n 1).

164  Matthias Jestaedt of both legal strata and emphasises a strong reading of ‘law’ and ‘legal obligation’.214 The Court faces up to this ‘difficult balancing act’215 in a courageous, prudent, professional and sustainable manner. This effort to achieve compatibility, which regularly appears as a strategy to avoid conflict, is reflected in at least five strategic aspects across all decisions: (1) The BVerfG always emphasises both the elements of the Basic Law that promote and that limit integration. Nevertheless, its jurisprudence is characterised by the effort to accentuate the harmonious embedding of the Basic law into the EU’s legal order. In individual cases, the BVerfG tries to find an interpretation that renders the disputed EC or EU legal rules compatible with the Basic Law. One could mention, pars pro toto, the meticulous and detailed explanations concerning the compatibility of the EU’s legal regulations on the financial and public-debt crisis with the Bundestag’s budgetary autonomy, as guaranteed by the Basic Law. It was largely this style of argument that led the BVerfG to be criticised as a dog that barks at the ECJ/CJEU but does not bite in an emergency – an accusation somewhat too simplistic in light of the challenges the Court faced.216 (2) On the contrary, the BVerfG frequently emphasises – in particular in its more recent decisions – that the reservations on constitutional identity which it has dissected from the Basic Law do not call the very existence of EU law into question. It also emphasises that the constitutional law and constitutional courts of other EU Member States as well as EU law itself (particularly Article 4(2) TEU) are considerate of the Member States’ national identity or (which is not necessarily the same) of their constitutional identity.217 Without saying so explicitly,218 the BVerfG seems to proceed from the assumption that something like a common constitutional law of the EU and its Member States exists – not qua sources of law, but in a substantive sense. The BVerfG feels encouraged by the fact that the CJEU expresses a view similar to its own for the EU legal order. In the Lisbon decision, the Second Senate referred to the 2008 ECJ judgment in the Kadi case,219

according to which an objection to the claim of validity of a United Nations Security Council Resolution may be expressed citing fundamental legal principles of the Community … The Court of Justice has thus, in a borderline case, placed the assertion of its own identity as a legal community above the commitment that it otherwise respects. Such a legal construct

214 Significantly, the current president of the BVerfG considers his Court’s contribution to European integration to be an ‘integration through law’ (Voßkuhle, ‘Integration’ (n 42)). In this paper, he talks about, inter alia, ‘the danger of a departure from the law’ (165), the ‘making the integration process a legal process’ (166) and the ‘belief in the integrating power of a European legal community’ (167). 215 Voßkuhle, ‘Integration’ (n 42) 165, mentions a ‘difficult balancing act’ with regard to the realisation of the idea of a legal community through the Member States. 216 See Joseph HH Weiler, ‘The “Lisbon Urteil” and the Fast Food Culture’ (2009) 20 European Journal of International Law 505: ‘in its internationally-related case law, the German Constitutional Court has a well-earned reputation of the Dog that Barks but does not Bite’. 217 See, in particular, BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 44, 46 and particularly 47, on the jurisprudence limiting the Union law’s primacy of application on the basis of national constitutional law with references from a further 10 EU Member States: Denmark (Højesteret), Estonia (Riigikohus), France (Conseil constitutionnel, Conseil d’État), Ireland (Supreme Court of Ireland), Italy (Corte Constituzionale), Latvia (Satversmes tiesa), Poland (Trybunal Konstytucyjny), Spain (Tribunal Constitucional), Czech Republic (Ústavní Soud), United Kingdom (High Court, UK Supreme Court); see also the earlier case, BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 30. 218 The quadripartite description of the EU as an ‘association of states, constitutions, administrations and jurisprudences’ recently coined by the Court could even be seen as an explicit confirmation (BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 44). 219 See ECJ 03.09.2008 Joined Cases C-402/05 P and C-415/05 P (Yassin Abdullah Kadi and Al Barakaat International Foundation v Council of the EU and Commission of the EC) ECLI:EU:C:2008:461.

Germany  165 is not only familiar in international legal relations as a reference to the ordre public as the boundary of a treaty commitment; it also corresponds, if used constructively, to the idea of contexts of political order which are not structured according to a strict hierarchy.220

(3) From the very beginning, the BVerfG’s jurisprudence was suffused with general reflections on the relationship between the EC/EU legal order and EU Member State’s domestic law as well as on the relationship between the EU and the its Member States’ courts and their respective jurisprudence (at least as regards constitutional courts/jurisprudence). Constitutional limits to integration are never mentioned without simultaneously referring to the principle of the Basic Law’s openness to integration, and thus relativising them.221 This coexistence and cooperation is analysed and reflected upon explicitly and in depth, inter alia drawing on arguments that are more at home in constitutional theory rather than constitutional law. Such is the Court’s set phrase of the EU Member States as the ‘masters of the treaties’ (Herren der Verträge)222 or of the EU as an ‘association of sovereign states’ (Staatenverbund).223 In its order of 15 December 2015, the Second Senate even calls the EU ‘an association of sovereign states (Staatenverbund), of constitutions (Verfassungsverbund), of administrations (Verwaltungsverbund) and of courts (Rechtsprechungsverbund)’.224 Thus, probably not by accident, it alludes to the ‘European association of constitutional courts’225 (europäischer Verfassungsgerichtsverbund), a phrase coined by the present president of the BVerfG. Also belonging to this sub-category is the ‘responsibility for integration’ (Integrationsverantwortung), which is also considered a duty arising out of the Constitution. However, this duty is supposed to go beyond the national strictures and to extend to all EU Member States as well as to the EU itself.226 (4) The BVerfG’s effort to reconcile claims of national constitutional and supranational law is clearly reflected in the fact that in its jurisprudence, the BVerfG sees itself bound to give due consideration to EC and EU law in general and to the CJEU’s jurisprudence in particular. Although all of these obligations are unwritten, they are nonetheless considered to be founded in the Constitution. In this respect, three arguments figure most prominently and have the greatest impact. The first is the ‘cooperative relationship’227 with the ECJ/CJEU, which provides the framework for the BVerfG’s jurisprudence. The second is the principle of ‘openness to European law’, which governs the interpretation of the Basic Law’s substantive as well as its procedural provisions.228 The third is the ‘responsibility for integration’,229

220 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 401 (BVerfG translation MN 340). 221 On this, see the Court’s summary of its own case law in BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 17–32. 222 BVerfG 2 BvR 687/85 (Kloppenburg) (n 46) 242; BVerfG 2 BvR 2134 (Maastricht) (n 43) 190; BVerfG 2 BvE 2/08 (Lissabon) (n 44) 348–9, 381 et seq; BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 302–3; BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 26; BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 44. 223 The first precedent was set in BVerfG 2 BvR 2134 (Maastricht) (n 43) 156 (Headnote 8), 181, 184; BVerfG 2 BvE 2/08 (Lissabon) (n 44) Headnote 1, 348. 224 This characterisation is all the more remarkable in that it does not speak of a ‘legislative association’ – and thus of the association between the EU and the Member States that can be most easily demonstrated in the applicable law (eg, when the Union and the Member States co-operate in the creation and implementation of directives). 225 Voßkuhle, ‘Verfassungsgerichtsverbund’ (n 131). 226 Thus explicitly in BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 47 with further references. 227 First mentioned in BVerfG 2 BvR 2134 (Maastricht) (n 43) 175. 228 See BVerfG 2 BvE 2/08 (Lissabon) (n 44) 346–7; BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 303 (BVerfG translation MN 341); BVerfG 2 BvR 987/10 (Euro-Rettungsschirm) (n 123) 172; BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 24; BVerfG 2 BvR 2735/14 (Europäischer Haftbefehl II) (n 46) MN 46. 229 BVerfG 2 BvE 2/08 (Lissabon) (n 44) 352, 353, 356 et seq (BVerfG translation MN 236) was a trail-blazing judgment in this respect.

166  Matthias Jestaedt which rests with both the EU as well as its individual Member States and their (constitutional) organs.230 (5) Finally, the BVerfG’s emphasis on its ‘reserve competence [mandated by the Constitution]’,231 and on the estimation that ultra vires acts of the EU’s organs (ultimately: of the CJEU) will be infrequent ‘borderline cases’232 is an expression of its reconciliatory efforts. This means that ordinarily, the constitutional order will adapt to the EC and EU law – not vice versa. The extent of the BVerfG’s effort to ensure that, on the one hand, the German constitutional law and EU law (on the general level) and, on the other hand, the BVerfG’s and CJEU’s jurisprudence (on a more particular level) cooperate and coexist harmoniously can best be shown in contrast to the CJEU’s conduct in a comparable setting. In its Opinion 2/13 on the EU’s accession to the ECHR,233 the CJEU found itself in a situation vis-à vis the ECHR that bears a striking resemblance to that of constitutional courts of EU Member States vis-à-vis the CJEU. In this situation, the CJEU neither attempted to minimise the damage, nor did it let itself be guided by a ‘cooperative relationship’, exercising any ‘reserve competence’ only in borderline cases. Rather, the CJEU reacted in a manner that was seen by many observers as being too blunt and overly formalistic by flatly refusing to entertain the possibility of the EU acceding to the ECHR without pointing a way out of this dilemma.234

References M Abels, Das Bundesverfassungsgericht und die Integration Europas (Munich, AVM, 2011). M Bäcker, ‘Das Grundgesetz als Implementationsgarant der Unionsgrundrechte’, (2015) 50 Europarecht 389. C Behrendt, ‘Der belgische Verfassungsgerichtshof ’, in A v Bogdandy, C Grabenwarter and P Huber (eds), Handbuch Ius Publicum Europaeum, Vol VI (Heidelberg, Müller, 2016) § 96. K v Beyme, ‘Modell für neue Demokratien? Die Vorbildrolle des Bundesverfassungsgerichts’, in R van Ooyen and M Möllers (eds), Handbuch Bundesverfassungsgericht im politischen System, 2nd edn (Wiesbaden, Springer, 2015) 927. G Britz, ‘Verfassungsrechtliche Effektuierung des Vorabentscheidungsverfahrens’, (2012) Neue Juristische Wochenschrift 1313. C Calliess, ‘Die europarechtliche Ultra-Vires-Kontrolle des Bundesverfassungsgerichts’, in C Calliess (ed), Herausforderungen an Staat und Verfassung: Völkerrecht – Europarecht – Menschenrechte. Liber amicorum für Torsten Stein zum 70. Geburtstag (Baden-Baden, Nomos, 2015) 446. P Cramer, Artikel 146 Grundgesetz zwischen offener Staatlichkeit und Identitätsbewahrung (Berlin, Duncker & Humblot, 2013). JP Derosier, Les limites constitutionnelles à l’intégration européene. Étude comparée: Allemagne, France, Italie (Paris, LGDJ, 2015). 230 Of central importance: BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 47: ‘The Member States and their constitutional organs – next to the institutions of the European Union – have to ensure that the integration programme is observed (responsibility with respect to integration, cf BVerfGE 123, 267 (352 et seq, 389 et seq, 413 et seq); 126, 286 (307); 129, 124 (181); 132, 195 (238 and 239) n 105). In the Federal Republic of Germany, it is the task of all constitutional organs to meet this responsibility with respect to integration’. 231 Eg BVerfG 2 BvE 2/08 (Lissabon) (n 44) 401 (BVerfG translation MN 341) emphasis added. 232 See BVerfG 2 BvR 2661/06 (Honeywell) (n 46) 303 (BVerfG translation MN 57); BVerfG 2 BvR 2728/13 (OMT (Vorlage an EuGH)) (n 46) MN 26. 233 CJEU 18.12.2014 Avis 2/13 (Accession of the Union to the ECHR) ECLI:EU:C:2014:2454, MN 153 et seq, particularly 178 et seq, summary at 258. 234 As a cross-check, one might ask whether it is imaginable that the CJEU would discover and formulate a legal principle of the primary law’s openness to the ECHR or even the ECtHR, comparable to the ‘principle of the Basic Law’s openness to European law’; Art 6(2)(3) TEU could serve as a textual basis which would be no less valid than Art 23(1) GG.

Germany  167 A Fischer-Lescano, C Joerges and A Wonka (eds), The German Constitutional Court’s Lisbon Ruling. Legal and Political-Science Perspectives (Bremen, ZERP, 2010). C Franzius, Vom Nationalstaat zum Mitgliedstaat und wieder zurück? Modifikationen ‘offener Staatlichkeit’ durch das Lissabon-Urteil des Bundesverfassungsgerichts, TranState Working Papers 124 (Bremen, 2010). C Franzius, ‘Grundrechtsschutz in Europa’, (2015) 75 Zeitschrift für auländisches öffentliches Recht und Völkerrecht 383. C Franzius, FC Mayer and J Neyer (eds), Grenzen der europäischen Integration. Herausforderung für Recht und Politik (Baden-Baden, Nomos, 2014). A Haratsch (ed), Nationale Verfassungen und Europarecht (Berlin, BWV, 2014). A Hatje, J Terhechte and U Everling (eds), Grundgesetz und europäische Integration. Die Europäische Union nach dem Lissabon-Urteil des Bundesverfassungsgerichts (Baden-Baden, Nomos, 2010). C Hillgruber, ‘Der Nationalstaat in der überstaatlichen Verflechtung’, in J Isensee and P Kirchhof (eds), Handbuch des Staatsrechts der Bundesrepublik Deutschland, Vol II, 3rd edn (Heidelberg, Müller, 2004) § 32. M Höreth, ‘Wenn Richter mitregieren wollen: Selbstautorisierung beim BVerfG und dem EuGH im Vergleich’, in R van Ooyen and M Möllers (eds), Handbuch Bundesverfassungsgericht im politischen System, 2nd edn (Wiesbaden, Springer, 2015) 875. PM Huber (ed), Der Vertrag von Lissabon und das nationale Verfassungsrecht. Europäische Verwaltungsnetzwerke (Stuttgart, Boorberg, 2013). PM Huber and A Paulus, ‘Cooperation of Constitutional Courts in Europe: The Openness of the German Constitution to International, European, and Comparative Constitutional Law’, in M Andenæs and D Fairgrieve (eds), Courts and Comparative Law (Oxford, OUP, 2015) 281. SP Hwang, ‘Der deutsche Verfassungsstaat im europäischen Mehrebenensystem: Überlegungen zur Auseinandersetzung zwischen den integrationsfreundlichen und -skeptischen Ansätzen’, (2015) 50 Europarecht 703. A Ingold, ‘Die verfassungsrechtliche Identität der Bundesrepublik Deutschland. Karriere – Konzept – Kritik’ (2015) 140 Archiv des öffentlichen Rechts 1. J Isensee, ‘Die Verfassung als Vaterland. Zur Staatsverdrängung der Deutschen’, in A Mohler (ed), Wirklichkeit als Tabu. Anmerkungen zur Lage (Munich, Oldenbourg, 1986) 11. R Ismer and D Wiesner, ‘Die OMT-Vorlage des Bundesverfassungsgerichts. Eine dogmatische Kritik auf Grundlage juristisch-ökonomischer Analyse’, (2015) 68 Die Öffentliche Verwaltung 81. M Jestaedt, ‘Bundesstaat als Verfassungsprinzip’, in J Isensee and P Kirchhof (eds), Handbuch des Staatsrechts der Bundesrepublik Deutschland, Vol II, 3rd edn (Heidelberg, Müller, 2004) § 29. M Jestaedt, ‘Phänomen Bundesverfassungsgericht: Was das Gericht zu dem macht, was es ist’, in M Jestaedt, O Lepsius, C Möllers and C Schönberger, Das entgrenzte Gericht (Frankfurt/M, Suhrkamp, 2011) 77. M Jestaedt, ‘Die deutsche Staatsrechtslehre im europäischen Rechtswissenschaftsdiskurs – Kennzeichen, Kernkompetenzen, Rezeptionshindernisse’, (2012) 67 JuristenZeitung 1. M Jestaedt, ‘Wissenschaft im Recht. Rechtsdogmatik im Wissenschaftsvergleich’, (2014) 69 JuristenZeitung 1. O Jouanjan, ‘Verfassungsrechtsprechung in Frankreich’, in A v Bogdandy, C Grabenwarter and P Huber (eds), Handbuch Ius Publicum Europaeum, Vol VI (Heidelberg, Müller, 2016) § 99. M Kerber, Der Kampf um den Lissabon-Vertrag (Stuttgart, Lucius & Lucius, 2010). R Kerssenbrock, Die Vereinigten Staaten von Europa. Die Lissabon-Entscheidung und die Notwendigkeit, Volkssouveränität neu zu verstehen (Berlin, epubli, 2013). P Kielmansegg, ‘Das Bundesverfassungsgericht und Europa’, P Kielmansegg (ed), Wohin des Wegs, Europa? Beiträge zu einer überfälligen Debatte (Baden-Baden, Nomos, 2015) 114. P Kirchhof, ‘Der deutsche Staat im Prozeß der europäischen Integration’, in J Isensee and P Kirchhof (eds), Handbuch des Staatsrechts der Bundesrepublik Deutschland, Vol X, 3rd edn (Heidelberg, Müller, 2012) § 214. M Klatt, Die praktische Konkordanz von Kompetenzen (Tübingen, Mohr Siebeck, 2014). H Klein, Europäische Integration und demokratische Legitimation (Baden-Baden, Nomos, 2011). D Knop, Völker- und Europarechtsfreundlichkeit als Verfassungsgrundsätze (Tübingen, Mohr Siebeck, 2013).

168  Matthias Jestaedt K Lenaerts, ‘Kooperation und Spannung im Verhältnis von EuGH und nationalen Verfassungsgerichten’, (2015) 50 Europarecht 3. O Lepsius, ‘Die maßstabsetzende Gewalt’, in M Jestaedt, O Lepsius, C Möllers and C Schönberger, Das entgrenzte Gericht (Frankfurt/M, Suhrkamp, 2011) 159. O Lepsius, ‘Souveränität und Identität als Frage des Institutionen-Settings’, (2015) 63 Jahrbuch des öffentlichen Rechts der Gegenwart N.F. 62. R Lhotta and J Ketelhut, ‘Bundesverfassungsgericht und europäische Integration’, in R van Ooyen and M  Möllers (eds), Handbuch Bundesverfassungsgericht im politischen System, 2nd edn (Wiesbaden, Springer, 2015) 845. R Lhotta, H Schöne and J Ketelhut (eds), Das Lissabon-Urteil. Staat, Demokratie und europäische Integration im ‘verfassten politischen Primärraum’ (Wiesbaden, Springer VS, 2013). EJ Lohse, ‘The German Constitutional Court and Preliminary References – Still a Match not Made in Heaven?’, (2014) 16 German Law Journal 1491. A López Castillo, Über die Gemeinschaftsrechtsprechung des deutschen Bundesverfassungsgerichts: Feder oder Schwert der europäischen Integration?, in C Calliess (ed), Herausforderungen an Staat und Verfassung: Völkerrecht – Europarecht – Menschenrechte. Liber amicorum für Torsten Stein zum 70. Geburtstag (Baden-Baden, Nomos, 2015) 718. M Ludwigs, ‘Der Ultra-vires-Vorbehalt des BVerfG – Judikative Kompetenzanmaßung oder legitimes Korrektiv?’, (2015) 34 Neue Zeitschrift für Verwaltungsrecht 537. B Maier, Grundrechtsschutz bei der Durchführung von Richtlinien (Baden-Baden, Nomos, 2014). A Mangold, Gemeinschaftsrecht und deutsches Recht. Die Europäisierung der deutschen Rechtsordnung in historisch-empirischer Sicht (Tübingen, Mohr Siebeck, 2011). J Masing, ‘Einheit und Vielfalt des Europäischen Grundrechtsschutzes’, (2015) 70 JuristenZeitung 477. C Möllers, ‘Legalität, Legitimität und Legitimation des Bundesverfassungsgerichts’, in M Jestaedt, O Lepsius, C Möllers and C Schönberger, Das entgrenzte Gericht (Frankfurt/M, Suhrkamp, 2011) 281. A Nußberger, ‘Europäische Menschenrechtskonvention’, in J Isensee and P Kirchhof (eds), Handbuch des Staatsrechts der Bundesrepublik Deutschland, Vol X, 3rd edn (Heidelberg, Müller, 2012) § 209. R van Ooyen, Die Staatstheorie des Bundesverfassungsgerichts und Europa. Von Solange über Maastricht zu Lissabon und Euro-Rettung, 5th edn (Baden-Baden, Nomos, 2014). R van Ooyen, ‘Eine “europafeindliche” Kontinuität? Zum Politikverständnis der Lissabon-Entscheidung des Bundesverfassungsgerichts’, in R van Ooyen, Bundesverfassungsgericht und politische Theorie (Wiesbaden, Springer VS, 2015) 129. A Proelß, Bundesverfassungsgericht und überstaatliche Gerichtsbarkeit (Tübingen, Mohr Siebeck, 2014). S Recker, Subjektivierung der Staatsstruktur. Schutzmechanismus nationaler Identität in der Europäischen Union (Cologne/Munich, Heymann, 2014). H Sauer, ‘Doubtful it Stood …: Competence and Power in European Monetary and Constitutional Law in the Aftermath of the CJEU’s OMT Judgment’, (2015) 16 German Law Journal 971. C Schönberger, ‘Anmerkungen zu Karlsruhe’, in M Jestaedt, O Lepsius, C Möllers and C Schönberger, Das entgrenzte Gericht (Frankfurt/M, Suhrkamp, 2011) 9. C Schönberger, ‘Höchstrichterliche Rechtsfindung und Auslegung gerichtlicher Entscheidungen’, in W Höfling (ed), Grundsatzfragen der Rechtsetzung und Rechtsfindung, Veröffentlichungen der Vereinigung der Deutschen Staatsrechtslehrer 71 (Berlin, De Gruyter, 2012) 296. C Schönberger, ‘Identitäterä. Verfassungsidentität zwischen Widerstandsformel und Musealisierung des Grundgesetzes’, (2015) 63 Jahrbuch des öffentlichen Rechts der Gegenwart N.F. 41. M Schönemeyer, Die Pflicht des Bundesverfassungsgerichts zur Vorlage an den Gerichtshof der Europäischen Union gem. Art. 267 Abs. 3 AEUV (Berlin, Duncker & Humblot, 2014). A Schwerdtfeger, ‘Europäisches Unionsrecht in der Rechtsprechung des Bundesverfassungsgerichts – Grundrechts-, ultra-vires- und Identitätskontrolle im gewaltenteiligen Mehrebenensystem’, (2015) 50 Europarecht 290. S Städter, Noch Hüter der Verfassung? Das Bundesverfassungsgericht und die europäische Integration (Marburg, Metropolis, 2013).

Germany  169 S Städter, ‘Das OMT-Verfahren in Luxemburg und Karlsruhe – ein wesentlicher Schritt der europäischen Krisenbewältigung?’, (2015) 51 Recht und Politik 20. D Sternberger, ‘Verfassungspatriotismus’ (1979), in P Haungs (ed), Dolf Sternberger. Schriften, Vol X (Frankfurt/M, Insel, 1990) 3. R Ullerich, Rechtsstaat und Rechtsgemeinschaft im Europarecht (Baden-Baden, Nomos, 2011). A Voßkuhle, ‘Der europäische Verfassungsgerichtsverbund’, (2010) 29 Neue Zeitschrift für Verwaltungsrecht 1. A Voßkuhle, ‘Die europäische Integration und das Bundesverfassungsgericht’, in Juristische Studiengesellschaft Karlsruhe, Jahresband 2014 (Heidelberg, Müller, 2015) 207. A Voßkuhle, ‘Integration durch Recht – Der Beitrag des Bundesverfassungsgerichts’ (2016) 71 JuristenZeitung 161. C Walter and M Vordermayer, ‘Verfassungsidentität als Instrument richterlicher Selbstbeschränkung in transnationalen Integrationsprozessen. Vergleichende Überlegungen anhand der Rechtsprechung von EuGH und EGMR’, (2015) 63 Jahrbuch des öffentlichen Rechts der Gegenwart N.F. 129. J Weiler, ‘The “Lisbon Urteil” and the Fast Food Culture’ (2009) 20 European Journal of International Law 505. M Wendel, Permeabilität im europäischen Verfassungsrecht (Tübingen, Mohr Siebeck, 2011). T Wischmeyer, ‘Nationale Identität und Verfassungsidentität. Schutzgehalte, Instrumente, Perspektiven’ (2015) 140 Archiv des öffentlichen Rechts 415.

170

6 Estonia HENT KALMO AND ANNELI ALBI1

I.  Introduction: Shifts in Estonian Constitutionalism in the Light of a ‘Disconnect’ between National and EU Legal Discourses As EU integration has come to impact national legal orders in virtually all areas of law, causing tensions in the process, a growing number of scholars have raised concerns about a ‘disconnect’ or a lack of meaningful dialogue between the national and EU legal discourses. Daniel Thym, for example, has evocatively summarised many of the issues in his Verfassungsblog post ‘The Solitude of European Law Made in Germany’: foreign languages as ‘an [… in]surmountable hurdle’, the lack of citations of national literatures in influential EU law journals, that ‘smaller jurisdictions do not have enough manpower to sustain debates about specialised questions of EU law’, etc.2 Thym suggests that ‘we have to find ways to link national and European debates’ in order ‘to overcome the disconnect’. In a follow-on blog post, Päivi Leino and Janne Salminen share similar concerns in the context of Finland, and observe, inter alia, that much of the research in European law

1 Sections II–VI (constitutional adaptations for EU law, sovereignty, supremacy, constitutional review) have been written by Hent Kalmo. Section VII (perspectives on substantive constitutional law and the social, democratic state based on the rule of law (Rechtsstaat), including issues raised in the ESM Treaty case) have been written by Anneli Albi. Given the range of the themes covered, we would like to add the following note for the information of the reader. Both co-authors share the broader concern about the pragmatisation of the Estonian constitutional system through adaptations to EU law, which is identified in the present chapter. Hent Kalmo holds the view that the turn to pragmatism should not exclusively, or even mainly, be attributed to EU law, but reflects a broader transformation of constitutionalism associated with the principle of proportionality. Anneli Albi in her sections of the chapter will point out that pragmatisation is additionally part of a broader ongoing shift to the EU’s autonomous paradigm of law and governance predicated inter alia on neofunctionalism and neoliberalism, and that this has caused similar problems for many other Member States; these have, however, not been discussed due to structural issues in EU discourses. The responsibility for the reasons identified in EU law and discourses lies only with Anneli Albi, who has developed these themes in further publications that will be referred to. The authors would like to gratefully acknowledge the grant funding of the following institutions. Hent Kalmo’s work was supported by the Estonian Research Council grant (PRG 942). Anneli Albi’s research was carried out in the framework of the research project ‘The Role and Future of National Constitutions in European and Global Governance’, funded by European Research Council grant No 284316. The views cannot be attributed to the Estonian Research Council or to the European Research Council. 2 Daniel Thym, ‘The Solitude of European Law Made in Germany’, Verfassungsblog 29.05.2014, https://verfassungsblog.de.

172  Hent Kalmo and Anneli Albi published in national languages in national journals risks being ‘ignored simply because important knowledge might be shut in rather closed circles.’ They argue that Keeping in mind the close relationship between law, legal research and culture, a situation where the European legal elites would discuss and publish legal research only in one language and in international fora could risk research being increasingly isolated from everyday societal debates and develop the results of such studies into being something for a closed epistemic community.3

Where national constitutions and constitutional courts do figure in the mainstream EU discourses, this tends to be predominantly through a very specific lens, such as whether the Member State and its courts and institutions are Euro-friendly in the application of EU law, or deviant and Eurosceptic by protecting sovereignty and national constitutional identity.4 In the prevailing EU narratives, what is conspicuously absent is coverage of, or indeed interest in, the problems that EU law and integration have posed to the constitutional orders of the Member States, especially to substantive constitutional values such as protection of fundamental rights, rule of law principles, judicial protection and separation of powers. One of the co-authors of this chapter, Anneli Albi, who has been puzzled with this prevailing approach, has documented the use of the above narratives with references to literature, and identified some of the potential underlying reasons in an earlier paper.5 These include broader dynamics elucidated in the literature on the epistemology of EU law and integration, especially that for a long time there has been scarcity of critical discussion on EU integration – which some have described as ‘silencing’ or ‘exclusion’ of ‘dissident voices’.6 The path to human progress is seen in a gradual shift to autonomous, self-referential EU law and legal consciousness that is decoupled from national cultures, histories and national legal languages.7 Indeed, the specificity of EU law and disconnect from comparative law is a core element of the ‘neo-functionalist 3 Päivi Leino and Janne Salminen, ‘Languages and EU law discourse: A view from a bilingual periphery’, Verfassungsblog 02.04.2014. Even in the UK, where the issue of language barriers is not present, Danny Nicol observed that ‘British constitutional scholarship for the most part tends to focus on Britain’s internal institutions’. He also observed a ‘reluctance to engage with globalisation’ and especially with the constitutional entrenchment of neoliberalism through EU law; for a detailed discussion, see Danny Nicol, The Constitutional Protection of Capitalism (Oxford, Hart, 2010) 5 ff. 4 For extensive documentation of examples of the very critical views in EU law literature with regard to constitutional courts, and for the broader structural reasons in the discourse, see Anneli Albi, ‘Erosion of constitutional rights in EU law: A call for “substantive co-operative constitutionalism”’, (2015) 9 (2) Vienna Journal of International Constitutional Law, 151 (Part 1 of the article), especially sections I and IV; and No 3, 291 (Part 2). 5 For examples of the narratives, see previous footnote; for the underlying reasons, see Albi, ‘Erosion’ (n 4), Part 2, 307 ff. 6 Ian Manners and Richard Whitman, ‘Another Theory is Possible: Dissident Voices in Theorising Europe’, (2016) 54 (3) Journal of Common Market Studies, 5–9, 12. See also below n 137 and the accompanying text. 7 Albi, ‘Erosion’ (n 4), Part 2, 307 ff. Different aspects of the shift from comparative law to autonomous, self-referential EU law have been pointed out by the following authors. In the context of private law, Simone Glanert has summarised debates amongst private law experts regarding the European Commission’s programme of gradually replacing national laws, which have been embedded in a tradition and culture, by a unique ‘European’ law; see Simone Glanert, ‘Speaking Language to Law: The Case of Europe’, (2008) 28 Legal Studies 163 f. Antoine Vauchez has documented the extensive EU funding for EU law journals and university centres to promote the specificity of EU law; see Antoine Vauchez, ‘The transnational politics of judicialization. Van Gend en Loos and the making of the EU polity’, (2009) 16 European Law Journal 4 and 26ff, 20–21. Anne-Lise Kjaer has perceptively raised the more general question ‘what will happen to the mutually divergent national languages and cultures of law when independency and autonomy of a common European law are presumed by an increasing number of European lawyers; when they … involve themselves in an increasingly self-referential European legal discourse with lawyers from other European countries; and when communicating about law and speaking the law are no longer conducted in divergent national legal languages, but in a Europeanized legal language with no reference to the domestic laws of the Member States’; see Anne-Lise Kjaer, ‘Theoretical Aspects of Legal Translation in the EU: The Paradoxical Relationship between Language, Translation and the Autonomy of EU Law’, in S Sarcevic (ed), Language and Culture in EU Law. Multidisciplinary Perspectives (Burlington, Ashgate, 2015) 91, 98 and 105. The direction of travel towards post-national law tallies more broadly with the observation by Alexander Somek that the ‘[m]ainstream discourse … is blinded by the belief that the one remaining obstacle to human progress is the nation state’; Alexander Somek, The Cosmopolitan Constitution (Oxford, OUP, 2014), p vii.

Estonia  173 paradigm of integration-through-law’, which, as extensively documented by Antoine Vauchez, has been the EU’s most established and powerful meta-narrative and has come to be the ‘new common sense’ of the EU legal and political discourses.8 We would invite the reader to bear the above observations in mind and consider their implications when reading the present chapter on the adaptation of the Estonian Constitution and constitutional system for EU integration. The Estonian case provides perhaps one of the clearest examples of the way in which EU law has profoundly changed a national constitutional system in a short period of time. Having regained its independence from the Soviet Union in 1991 and established a constitutional system modelled on the German post-totalitarian constitutional Rechtsstaat, Estonia joined the EU in 2004 along with nine other countries, mostly from Central and Eastern Europe. The country has been at pains to be seen as a ‘good European’, and has therefore implemented EU law predominantly without reservations. It will be seen that the national constitution has by and large been suspended in areas that fall within the scope of EU law, and courts have granted to EU law exceptionless primacy. From the perspective of EU law, this could be regarded as an exemplary or indeed an ideal approach. However, from the perspective of the Estonian constitutional system, we have observed the following broader shift, which we would like to submit for the consideration of both the Estonian and European/international readership. It is that the ongoing processes – as referred to in the title of the chapter – have brought about a profound but largely undiscussed shift from the binding rules of the post-totalitarian constitutional Rechtsstaat to pragmatism.9 In the account that follows, we will seek to point out how the above-mentioned shift from rules to pragmatism finds expression, in particular, in three developments. All three can be seen as introducing a caveat to the text of the Constitution, in the sense of driving a wedge between what appears from the letter of constitutional provisions and the actual legal state of affairs. The first is a shift in relation to the binding nature of constitutional provisions, not only the provisions on sovereignty but a wide range of provisions that overlap with the field of regulation of EU law. This shift has been brought about both through the exceptional way in which the Constitution was amended, as well as through judicial interpretation. It will be seen that a profound change has occurred through the Supplementary Act in 2003, which was adopted with the aim of accommodating EU integration by supplementing the Constitution with a separate Act that generically provides that the Constitution ‘is applied without prejudice to the rights and obligations arising from the Accession Treaty’ (Art  2). It will be seen that in the judgments of the Supreme Court, this provision, along with the generic clause on the respect for the ‘fundamental principles of the Constitution’, have been interpreted in a way that grants EU law exceptionless primacy. Thus the Act introduced and, in turn, spurred on the wider development of representing the Constitution as a set of flexible principles as opposed to a set of strict, binding rules. Through the Supplementary Act, the Estonian judiciary was able to cast the relationship between EU commitments and the Estonian Constitution as similar to the infringement of a fundamental right, and to assess their constitutionality through the lens of the principle of proportionality. As will be explained below, this doctrinal shift was in the making for nearly 10 years and was brought into full view in the ESM case in 2012.

8 Antoine Vauchez, ‘“Integration-through-Law”. Contribution to a Socio-history of EU Political Commonsense’, EUI Working Paper RSCAS 2008/10, http://cadmus.eui.eu, 1 f, 21 f. The ‘neo-functionalist paradigm’ of ‘integration-throughlaw’ is explained by Vauchez in fn 4 of the paper. On the centrality of neofunctionalism to autonomous EU law and governance, see also below n 124 and the accompanying text. 9 A similar observation about ‘flexibilisation’ and even obsolescence of constitutions has been observed by Italian scholars; see below nn 119–120 on the work of Simoncini and the accompanying text.

174  Hent Kalmo and Anneli Albi The second shift is from an all-encompassing approach to constitutional review by the Supreme Court to its near abandonment in the very broad areas of law that are more or less ‘tied’ to EU law. Sections II–VI, which explore these first two shifts, are written by Hent Kalmo. The third shift, outlined by Anneli Albi in section VII, concerns the broader post-totalitarian, German-influenced, Rechtsstaat model conceived in terms of substantive constitutional values. Examples will be brought from various areas of law where the well-established protection to fundamental constitutional rights and rule of law safeguards have gradually been levelled downwards to comply with the priorities of the EU legal order, which are oriented more towards effectiveness, teleological interpretation, the market, and tend to represent neofunctionalist approach to law and legal thinking (section VII.B). Examples include the principle of legitimate expectations and rules on publication of laws in the sugar stocks cases, judicial protection in European Arrest Warrant cases, and the disappearance of the difference between administrative penalties and criminal punishments in what has been described by Estonian lawyers as a new, combined ‘criminstrative law’.10 Greater space will be devoted to concerns with regard to the requirements of the social, democratic state based on the rule of law, which were expressed by nine dissenting justices in the ESM Treaty case (section VII.A). Indeed, in that case, the majority of the Supreme Court eventually gave an indication that potential limits may exist to the constitutionality of future EU treaties. Anneli Albi will further point out that the issues arising with regard to substantive as well as formal constitutional law have arisen in a large number of Member States, and thus there seems to be a conceptual flaw in EU discourses that typically tend to reduce such matters to the notion of national constitutional identity, with connotations of idiosyncrasy. Constitutional issues which have arisen in Estonia are similar to, in particular, what Albi has categorised in her research as the post-totalitarian and post-authoritarian constitutional traditions, which have been modelled on the German Basic Law. Albi will additionally draw attention to broader concerns of constitutional scholars Europe-wide that the effects of EU law, especially the euro crisis management, are increasingly leading to the ‘twilight’ of constitutionalism, ‘the end of constitutionalism as we know it’, o ­ bsolescence of national constitutions, and even a change from a constitutional law paradigm to a new paradigm of governance.11 We will conclude with a broader observation that the above, profound changes in Estonian constitutionalism have been accepted in Estonia – besides the keenness to be ‘good Europeans’ – also due to acute geopolitical concerns. However, we would like to invite the reader to consider, when taking a look at the material, whether the shift to pragmatism is the right long-term direction of travel, or, as in our view, ought to remain a temporary, short-term solution, with an eventual return to the binding rules of the constitutional Rechtsstaat. In the latter case, due to geopolitical concerns as well as the constraints of the EU legal order such as costly enforcement proceedings, neither Estonia nor its legal community are in a position to change the direction of travel. We will put forward the suggestion that constitutional problems encountered by the Member States due to EU law ought to be formally given attention as part of the ongoing debate on the future of Europe. In the conditions where there has by and large been a ‘disconnect’ and a lack of meaningful, joined-up discussion between the national constitutional discourses and the EU law discourses, it would seem premature to proceed to reforms that may well have far-reaching consequences to the Member States’ constitutional orders. Such a wider, joined-up discussion

10 See 11 See

below n 104 and the accompanying text. below nn 118–124 and the accompanying text.

Estonia  175 also seems pressing for the reason that whereas in the mainstream EU discourses the voters’ turn to Eurosceptic and extreme parties has been attributed to populism, nationalism and problems in the individual Member States, the causes may also lie partly in some structural issues of EU law and integration. Indeed, in the United Kingdom, which has meanwhile left the Union, one salient theme in the media debates leading up to Brexit was that of ‘a slow and invisible process of legal colonisation, as the EU infiltrates just about every area of public policy’.12

II.  Main Characteristics of the National Constitutional System The present Estonian Constitution was adopted in the wake of what is known as re-independence, achieved in the confusion surrounding the conservative Moscow coup of August 1991. The years leading up to this pivotal moment were marked by the moderately liberal atmosphere of perestroika, and landmark events such as the Baltic Way, where more than two million people formed a human chain from Tallinn to Vilnius in August 1989 to protest peacefully against the Soviet occupation. During these years, the former Soviet republic took a series of steps towards selfgovernment. Already in the spring of 1990, the Supreme Soviet had declared its intention to embark on a step-by-step process towards independence. From the point of view of constitutional theory, what mattered most was that Estonia’s independence was not achieved by way of seceding from the Soviet Union. The process was seen rather as a restoration of the country’s rightful legal status which had been violated in 1940.13 The restoration of independence was thus grounded in the idea of legal continuity with the pre-war republic, implying that the Soviet interlude, durable as it was, had been no more than a brute fact, without any valid legal consequences.14 The doctrine of legal continuity with the inter-war republic became a foundational principle in building up Estonia’s constitutional system after 1991. Some politicians insisted on taking it to the extreme of simply declaring the continuing validity of the Constitution of 1938. A more pragmatic solution prevailed in the end. The Constitutional Assembly, the body entrusted with the task of working out a new constitutional text, borrowed from the pre-war Constitutions features which were felt to be still appropriate. The first Constitution of 1920 was thought to have been too parliamentarian, whereas that of 1938 had given excessive power to the head of state. The Constitution of 1992 accordingly represented an attempt to find a middle ground between an overbearing executive, on the one hand, and the opposite extreme of an all-powerful Parliament leaving little room for a consistent governmental policy, on the other. This resulted in the present system where the powers of Parliament and the Government are clearly delimited, both having their independent spheres of action. 12 Boris Johnson, ‘There is only one way to get the change we want – Vote Go’. Editorial, The Daily Telegraph, 22.02.2016, 18. The word ‘colonisation’ has been used with regard to EU law in the work of a number of leading UK EU law scholars. 13 In that year, the Soviet leadership had profited from the presence of its army in the Estonian territory to mount a coup, dressed up as a revolution, and to install an obedient regime that immediately applied for membership of the Soviet Union. These actions had followed the signing of the Nazi-Soviet Pact of August 1939 (the Molotov-Ribbentrop Pact) with its secret protocol allotting Estonia to the Soviet Union. Once the secret protocol had been declared illegal and void ab initio in 1989, it appeared to follow that the whole Soviet regime was without any legal basis in Estonia. 14 The doctrine of state continuity had started to gain ground long before the draft of the Constitution was worked out in the Constitutional Assembly. It had become the rallying cry of the movement which contested the legitimacy of the Estonian Supreme Soviet, designating it as an organ of an occupying country (the Soviet Union). In order to create a body representing what they took to be the true bearer of sovereignty, the leaders of this movement started to compile lists of the citizens of the interwar-war Estonian republic and their descendants. Those included in the lists were then called on to elect the Estonian Congress – an assembly in open constitutional rivalry with the Soviet authorities. By the summer of 1991, Estonia thus had two elected bodies – the Estonian Supreme Soviet and the Estonian Congress – both claiming to express the will of the people.

176  Hent Kalmo and Anneli Albi Estonia, nevertheless, remains a parliamentary republic where the Government is accountable to the legislature. The latter also elects the President who possesses the power of suspensive legislative veto and can initiate constitutional review in the Supreme Court in case Parliament overrides this veto. Although the President is sometimes described as a mere figurehead with very limited powers, the political practice has revealed this to be a misrepresentation of his or her position in the constitutional system. The suspensive veto – backed up by the possibility of constitutional review – is a lever that has sometimes proved remarkably effective in the hands of an activist holder of the office. Estonia has never had a separate constitutional tribunal.15 All courts, both civil and administrative, of first instance and appellate jurisdiction, apply the Constitution and can set aside, either ex officio or upon request, a piece of legislation that they deem unconstitutional. The role equivalent to a constitutional court is, however, in reality, held by the Constitutional Review Chamber (or the full bench) of the Supreme Court (Riigikohus16), which has the power to rule definitively on the validity of a statute and strike it down. The disapplication of a statute in a lower court only has the effect of triggering the proceedings higher up in the Supreme Court. The review initiated within the judicial system is always concrete review, ie the constitutionality of statutes is tested with regard to the specific facts of an actual controversy. But the Supreme Court can also be asked by some institutions to conduct abstract review of legislation. One peculiarity of the Estonian constitutional system is the presence of the Chancellor of Justice with powers to apply for constitutional review. Through this channel, the Supreme Court may be confronted with disputes unlikely to have come up from lower, ordinary courts, as the latter can only set aside and send for review by the Supreme Court statutes directly relevant to a case at hand.17 As already mentioned, the President also has the power to ask for ex ante constitutional review if the Riigikogu overrides his or her veto. Cities and municipalities can request abstract constitutional review at the Supreme Court only when their constitutionally guaranteed right of self-government is at play. One may add that determining the limits of local self-government has been one of the most abiding concerns for Estonian constitutional lawyers, in spite of the fact that, by now, they can draw on a very substantial body of judgments.

III.  Constitutional Culture As in many other countries, the character of the Estonian constitutional culture is determined largely by the central role of the Supreme Court, which enjoys the monopoly of constitutional review. The constitutional disputes provoking debate and attracting the public’s attention are generally those either already before the Supreme Court or having at least the potential of ending up there. Overall, the Court’s decisions enjoy a very high level of compliance. Their authoritativeness is such that the discussion of even the most divisive issues tends to die down after the Court has rendered its decision. Perhaps one deep-seated reason is that, in Estonia, constitutional law is widely seen as a field amenable to objective, scientific analysis. Judicial decisions are almost

15 When the idea was put forward in the Constituent Assembly which adopted the first Constitution in 1920, it was argued that there was no need for such a specialised tribunal as all courts had to apply the Constitution anyway. 16 The literal translation of the Riigikohus would be the ‘State Court’; however the official term used by the Court in English is ‘Supreme Court’. 17 The ESM case is one important example illustrating this. For this case, see below section VII.A.

Estonia  177 never discussed in political terms, ie as an expression of political bias. This is so to some extent even in the case of presidential vetoes. The President is expected to be a non-partisan figure and an openly political veto, appearing to use the argument of unconstitutionality as a mere pretext, would most likely be seen as abusive. Three factors have been especially important in forging the character of Estonia’s present constitutional culture. The first is the doctrine of state continuity, mentioned above. This doctrine not only prompted the choice of using the pre-war republic as a pattern for moulding the new constitutional system. It was also used to justify a series of thoroughgoing political changes, such as the comprehensive ownership reform, initiated in 1991 and aimed at restoring the rights of those who had lost property after 17 June 1940 (the beginning of Soviet occupation). Legal continuity with the pre-war republic is still taken to be a fundamental principle underlying the constitutional text. To give just one example of its continuing relevance, it has been at the centre of the arguments about the constitutionality of the recently signed new border treaty between Estonia and Russia. The second dimension of the Estonian constitutional culture is the above-mentioned influence of the German Basic Law, including its influence on the drafting process of the Constitution. This is part of the broader impact of the post-totalitarian, Rechtsstaat-based constitutional tradition both in the Western and Central-Eastern parts of Europe, which will be explored in section VII.B below. The third cultural factor informing attitudes to the constitutional text has been the diffusion of a balancing model of legal reasoning. This model was borrowed largely from the case law of the European Court of Human Rights and conceptualised with the help of the theory of principles developed by Robert Alexy, one of the foreign experts asked in the mid-1990s to assess whether the Estonian Constitution needed amendment. In the practice of the Supreme Court, the theory of principles first became prevalent in fundamental rights cases where it provided a ready tool to flesh out the constitutional requirement of necessity. There was nothing idiosyncratic about such use of balancing in the area of fundamental rights, which reflected a global trend and, in fact, was almost mandated by the European Convention on Human Rights (ECHR). More noteworthy, and often subtle, are the wider ramifications of the emphasis on the principle of proportionality. The principle has been extended from fundamental rights to other areas of law, so that balancing has now become the dominant technique of resolving all kinds of constitutional disputes. Perhaps most controversially, as already mentioned, the Supreme Court used proportionality as a criterion for assessing limitations to sovereignty in the ESM case.18 This aspect of the decision provoked a dissenting opinion by two judges who contested the appropriateness of applying proportionality analysis outside the field of fundamental rights, all the more since the text of the Constitution speaks of necessity only in the context of fundamental rights.19 Although contested at times, the balancing model of constitutional reasoning has become embedded in the Estonian legal culture. It is likely that future constitutional challenges to European integration will be addressed in the same pragmatic way: weighing the infringement of constitutional values against the benefits of the legal measures challenged.

18 See Carri Ginter, ‘Constitutionality of the European Stability Mechanism in Estonia: Applying Proportionality to Sovereignty’, (2013) 9 European Constitutional Law Review 335. 19 Constitutional Review Chamber of the Supreme Court (hereinafter CRCSC) 11.05.2006, Opinion 3-4-1-3-06, Dissenting opinion of Justice Tampuu. English versions of the most important CRCSC jurisdiction are provided at www. riigikohus.ee/en/judgements/constitutional-judgments.

178  Hent Kalmo and Anneli Albi

IV.  Constitutional Foundations of EU Membership and Closely Related Instruments A.  Pre-Accession Debate and Proposed Constitutional Changes Estonia was invited to start negotiations to join the EU in 1997. As already noted, there was first very little discussion about possible constitutional obstacles to accession. The initial absence of a scholarly debate appears somewhat surprising in retrospect, given that the text of the Constitution contains a formula invoking a highly conservative notion of sovereignty. Article 1 of the Constitution provides that ‘Estonia is an independent and sovereign democratic republic wherein the supreme power of state is vested in the people. The independence and sovereignty of Estonia are timeless and inalienable.’ Whilst these ringing words were not seen as a barrier to accession, they were nevertheless something of an embarrassment to those lawyers and political commentators who were at pains to downplay the idea that the Constitution barred Estonia from joining the EU. As one scholar put it, the ‘notion of a timeless sovereignty has itself outlived its time’.20 Paradoxical as it may sound, politicians, many of whom had been members of the body which framed the Constitution, argued that a provision adopted less than a decade ago was hopelessly out of date. But the debate was not limited to the question of whether membership in the EU would be compatible with Estonia’s sovereignty. An additional problem arose from Article 123 of the Constitution, according to which ‘The Republic of Estonia may not enter into international treaties which are in conflict with the Constitution.’ Whatever the uncertainty regarding sovereignty, it was clear that EU membership would entail obligations that conflicted with the Constitution, eg the equal treatment of all Union citizens. Not all conflicts could be ironed out merely by interpretation. In order to fully appreciate the debate that followed, it is helpful to know something about the historical background to the adoption of the Estonian Constitution in 1992. Several experiences might explain why the framers embraced an apparently isolationist souverainiste position. First, and most importantly, the declaration that the Republic’s ‘independence and sovereignty’ are ‘timeless and inalienable’ can be seen as the Estonian ‘nie wieder’ – a constitutional commitment to avoid a repeat of the scenario of 1940, when a series of unilateral legal steps taken by the then head of state paved the way for the country’s incorporation into the Soviet Union. In the summer of 1992, at the time the referendum on the Constitution was held, the notion of sovereignty also resonated strongly with the developments that preceded the break-up of the Soviet Union and, more particularly, with the constitutional conflict that had pitched the small republic against the central Union authorities in 1988. In that year, the Estonian Supreme Soviet had found itself at the forefront of the opposition to a series of centralising amendments to the Soviet Constitution. Its response was the adoption of a curious and legally anomalous act – the Declaration of Sovereignty (November 1988).21 Drawing directly on Jean Bodin, the Declaration remarked: The sovereignty of the Estonian Republic means that, through its supreme bodies of power and administration and judicial bodies, it wields supreme power in its territory. The Sovereignty of the Estonian 20 Mario Rosentau, ‘Suveräänsus Euroopa Liidus’ [‘Sovereignty in the European Union’], Riigikogu Toimetised 2003/8, 11. 21 Added to the declaration were a number of amendments to the Soviet Estonian Constitution which were nothing short of revolutionary, for they effectively made the application of Soviet central legal acts optional in the territory of the Estonian SSR.

Estonia  179 Republic is one and indivisible … [The Estonian Supreme Soviet] declares the priority of its laws within the territory of Estonian SSR.22

No wonder, then, that once accession to the EU appeared in the offing in the second half of 1990s, politicians and legal scholars were confronted with a conundrum: how to apply Article 1 of the Constitution, formulated with the fresh memory of the Soviet experience in mind, to a Union that was perceived as benign and beneficial to Estonia? Should there be a double standard? Were any benefits that might follow from the alleged loss of sovereignty or independence legally relevant at all? A commission of legal experts, convened by the Government to assess whether the Constitution needed to be changed, struck a note of compromise. Its concluding report remarked that, since the EU had remained a confederation, membership would not result in a loss of independence. Estonia’s sovereignty, on the other hand, would be considerably reduced by the obligation to refrain from legislating in some areas and to implement legal measures adopted by majority voting.23 The commission of experts argued that there was no ground in the Constitution for permitting a foreign body to exercise Estonia’s sovereign powers. It thus concluded that, in order to make it possible for Estonia to join the EU and enforce the primacy of Union norms, the Constitution had to be amended. The experts suggested that this be done in the form of a set of provisions, one of which would explicitly allow Estonia, under the principle of reciprocity and equality, to delegate state powers stemming from the Constitution to the institutions of the European Union for their joint exercise by Member States of the European Union to the extent that is necessary for the implementation of the agreements underlying the Union, and on condition that it is not opposed to the foundational principles and tasks of the Estonian statehood set out in the preamble to the Constitution.24

The expert commission stressed that all deviations from the principles mentioned in the Constitution should be seen as exceptional. Its members thought that this idea is best expressed by opting for a general enabling provision rather than introducing amendments to all Articles of the Constitution which might be affected by membership in the EU. For example, the commission saw no need to change the wording of Article 59 of the Constitution, according to which ‘Legislative authority is vested in the Riigikogu’. Some foreign experts had taken a different view. The legal opinion prepared by McKenna & Co recommended numerous specific amendments, eg adding the words ‘and in the institutions of the European Union’ to Article 59 of the Constitution. This difference could perhaps be seen as a matter of style. But there was also a much more fundamental point with regard to which the expert commission disagreed with the assumptions of McKenna & Co. The latter had proposed to complement the second sentence of Article  1 of 22 Declaration of the Estonian Supreme Soviet on the Sovereignty of the Estonian SSR, 16 November 1988. For context and aftermath, see Edward W Walker, Dissolution. Sovereignty and the Breakup of the Soviet Union (Lanham MD, Rowman & Littlefield, 2003). The Soviet Union was nominally governed by a Constitution which described the Union Republics as ‘sovereign’ and endowed them with the right of secession, there being, however, no realistic possibility of exercising this right before the end of the 1980s. On the other hand, one should not rush to conclude that the letter of the Soviet Constitution of 1977 was entirely without relevance. In the favourable circumstances which followed the beginning of perestroika, the text of the Soviet Constitution, and particularly the ambiguity of the word ‘sovereignty’, could be played upon by a whole gamut of political movements in the member republics in order to further their agenda, from local communists eager to expand their autonomy within the Soviet system to those making an explicit bid for the restoration of independence. Significant legal and political changes could be justified as mere conclusions from the constitutionally recognised status of the Union Republics – a strategy which was all the more effective as Moscow struggled to formulate an alternative line of constitutional interpretation that could be used to counter the Baltic claims. 23 Võimalik liitumine Euroopa Liiduga ja selle õiguslik tähendus Eesti riigiõiguse seisukohalt [‘Possible accession to the European Union and its legal meaning from the perspective of Estonian constitutional law’], Final Report of the Legal Expertise Committee for the Constitution of the Republic of Estonia (1998). 24 Ibid.

180  Hent Kalmo and Anneli Albi the Constitution (which reads: ‘The independence and sovereignty of Estonia are timeless and inalienable’) with the following words: ‘and can be transferred to the institutions of international organisations only in accordance with Article 3’.25 The expert commission resolutely rejected the notion that state powers, let alone Estonia’s independence and sovereignty, could be transferred to a foreign institution. What was possible in its opinion was the delegation of the exercise of these powers – and this under very strict conditions. The expert commission also diverged from the view of McKenna & Co that the authorisation to delegate state powers should be extended to the broader class of ‘international organisations’ (of which the European Union would be only an instance). Consonant with the exceptional character of the limitations that membership in the EU would put on fundamental constitutional principles, the amendments would have to be tailored to the unique nature of this organisation.26 Such fundamental questions, and especially the issue of sovereignty, receded into the background in the ensuing debate on the constitutional implications of joining the EU. Much more attention was devoted to the question of whether a sweeping authorisation would be preferable to a set of piecemeal amendments throughout the text of the Constitution. At the beginning of 2002, another expert group was convened to explore the matter; it proposed a novel solution of leaving the constitutional text unchanged and supplementing it instead with a separate legal instrument.27 The then Minister of Justice Märt Rask threw his weight behind this solution. He suggested that the added instrument should state: ‘When Estonia has acceded to the European Union, the Constitution of the Republic of Estonia is applied without prejudice to the rights and obligations arising from the Accession Treaty.’ There would be no imperative need to introduce changes into other provisions, the minister argued. ‘To speak of amending would be justified’, he wrote in an opinion piece, ‘if there was anything wrong in the text.’28 In this case, the change did not stem from something being wrong, he claimed, but from progressing to a new era. Seeking to summarise the divergence of views among politicians, constitutional scholars and the general public, the minister wrote that ‘changing the main text of the Constitution is, in a way, a question of religion – to many people, the Constitution is a sacred document, so that changing the position of a comma would be a sacrilege; others are willing to amend it every year to finally achieve juridico-technical perfection.’29 This quote serves as a reminder that the amendment debate was 25 Art 3 reads: ‘Governmental authority is exercised solely pursuant to the Constitution and laws which are in conformity therewith. Generally recognised principles and rules of international law are an inseparable part of the Estonian legal system. Laws are published in accordance with prescribed procedure. Only published laws may have binding force.’ 26 Other foreign experts offered different suggestions as to how to amend the Constitution. Robert Alexy, for example, proposed that Art 1 be added to with the following words: ‘The independence and sovereignty of Estonia do not exclude membership in the European Union.’ Robert Alexy, ‘Põhiõigused Eesti põhiseaduses’ [‘Fundamental Rights in the Estonian Constitution’], (2001) Special Edition Juridica 5. 27 A more detailed overview of the drafting process and of the debates that ensued is developed in Madis Ernits, Carri Ginter, Saale Laos, Marje Allikmets, Paloma Krõõt Tupay, René Värk and Andra Laurand, ‘The Constitution of Estonia: The Unexpected Challenges of Unlimited Primacy of EU Law’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law – National Reports (The Hague, TMC Asser Press, 2019) 887. That study traces back that, according to MP Mart Nutt, the following experts played a central role in the drafting process of what became the Supplementary Act, submitted to the Riigikogu on 16 May 2002: Kalle Merusk, Jüri Põld, Ülle Madise, Julia Laffranque and Märt Rask, see minutes of the IX Riigikogu, VII session (11 June 2002), (available in Estonian) http://stenogrammid.riigikogu.ee/et/200206111000#PKP-2000008924. The above study further notes that the Supplementary Act solution was earlier brought into the limelight by Laffranque, who listed possible ways to amend the Constitution for the purpose of EU membership; the list summarised proposals presented by other legal experts, inter alia the possibility of introducing a third supplementing act to the Constitution (Julia Laffranque, ‘Constitution of the Republic of Estonia in the Light of Accession to the European Union’, (2001) 6 Juridica International, 207, 215 f). 28 Märt Rask, ‘Kas muuta põhiseadust või mitte?’ [‘Should the Constitution be Amended or not?’], Postimees 15.01.2002. 29 Ibid.

Estonia  181 not merely a technical exchange of views about doctrinal intricacies or about the pros and cons of various methods of constitutional drafting. Even scholarly opinions often carried a strong evaluative subtext about the danger of marring the beauty of the familiar formulas contained in the Constitution with additions that might be required from a narrow legal point of view. Such considerations could help to explain why the idea of adding a separate instrument to the Constitution prevailed in the end. Of course, the solution of limiting the change to a general laconic authorisation also had the enormous political advantage of avoiding disagreements about the precise wording of the numerous amendments that would otherwise be needed – not to forget the additional benefit of creating the illusion in the public’s mind that the constitutional order would be well-nigh unaffected by accession. Unsurprisingly, the solution at first provoked strong criticism. One prominent Estonian judge and constitutional lawyer went so far as to describe it as ‘a rape of the Constitution or constitutional nihilism’.30 Two main arguments were directed against the proposal. It was objected, first, that leaving the main text of the Constitution unchanged and relying on a vague tacked-on provision would give rise to a dangerous situation, where the letter of the Constitution ceased to reflect the actual mechanisms of exercising authority. This would, in turn, undermine the normative status of the Constitution. The other main objection was that the provision endorsed by the Minister of Justice was recklessly generous in failing to attach any conditions to the applicability of EU law. ‘No Member State of the European Union has subjected its constitution to so bold a treatment’, wrote the then Chancellor of Justice Allar Jõks in a critical op-ed.31

B.  Changes Introduced to Enable Accession Despite criticism, it was the sweeping authorisation proposed by the Minister of Justice that became the starting point for drafting the accession amendment. Rather than changing the wording of Article 1 of the Constitution, the people would be asked to vote on a Supplementary Act to the Constitution, which provides that Estonia may belong to the European Union. In order to meet the concerns of the experts unhappy with the absolute nature of the authorisation, it was complemented with what came to be called a safeguard clause, making membership in the EU conditional upon respect for the ‘fundamental principles of the Constitution’. The text finally approved by a referendum in 2003, in the Act entitled ‘The Constitution of the Republic of Estonia Supplementing Act’, reads as follows: In a referendum held on 14 September 2003 pursuant to Article 162 of the Constitution of the Republic of Estonia, the people of Estonia adopted the following Act to amend the Constitution: §1. Estonia may belong to the European Union, provided the fundamental principles of the Constitution of the Republic of Estonia are respected. §2. When Estonia has acceded to the European Union, the Constitution of the Republic of Estonia is applied without prejudice to the rights and obligations arising from the Accession Treaty. §3. This Act may only be amended in a referendum. §4. This Act enters into force three months after the date of its promulgation.

It ought to be noted that, strictly speaking, the Estonian legal system does not provide for such a thing as ‘supplementing the Constitution’. The only way to introduce new legal rules of 30 Rait Maruste, ‘Põhiseadust tuleks siiski muuta’ [‘The Constitution Should Nevertheless be Amended’], Postimees 29.01.2002. 31 Allar Jõks, ‘Põhiseaduste muutuste künnisel’ [‘Constitution on the Verge of Changes’], Postimees 01.04.2002.

182  Hent Kalmo and Anneli Albi constitutional rank is to follow the procedures set out in Chapter XV of the Constitution. The Supreme Court later described what had happened in the following (slightly confusing) terms: The Constitution was added to by a referendum of 14 September 2003, resorting to the model of constitutional amendment in which the additions to the Constitution are enacted as separate constitutional acts and the provisions of the Constitution are not formally amended.

The Court added that, ‘at the same time the text of the Constitution must always be read with the additions and only that part of the constitutional text shall be applied which is not in conflict with the additions.’32 This last directive seems clear enough. However, the Supreme Court’s dicta prompted a difficult and largely unexplored question. Is the Supplementary Act a lex posterior that carves out an exception from the general rule that Estonia is an independent and sovereign democratic republic? In this case, it clearly constitutes an amendment to Article 1 of the Constitution. Or is it a specification of the meaning of Article 1 of the Constitution, clarifying that Estonia’s independence and sovereignty are compatible with membership in the EU, ie that these two fundamental characteristics are not to be understood so as to be compromised by joining an organisation such as the EU? It stands to reason that this kind of specification would, strictly speaking, also be an amendment to Article 1 of the Constitution and that the accession referendum was therefore not optional – as was sometimes implied in the pre-accession debate – but obligatory. This follows from Article 162 of the Constitution, which provides that ‘Chapter I (General Provisions) and Chapter XV (Amendment of the Constitution) of the Constitution may only be amended by referendum.’ What is at stake here? Most importantly, the two conceptions can lead to different conclusions regarding the obligation to hold a new referendum once Estonia begins to contemplate the signing of a new treaty that would entail a closer form of integration among the members of the EU. If the Supplementary Act’s statement that ‘Estonia may belong to the European Union’ is construed as an exception with respect to Article 1 of the main text of the Constitution (‘Estonia is an independent and sovereign democratic republic …’), then it appears that any additional loss of sovereignty would require a new amendment and therefore a referendum. Things stand differently if Article 1 of the Supplementary Act is a mere specification. Under this second reading, a referendum could also be mandatory, but it would depend on how the following question is answered: Is membership in the new form of organisation produced by the treaty waiting to be signed also compatible with Estonia’s independence and sovereignty (as membership in the then EU was affirmed to be in the Supplementary Act). If not, Article 1 of the Constitution needs to be amended once again and this can only happen through a referendum.

V.  Constitutional Limits to EU Integration: Exceptionless Primacy of EU Law As noted above, the limiting clause in the accession amendment was couched in terms not free from ambiguity. Article 1 of the Supplementary Act reads as follows: ‘Estonia may belong to the European Union, provided the fundamental principles of the Constitution of the Republic of Estonia are respected.’ Quite apart from the question as to which principles are fundamental, the



32 CRCSC,

Opinion 3-4-1-3-06 (n 19) para 14.

Estonia  183 formulation leaves open the nature of the limitation implied by the word ‘provided’. Does it mean that what is conditional is Estonia’s membership in the EU, so that, if it is found that the fundamental principles of the Constitution cease to be respected, the Supplementary Act mandates withdrawal from the Union? Or, are we dealing with a limitation on the applicability of individual EU acts? Sparse and allusive as it is, the judgment of the Supreme Court suggests something closer to the first reading. Yet, in the decisions of the Court, the word ‘provided’ has received an interpretation that makes it into a Solange with a difference. The threat is not expected to come from the lowering of the standard of protection accorded to fundamental rights by the Court of Justice of the European Union (CJEU), but exclusively from Treaty changes. Binding EU law is immune to any challenge in the Estonian legal system, the Supreme Court has proclaimed. Fundamental principles of the Estonian Constitution are allowed to come into play only at the moment when the present European Union undergoes a legal transformation in the hands of the Member States. It was not until several years after accession, in 2006, that the Court chose to specify the position of EU law within the Estonian legal system. The Supplementary Act provided that the Constitution is to be applied ‘without prejudice to the rights and obligations arising under the Accession Treaty’. The wording left open the question as to what exactly does it mean to interpret the Constitution in an EU-friendly manner, ie in a way that allows Estonia to fulfil its Treaty obligations. Such lack of clarity soon created difficulties for MPs. A statutory provision was therefore introduced in 2005 according to which Parliament may request an opinion of the Supreme Court on how to interpret the Constitution in conjunction with EU law. The possibility of asking for the Court’s interpretation was explicitly limited to situations where obtaining it is of critical importance for the adoption of a bill needed to enable Estonia to fulfil its obligations under the Treaty. Parliament very soon took advantage of this novel procedure. At the beginning of 2006, when making legislative arrangements for the adoption of the euro (an obligation under the Treaties), it decided to solicit the opinion of the Supreme Court on the interpretation to be given to Article 111 of the Constitution, which reads: ‘The Bank of Estonia has exclusive right to issue Estonian currency. The Bank of Estonia administers the circulation and upholds the stability of the currency.’ It was obvious that this statement would become a misrepresentation of the true legal state of affairs upon Estonia’s entry into the eurozone. It therefore represented an especially glaring instance of a contradiction difficult to simply interpret away. The Supreme Court took this opportunity to embark on a more general reflection on the effects of accession, setting out to clarify ‘what the adoption by referendum of the amendment to the Constitution entailed for the Estonian constitutional order as a whole.’33 The Court remarked that EU law has now ‘became one of the grounds for the interpretation and application of the Constitution.’ The Court went on to declare that the Supplementary Act had the effect of amending the Constitution ‘to the extent to which it does not correspond to EU law.’ It made clear, however, that this only meant primacy in application and not – as the word ‘amend’ could suggest – loss of validity of those constitutional provisions that run afoul of EU rules: To find out which part of the Constitution is applicable, it has to be interpreted in conjunction with European Union law, which became binding for Estonia through the Accession Treaty. Only that part of the Constitution is applicable, which is in conformity with European Union law or which regulates relationships that are not regulated by European Union law. The effect of those provisions of the Constitution that are not compatible with European Union law and are thus inapplicable, is suspended. This means that within the spheres of the exclusive competence of the European Union or of shared



33 CRCSC,

Opinion 3-4-1-3-06 (n 19) para 13.

184  Hent Kalmo and Anneli Albi competence with the European Union, European Union law shall apply in the case of a conflict between Estonian legislation, including the Constitution, with European Union law.

This was a resounding statement of the primacy of EU law, with no explicit exceptions being made for any part of the Constitution. Should a conflict occur in an area of exclusive or shared competence of the EU, its law would always trump the Estonian Constitution. As was only to be expected, there were opposing voices on the bench. One dissenting opinion rebuked the majority for having ‘overrated’ the primacy of EU law in the Estonian legal system.34 Another deplored the exclusive emphasis on Article 2 of the Supplementing Act, whilst Article 1 of the Constitution (ie the requirement to respect the ‘fundamental principles of the Constitution of the Republic of Estonia’) had been given no consideration at all by the majority.35 Judge Kergandberg remarked that an analysis of the respective roles of Article  1 and Article  2 of the Supplementary Act would have been all the more necessary since the Court had explicitly proceeded to set forth ‘what the adoption by referendum of the amendment to the Constitution entailed for the Estonian constitutional order as a whole.’36 Placed in such a context, the lack of a reference to any limitations was bound to be seen as significant. ‘How is it possible’, one prominent scholar asked, to take from an Act that forms part of the Constitution, and which includes two articles addressing the same topic, only one and entirely ignore the other? It doesn’t really conform to what first-year law students are taught about legal interpretation. At this level, lawyers working at the Supreme Court do not ‘sin’ against the rules of legal handicraft for nothing. This ‘sinning’ has therefore been willing; it reflects ideas and preferences.37

No doubt. But the idea might have been the prudential one of avoiding making inroads into an unfamiliar doctrinal area when it was not imperative to do so within the narrow perspective of the case.38 In other words, it was not yet obvious in 2006 that the Court’s failure to refer to the ‘fundamental principles’ mentioned in Article 1 of the Supplementary Act signalled a recognition that the primacy of EU law could never be resisted by invoking some higher constitutional criteria hiding in the background. Be that as it may, in 2008, the Supreme Court – this time the Administrative Law Chamber – reaffirmed its position of unqualified primacy from a slightly different angle. The Court addressed the question as to whether it has the power to review the constitutionality of EU measures. The ruling referred to the CJEU case law and noted that it is only for the latter to assess the legality of secondary EU law, whose ‘validity in a Member State cannot be affected by its lack of conformity to the constitution of this Member State’. If the CJEU has found secondary EU law to be in conformity with the Treaties, ‘there is no need to review the constitutionality of these Union acts and, in the general case, also of the Estonian law adopted on their basis.’39 According to the Court, the impossibility of reviewing domestic provisions ‘tied to EU law’ does not exclude subjecting ‘changes in EU Treaties, and the provisions delegating additional

34 Ibid, dissenting opinion of Justice Kõve, para 2. 35 Ibid, dissenting opinion of Justice Kergandberg, para 2. 36 See CRCSC, Opinion 3-4-1-3-06 (n 19) para 13. 37 Lauri Mälksoo, ‘Eesti suveräänsus 1988–2008’ [‘Estonian Sovereignty 1988–2008’], in H Kalmo and M Luts-Sootak (eds), Iganenud või igavene? Tekste kaasaegsest suveräänsusest [Old-fashioned or Permanent? Texts on Contemporary Sovereignty] (Tartu, Tartu Ülikooli kirjastus, 2010) 147. 38 It bears to mention, in addition, that the Court delivering the opinion was not sitting in full bench. 39 CRCSC 07.05.2008, Ruling 3-3-1-85-07, para 39. A few weeks later, the Court remarked that reviewing the constitutionality of secondary EU legislation would indeed be contrary to EU law; see CRCSC 26.06.2008, Ruling 3-4-1-5-08, para 29.

Estonia  185 powers to the EU that these changes entail, to constitutional review with the fundamental principles of the Constitution and the safeguard clause of the Supplementary Act in mind.’40 Appearing  after another affirmation of exceptionless primacy, this caveat clearly carried the implication that the fundamental principles of the Constitution are not a counterweight to the applicability of binding EU law. The review under these principles is limited to the stage where what is under scrutiny is the constitutionality of the adoption of a new EU Treaty (or an amendment to existing Treaties). There is something puzzling about the Court’s reading of the Supplementary Act. It is not that its statements lack internal consistency, nor the fact that they very substantially reduce the scope of constitutional review, as many Estonian scholars have critically observed. It is rather the apparent assumption – suggested by the language used in the 2008 ruling of the Administrative Law Chamber – that future changes to EU Treaties are to be assessed only under the fundamental principles of the Constitution and not under the Constitution as a whole. Article 123 of the Constitution provides that ‘the Republic of Estonia may not enter into international treaties which are in conflict with the Constitution.’ Before Estonia joined the European Union, and before the Supplementary Act was adopted by a referendum in 2003, it was generally accepted that Article 123 of the Constitution required treaties binding upon Estonia to be in conformity with all parts of the Constitution. How is the signing of a new EU Treaty different from accession? The Supreme Court’s reasoning seems to be premised on the idea that, read in conjunction with the Supplementary Act, Article 123 of the Constitution now dictates a more lenient test for all legal steps somehow covered by membership in the EU. Yet the signing of a new Treaty is not required by EU law. If it were, primacy would shield it from any kind of review by Estonian courts. Perhaps the Court’s meaning is that changes to the EU Treaties are not like any other international treaty – they are part of a process which has received a particular constitutional status through the adoption of Article 2 of the Supplementary Act. If this is indeed the Supreme Court’s understanding of the Supplementary Act, then it becomes especially important to identify the set of fundamental principles, because the latter will serve as a yardstick for assessing future proposed Treaty provisions. In fact, the issue of fundamental principles came up precisely in such a context in 2004 when legal experts were asked to consider whether Estonia could sign the draft Treaty Establishing a Constitution for Europe without amending the Constitution. The expert group concluded in its final report that the Treaty ‘is by its nature a constitutional act in a wider sense, which has to be considered as an international treaty for the purpose of the Constitution.’41 The experts asserted that the constitutional review of a new Treaty (or of changes to existing Treaties) does not cover only the fundamental principles of the Constitution. In other words, they embraced a position different from the one we tentatively ascribed to the Supreme Court, in that the experts did not consider the Treaty Establishing a Constitution for Europe as falling into a special category of EU treaties, subject to a less extensive scrutiny. The experts consulted in 2004, nonetheless, thought they had other reasons for seeking to list the fundamental principles of the Constitution. By contrast with the Supreme Court, they took the fundamental principles to constitute a limit to primacy. Intriguingly, the experts’ report also mentioned that the fundamental principles have ‘an irremovable legal force’.



40 CRCSC 41 Report

Ruling 3-3-1-85-07 (fn 39) para 39. of the working group on the European Constitutional Treaty, 15 December 2005.

186  Hent Kalmo and Anneli Albi This idea has not stimulated much discussion. On the face of it, the notion that some constitutional provisions enjoy ‘an irremovable legal force’ appears very difficult to reconcile with the text of the Constitution. Chapter XV, which sets out the procedures for amendment, describes no provisions as being immutable. There is thus little doubt that every Article of the Constitution can be amended, although some through a referendum only. Where does that leave the ‘irremovable legal force’ of the fundamental principles? This is not the place to probe deeper into this theoretical conundrum. Let us merely note with some regret that Estonian scholars have shown much less interest in the character of the fundamental principles than in drawing up various lists of them. In any case, on both points, they have received hardly any guidance from the Supreme Court.

VI.  Constitutional Rules and the Practice of Implementing EU Law The Supreme Court has also teased out the implications of Estonia’s accession to the EU for its system of constitutional review. First, it settled the question as to whether the Chancellor of Justice is empowered to challenge the validity of domestic statutes on the grounds of their lack of conformity with EU law. In a decision rendered in 2005, the Supreme Court stated that, under existing law, compliance with EU norms cannot be enforced through the mechanism of constitutional review. According to the Court, there are different possibilities for bringing national law in conformity with the European Union law and neither the Constitution nor the European Union law provide for the existence of constitutional review proceedings for this purpose. The European Union law has indeed supremacy over Estonian law, but taking into account the case-law of the European Court of Justice, this means the supremacy of application. The supremacy of application means that the national act which is in conflict with the European Union law should be set aside in a concrete dispute …42

The Court added: The legislator is competent to decide whether it wishes to regulate the procedure for declaring invalid Estonian legislation which is in conflict with European Union law, just as the legislator is free to choose whether it will or will not give the Chancellor of Justice the right to review the conformity of national legislation with European Union law.43

The Court has not granted any particular status to the EU Charter of Fundamental Rights (CFR) in this respect. A lack of conformity with the Charter is, in itself, not a ground for invalidating a domestic provision (as opposed to disapplying it). Nevertheless, the Court has used the Charter repeatedly to interpret the Estonian Constitution.44 The judgments of the Court are sometimes mistakenly assumed to have precluded the Chancellor of Justice from reviewing the compatibility of Estonian law with EU instruments. This was not quite what the Court said. Its conclusion may be represented as resulting from backward reasoning: if the Supreme Court itself lacks the power to invalidate a domestic rule on the ground that it is contrary to EU law (as opposed to disapplying it), it makes little sense

42 CRCSC

19.04.2005, Judgment 3-4-1-1-05, para 49. para 50. eg CRCSC 07.06.2011, Judgment 3-4-1-12-10, para 33.

43 Ibid, 44 See

Estonia  187 to recognise the right of the Chancellor of Justice to ask the Court to do this.45 The Chancellor of Justice can scrutinise domestic rules for their conformity with EU law, but may not turn to the Supreme Court with an application to invalidate such provisions (until given the respective power by the legislator). The Supreme Court followed a similar approach in 2008 when declaring that, after accession, some domestic provisions became immune to constitutional scrutiny because of their link to EU law.46 The decisive consideration was the fact that invalidating a domestic rule ‘tied to EU law’ could lead Estonia to breach its obligations under the Accession Treaty. Constitutional scrutiny of such provisions could thus be seen to lack a meaningful outcome. In order to ensure that ‘the Constitution of the Republic of Estonia is applied without prejudice to the rights and obligations arising from the Accession Treaty’ (as the Supplementary Act requires), the Supreme Court disapplied the procedural rule empowering it to strike down an unconstitutional domestic provision: The competence of the Supreme Court to declare invalid any Act or other legislation which is in conflict with the provisions or spirit of the Constitution, arises from § 152(2) of the Constitution and from the Constitutional Review Court Procedure Act, adopted for the implementation of the provision. If the Supreme Court exercised this competence in regard to the provisions of legislation of general application relating to EU law, it would create a conflict with European Union law.47

The reasoning of the Court on this point is in line with the judgments of the highest courts in other Member States that have found that, if a domestic provision simply reproduces an EU rule, then its review would effectively amount to a review of the EU rule itself. Yet, according to the definition offered by the Court, an Estonian provision does not need to reproduce EU law in order to become tied to it. A tied provision is one which ‘derives from EU law, clarifies or implements it’.48 At first blush this definition appears to imply that an Estonian provision becomes tied only if it was adopted with a view to either clarifying or implementing an EU act. Not so. In fact, even an Estonian rule pre-dating the accession can become tied to EU law, and hence immune to constitutional scrutiny, if it spells out an obligation deriving from an EU act. The possibility of such indirect and often unsuspected links has added a completely new layer to the activity of the courts and the Chancellor of Justice. Upon receiving a complaint that either a statute, an administrative regulation or an Act of a municipality is unconstitutional, they must now first satisfy themselves that, behind the challenged rule, there is not lurking some EU provision that would remain unapplied if the domestic rule was invalidated. It seems that the best way to understand the decision of the Supreme Court and to steer clear of all such complexities is to continually refer back to the reason why the so-called tied provisions enjoy immunity from constitutional review. As already said, it is the need to avoid an indirect scrutiny of EU law or, more precisely, the risk of leaving Estonia’s obligations under EU 45 ‘As the Supreme Court en banc has no possibility, on the basis of a request of the Chancellor of Justice, to declare national legal acts invalid because of conflict with the European Union law, the petition of the Chancellor of Justice regarding § 5(1) of the PPA shall be dismissed.’ CRCSC Judgment 3-4-1-1-05 (n 42) para 51. 46 CRCSC Ruling 3-3-1-85-07 (n 39) para 36. 47 Ibid, para 30. Apparently, in the light of its earlier case law, the Court could have also reached the conclusion that domestic provisions ‘tied to EU law’ simply cannot be unconstitutional, for the Constitution itself needs to be applied without prejudice to obligations under EU law. One would be dealing with the interesting phenomenon of the primacy of EU law extending to all domestic rules implementing it. Since the Court had earlier embraced a doctrine of absolute primacy, and had explicitly included the Constitution, its 2008 decision looks like an attempt to spell out the necessary consequences of what it had already said in 2006. 48 CRCSC Ruling 3-3-1-85-07 (n 39) para 38.

188  Hent Kalmo and Anneli Albi law unfulfilled. Rather than adhering to some formal definition of the notion of a ‘tied provision’, the question to ask is whether this risk is present in the case of the provision that is challenged. If it is, then it follows that the provision is tied. If not, then the way is clear for a review, with the possibility of taking the case to the Supreme Court. The same point may be expressed differently by saying that as much scope as possible for constitutional review is preserved under EU membership. Only these elements of the Estonian legal system are left out of scrutiny which are specifically demanded by EU instruments. It is not sufficient that some domestic rule or another is required for achieving the purposes of a directive. The word ‘implement’ in the formula provided by the Supreme Court must be construed narrowly and, even if it found that a rule is tied to EU law under this narrow definition, those aspects of the provision are still subject to constitutional review which are not determined by EU law: the mode of its adoption, other elements of formal constitutionality, etc. In sum, the Supreme Court can be seen as insisting on the duty of the Estonian legislator to continue to conform to the Constitution in all those interstitial areas that are not strictly governed by EU rules. Indeed, the Court has emphasised that a domestic provision does not become tied to EU law by the mere fact of being in conformity with it. The Court noted: ‘It does not follow from the fact that provisions are in conformity with EU law that these same provisions are in conformity with the Estonian Constitution, nor that EU law would be infringed by declaring such a provision unconstitutional and invalid.’49 As it happened, the Estonian Government was to embrace a much more expansive reading of the notion of ‘tied provisions’ when it defended the constitutionality of the Treaty Establishing the European Stability Mechanism (ESM) in 2012. The Chancellor of Justice had taken the view that the ESM Treaty contained a provision that was contrary to the Constitution because, under the emergency procedure foreseen by the Treaty, Estonia could incur very significant financial liabilities towards the ESM without the express approval of Parliament. The case came before the Supreme Court, which upheld – with a very narrow majority – the constitutionality of the Estonian statute ratifying the Treaty.50 The substance of the controversy will be explained below in more detail. What is important for present purposes is the claim of the Estonian Government that the impugned provision of the Treaty was outside the scope of constitutional review because it was part of ‘European Union affairs’. Significantly, this latter expression was taken from the statute regulating the activities of the EU Affairs Committee of the Estonian Parliament. The claim of the Government rested on the assumption that the set of tied provisions (in the sense discussed above) is equivalent to ‘European Union affairs’ within the meaning of the Standing Orders of the Estonian Parliament.51 Technical as all this may sound, it is in fact an enormously important issue. It hardly needs mentioning that the area of domestic law that is entirely untouched – whether directly or indirectly – by EU legislation is shrinking rapidly. Should the category of tied provisions be expanded to cover all parts of Estonian law that have a link to the policies of the EU or can, in a very broad sense, be described as ‘implementing’ such policies, the result will be a very considerable contraction of the extent of constitutional review. This remains a mere hypothesis, however. The Supreme Court did not follow the Government’s position on this point and is indeed very unlikely to do so in the future for, as mentioned earlier, it would directly contradict its own definition of a tied provision. 49 CRCSC 15.12.2015, Judgment 3-2-1-71-14, para 81. 50 CRCSC Ruling 3-3-1-85-07 (fn 39) para 39. 51 Given this view, it was hardly surprising that the Estonian Government argued that its position found support in the decision of the German Bundesverfassungsgericht which declared the ESM to belong to the group of European matters on which the German Government is obliged to report to the Bundestag on a regular basis.

Estonia  189

VII.  Resulting Relationship between EU Law and National Law: Perspectives Regarding Substantive Constitutional Law A. The ESM Treaty Case: The Very Large Financial Liabilities and the Concerns of Nine Dissenting Judges Regarding the Democratic, Rule-of-law-based Social State So far in the present chapter Hent Kalmo has traced Estonia’s journey from its post-independence Constitution to EU membership in relation to two main areas: (a) the changing approach to sovereignty, supremacy and the binding nature of the provisions of the Constitution, and (b) the curtailing of the formal scope of constitutional review. Here in the final section Anneli Albi will outline some areas where substantive constitutional values – from democracy and the social state to fundamental rights and rule of law safeguards – have come under strain in the context of EU law, which has further reinforced the broader ongoing shift from rules to pragmatism in the Estonian constitutional order. It was not in relation to EU law proper, but in a case concerning the constitutionality of a treaty – the ESM Treaty – that the Euro-friendliness of the Supreme Court was tested in the most conspicuous manner. The case raised, in particular, the question as to the limits that the requirement of sovereignty – along with the requirements of the social and democratic state based on the rule of law – may set to European integration in a broader sense (ie not only with regard to primacy). The case also highlighted the broader political considerations which underlay Estonia’s adherence to the doctrine of absolute primacy. The unconditionally EU-friendly approach has been part of a broader quest of Estonia to prove that the country belongs to Europe; Estonians are keen to be regarded as good Europeans. It has been a priority for Estonian institutions to seek to be at the top of the European Commission scoreboards in the implementation of EU law. For example, Estonia has been the fastest country to extradite individuals under the European Arrest Warrant,52 and, unlike in the case of several other Member States, there are no specific fundamental rights grounds for refusal of extradition. The Estonian political elite has also been keen to proceed to a federal Europe. This distinctly EU-friendly approach has an additional background element, which was often mentioned in the media coverage on the ESM case and pointed out by different ministers during the ESM court proceedings. It is the continued and acute sense of anxiety about the fragility and precariousness of Estonia’s independence, given its geographical position on the border of Russia, and the recent annexation of Crimea and the war in Eastern Ukraine. The EU is regarded as the main (even if soft) security guarantee. There is a very real sense of responsibility amongst politicians and diplomats that any constitutional challenges to EU law, let alone any hint in the international media that Estonia might be uncooperative or Eurosceptic, could undermine potential support for the country in the future, should it be needed. It is against this background that the change of approach by the Supreme Court in the case regarding the ratification of the ESM Treaty53 shows the particularly heightened sensitivity of the constitutional values at stake. The legal challenge was brought by the then Chancellor of

52 Julia Antonova, Maila Pardla and Imbi Markus, ‘Estonia’, in J Laffranque (ed), Reports of the XXV FIDE Congress Tallinn 2012, Vol 3: The Area of Freedom, Security and Justice, Including Information Society Issues (Tartu, Tartu University Press, 2012) 324. 53 CRCSC 12.07.2012, Ruling 3-4-1-6-12 on the ESM Treaty.

190  Hent Kalmo and Anneli Albi Justice Indrek Teder, who was concerned about the unprecedented magnitude of the financial liabilities under the ESM Treaty. According to revised figures in 2012, Estonia’s total liability that can be called in by the ESM amounts to 1.79 billion EUR, which in 2012 corresponded to about 11.7 per cent of its GDP and 29 per cent of the annual state budget.54 In comparison, the Chancellor of Justice pointed out that Estonia’s liabilities to the IMF were 108 million EUR, or about 0.7  per  cent of Estonia’s annual GDP. The Chancellor of Justice’s concerns included the fact that the use of such funds is not subject to democratic control or judicial review. The main, central concern was that the 85 per cent majority rule in the case of emergency voting under Article 4(4) of the ESM Treaty means that smaller states have no say over the use of the funds, and consequently over the likelihood and the size of the liabilities that may be called in from the Estonian state. The Chancellor of Justice additionally outlined potential scenarios where under the ESM Treaty instances may arise where Estonia’s liabilities, at least temporarily, could increase beyond the above maximum limit. Overall, according to the Chancellor of Justice, the interference with parliamentary democracy and Parliament’s budgetary reservation is ‘very intensive’. The Chancellor of Justice noted that the ESM Treaty has the potential to seriously limit the very ability of the state to ensure the functioning of state institutions, including the judiciary, and the protection of the rights and social benefits envisaged under the Constitution. The amount of the financial liabilities also caused concern in other Member States and led to adjudication in some other constitutional courts and supreme national courts. For example, the liabilities correspond to about a third of the state budget in Ireland,55 and in Germany to two-thirds of the federal level state budget.56 Questions were also raised as to whether an upper limit exists to the liabilities. Indeed, in Germany the applicants were concerned about the state taking ‘incalculable risks’, with Parliament no longer being able to exercise overall budgetary responsibility.57 The German Constitutional Court noted the ambiguities in the Treaty’s wording about whether a maximum limit exists, and requested Parliament to address this issue in the ratification process. Some of the above issues – albeit primarily with a view to questions of legal base and competence – were referred to the CJEU by the Irish High Court in Pringle. The CJEU approved the legality of the Treaty.58 However, it did so in a way that sparked a discussion in the literature about the CJEU’s approach of ‘if anything goes, then indeed anything goes’.59 Whilst the Estonian Supreme Court also approved the ratification, it did so by a notably narrow majority or 10 judges against 9. There were five separate opinions, of which four were dissenting opinions: a joint opinion of six judges (Justices Jõks, Järvesaar, Kergandberg, Kivi, Kull and Laarma), and further individual dissenting opinions by Justices Ilvest, Luik and Tampuu. The final separate opinion was that of Justice Kõve, who agreed with the Supreme Court en banc decision but expressed the view that the case ought to have been declared inadmissible. As this had

54 These calculations were provided by A Albi in her expert opinion of 23 April 2012, submitted to the Supreme Court on the request of the Court in Case No 3-4-1-6-12 on the ESM Treaty. These calculations revised the initial figures that had been provided in the application of the Chancellor of Justice, as the figures had subsequently been revised at the EU level. 55 MP Pringle, who brought the ESM proceedings in Ireland, pointed out that Ireland’s liability is 11 billion EUR in various forms of capital, which is equivalent to approximately one third of government tax revenue for 2011. See ‘Local Pringle takes legal challenge against government on EU treaties’, 17.04.2012 (www.thomaspringle.ie/?p=697). 56 Around 190 billion EUR; the annual budget of the Federal Republic without the budgets of the Länder is around 300 billion EUR. 57 BVerfG 18.03.2014, 2 BvR 1390/12 ECLI:DE:BVerfG:2014:rs20140318.2bvr139012 (ESM). 58 CJEU 27.11.2012 C-370/12 (Pringle) ECLI:EU:C:2012:756. 59 See Michal Bobek, ‘The Legal Reasoning of the Court of Justice of the EU’, (2014) 39 European Law Review 423, interpreting the concerns about the CJEU’s expansive, contra legem legal interpretation expressed by Gunnar Beck in The Legal Reasoning of the Court of Justice of the EU (Oxford, Hart, 2012) 446 ff.

Estonia  191 not been done, Justice Kõve agreed with the majority view. However, Justice Kõve additionally recorded his uncertainty about whether a maximum limit to the liabilities exists. The joint opinion of six dissenting judges – Justices Jõks, Järvesaar, Kergandberg, Kivi, Kull and Laarma – found that the Chancellor of Justice’s request should have been upheld.60 They expressed concern that the ESM judgment, whilst being ‘the most important in our constitutional review history so far’, ‘does not reflect by far all the important aspects of the problems which are the object of the court case and extremely significant from the point of view of Estonian statehood’.61 According to the dissenting judges, the case was decided ‘in a rush’, without any risk analysis, and the review of constitutionality was ‘with a very low intensity’.62 The six dissenting judges found that the ESM Treaty contravened ‘the principle of a state subject to the rule of law which is one of the most substantial principles’, and, further, that ‘[t]here is no doubt that the implementation of the Treaty harms the principle of a social state arising from §§ 10 and 28 of the Constitution’.63 Justice Ilvest, in an individual dissenting opinion, expressed support to the above dissenting opinion of six judges, and additionally found that a referendum would have been required to ratify the Treaty. Justice Ilvest posed the question whether the constitutional sovereign – the people – has, through a single referendum … granted to any composition of the Riigikogu holding office the right to waive, forever and ever, on current political considerations the sovereignty of the state of Estonia to any extent … I am of the opinion that the Constitution of the Republic of Estonia Amendment Act does not give the Riigikogu the right to ratify the Treaty without holding a referendum.64

In a further dissenting opinion, Justice Luik highlighted what in the view of this co-author really goes to the heart of why the ESM Treaty raises profound questions about justice and fairness, especially for low-income countries such as Estonia. Justice Luik started, similarly to the six dissenting judges above, with a concern that the Supreme Court ‘was forced to make a rushed judgment in a situation of confusion, uncertainty and legal vagueness dominating in the legal space of both Estonia and the European Union’. The judgment left him ‘overwhelmed with concern, perplexity and puzzlement’.65 Noting that the judgment ‘does not contain a word about the nature of the global, including of the euro area’s financial crisis, about political, economic and ideologic[al] reasons and about the actual situation’, Justice Luik made the following observation: As of the adoption of the euro the consumer prices have constantly risen in Estonia … Alleged economic success is seeming: Estonia is without a doubt one of the poorest countries in the euro area. A rise in prices due to opening of the electricity market lies ahead. The Constitution obliges us to develop the state which shall ensure to present and future generations social progress and welfare (preamble to the Constitution). Is social progress and welfare to be understood as a guarantee of well-being for few?66

Justice Luik continues: The Treaty attributes an opposite value also to the principle of solidarity: Estonia undertakes to guarantee with the taxpayer’s money the sustainability of the states of the euro area which are many times 60 CRCSC, Ruling 3-4-1-6-12 (fn 53) Dissenting opinion of Justices Jõks, Järvesaar, Kergandberg, Kivi, Kull and Laarma, introductory paragraphs. 61 Ibid, introductory paragraphs. Here and subsequently the unofficial translation provided on the Supreme Court’s website noted earlier is used, with some revisions by Anneli Albi. 62 Ibid, introductory paragraphs and paras 6 and 9. 63 Ibid, paras 8, 11. 64 Ibid, dissenting opinion of Justice Ilvest. 65 Ibid, dissenting opinion of Justice Luik, introductory paragraph. 66 Ibid, para 16.

192  Hent Kalmo and Anneli Albi wealthier than Estonia, including the sustainability of the private sector (banks) of the said states. The idea probably is that the inhabitants of those states are used to well-being which shall be maintained as long as possible. The living standard in Estonia may safely decrease. Estonia is a small country, the people are patient and understanding: it laboured to adopt the euro to meet the Maastricht criteria, when the economic depression came and in the interests of budget balance it agreed to restrictions without complaining. Is it not unjust? Or does it in the contemporary interpretation mean a state founded on liberty, justice and the rule of law (preamble to the Constitution). I am perplexed.67

By way of an addendum to Justice Luik’s observations, it is important to bear in mind that Estonia has over the years rigorously maintained a balanced budget regime, even where this has been at the cost of severe reductions to the already low child benefits, sickness, unemployment and other benefits as well as public sector salaries. It is clear that if the financial liabilities under the ESM Treaty should have to be paid – a scenario that the Finance Minister at the time insisted was only ‘extremely theoretical’68 but the likelihood of which increased after Greece threatened to default on its payments to the ESM in 2015 – this would cause genuine hardship for the people. The Supreme Court en banc, with a notably narrow majority of one judge, found the ESM Treaty to be constitutional. On the one hand, the Court did acknowledge that the Treaty interferes with the budgetary powers of Parliament in a way that ‘brings about also an interference with the principle of a democratic state subject to the rule of law and of the state’s financial sovereignty since indirectly the people’s right of discretion is restricted’.69 On the other hand, such a breach was found to be proportionate given the overriding need to ensure the financial stability of the eurozone. Furthermore, in the ESM case, the Court also explicitly adopted a modern approach to sovereignty, stating that the Constitution does not require, despite the strict wording of the sovereignty clause, observation of absolute sovereignty … [M]embership of the EU and in international organisations has become a natural part of sovereignty in this day and age.70

However, in a notable development, the Court en banc went on to state that limits may exist to European integration. According to the Court, the Supplementary Act ‘does not authorise the integration process of the European Union to be legitimised or the competence of Estonia to be delegated to the European Union to an unlimited extent’.71 According to the Court, a new legitimisation from the people by referendum may be required: If it becomes evident that the new founding treaty of the European Union or the amendment to a founding treaty of the European Union gives rise to a more extensive delegation of the competence of Estonia to the European Union and a more extensive interference with the Constitution, it is necessary to seek the approval of the holder of supreme power, i.e. the people, and presumably amend the Constitution once again. These requirements are to be considered also if the Treaty leads to amendments to the TFEU and TEU.72

In the ESM case, the Estonian Supreme Court was under exceptionally great pressure, as were also the highest courts elsewhere. Indeed, the possibility that the Treaty might not be ratified was

67 Ibid, para 17. 68 ‘Ligi: ESMi risk eelarvele on tagasihoidlik ja teoreetiline’ [‘Ligi: The Risk of ESM on the budget is modest and theoretical’], E24 Majandusuudised 12.03.2012 (www.e24.ee/770926/ligi-esmi-risk-eelarvele-on-tagasihoidlik-ja-teoreetiline/). 69 CRCSC, Ruling 3-4-1-6-12 (fn 53) para 153. 70 Ibid, para 130. Here and subsequently the unofficial translation provided on the Supreme Court’s website noted earlier is used. 71 Ibid, para 223 (emphasis added). 72 Ibid, para 223.

Estonia  193 seen in the media as leading to the country’s ‘isolation’ and pariah status in Europe, and the case was widely disparaged as ‘embarrassing’.73

B.  Decline of the Post-totalitarian, German Rechtsstaat Type of Constitutionalism through EU Law: Judicial Protection, the Principle of Legitimate Expectations, Publication of Laws and Other Rule of Law Safeguards Whilst the ESM Treaty brought into the limelight the strain caused by EU integration on the democratic and social state, over the years of EU membership there have been cases in several further areas where due to EU requirements, key elements of substantive Estonian constitutional law and rule of law safeguards have seen a decline. It would seem important to briefly outline these to invite discussion for several reasons. First, such cases have by and large not received attention in the literature due to the prevailing focus on sovereignty, national constitutional i­dentity and loyal implementation. Secondly, when looking at the cases in a joined-up manner, they point to a rather different understanding of the rule of law in EU law and in Estonian law; it will be pointed out that the latter reflects the Rechtsstaat approach to the rule of law which is adhered to more widely in post-totalitarian and post-authoritarian parts of Europe. Thirdly, the rule of law safeguards and their judicial protection in Estonia additionally need greater cognizance as the prevailing view in much of the EU and transnational discourses has erroneously been that the new Member States from Central Eastern Europe have never really lived up to the Western rule of law standards, and that the illiberal turn in Hungary and Poland, rather than being an aberration, is simply a clearer illustration of this. A brief note about the recent illiberal turn in parts of Central and Eastern Europe, including the rise of a more extreme party in Estonia, will be added at the end of this section. The substantive aspects of Estonian constitutional law have their origin in what the author of this part of the chapter, Anneli Albi, has described as a ‘post-totalitarian constitutional culture’, which is based on the German model of the constitutional Rechtsstaat.74 In fact, the eminent Italian constitutional scholar Cesare Pinelli regards this broader constitutional culture as a ‘continental Europe’s constitutional tradition’ (emphasis added), for the understanding of which a central necessary element is the collapse of the totalitarian regimes.75 Pinelli places in the posttotalitarian European constitutional tradition the constitutions of Germany, Italy, Spain, Portugal and Greece, and finds that this approach has subsequently been consolidated by its inclusion into the design of the post-communist constitutions of Central and Eastern Europe. Whilst some of the constitutions may be better described as post-authoritarian rather than post-totalitarian, they share a number of characteristic features, and for ease of reference will be referred to as ‘post-totalitarian constitutions’. The author of this part of the chapter has elsewhere

73 Sulev Vedler, ‘Riikliku julgeoleku küsimus’ [‘A Matter of National Security’], Eesti Ekspress 13.04.2012 (www.ekspress. ee/news/arvamus/arvamus/sulev-vedler-riikliku-julgeoleku-kusimus.d?id=64238735); ‘Ligi ESMist: Eesti suutis negatiivselt üllatada’ [‘Ligi on the ESM: Estonia Managed to Surprise Negatively’] Äripäev Online 14.03.2012 (http://live.ap3.ee/ article/2012/3/14/ligi-esmist-eesti-suutis-negatiivselt-ullatada). 74 Anneli Albi and Samo Bardutzky, ‘Revisiting the Role and Future of National Constitutions in European and Global Governance: Introduction to the Research Project’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law – National Reports (The Hague, TMC Asser Press, 2019) 13 ff. 75 Cesare Pinelli, ‘The Formation of a Constitutional Tradition in Continental Europe since World War II’, (2016) 22 European Public Law 257, 258.

194  Hent Kalmo and Anneli Albi identified a list of 12 typical features of the post-totalitarian or post-authoritarian constitutions. These features include (1) strong normative, binding force; (2) detailed and extensive bills of fundamental rights, which in many respects are more extensive and stringently worded than the ECHR and the CFR; (3) directly applicable and justiciable nature of the rights; (4) a set of core provisions that are unamendable or can be amended only by a more stringent procedure, such as recourse to a referendum; (5) strong mandate for constitutional review by the constitutional courts; (6) frequent annulment by the constitutional courts of legislative measures on the grounds of fundamental rights and the rule of law; and (7) the explicitly written duty for the courts and state institutions to protect fundamental rights.76 Furthermore, many of the rule of law safeguards and sub-principles are expressly written out in the constitutions and are, hence, protected in the quality as rights and rules (ie binding norms) rather than as general principles of law.77 These and other features represent the ‘Never again’ ethos against the background of totalitarian or authoritarian history, providing detailed guarantees against arbitrary exercise of power and curtailing of fundamental rights and individual freedoms. In particular, in the post-totalitarian constitutions, the deprivation of personal liberty and the use of coercive power are subject to stringent constitutional safeguards, as part of the protection of human dignity. A further key feature is that this type of constitution, along with judgments of constitutional courts, pre-determine the acceptable margins for political behaviour, and the constitution needs to be amended prior to effecting a significant constitutional or institutional change.78 The eminent scholar Alexander Somek, who regards ‘constitutionalism as a project of emancipation’, regards this type of constitution as ‘Constitutionalism 2.0’.79 This approach stands in a distinctive contrast with the political or historical-type constitutional culture, which marks the approach eg in Britain, the Netherlands, and the Nordic countries Sweden, Denmark and Finland.80 In these constitutional cultures, there is no constitutional court. In the Netherlands, constitutional review is even banned by the Constitution. Fundamental rights protection takes place primarily on the basis of a historically older and more limited bill of rights or on the basis of the ECHR, with greater scope for policy discretion in the event of limitation of rights. Further key differences in comparison with the post-totalitarian constitutions include the fact that in the political constitutional cultures, change is gradually incorporated into the text of the constitution ex post facto, and that the political constitutional cultures – especially the British and Dutch constitutional cultures – are regarded as more pragmatic, leaving to the authorities significantly greater scope for discretion.81 Indeed, the British constitution is unwritten. Returning to the Estonian Constitution, one study recalls that historically it was bound to the German legal culture. Old ties awoke from hibernation after Estonia’s restoration of independence, and the most influential model for reconstruction of vast parts of the Estonian legal 76 For a more detailed account, with references to individual constitutions and respective literature, see Anneli Albi, A Comparative Study on National Constitutions in EU and Global Governance (The Hague, TMC Asser Press, forthcoming 2022). A summary is available open access in Albi/Bardutzky, ‘Revisiting’ (n 74) 17. 77 On the qualitative difference between the protection of fundamental rights formulated through binding constitutional rules on the one hand, and through the general principles of law which are subject to balancing and proportionality on the other hand, also in the judgments of the Estonian Supreme Court, see discussion between Madis Ernits and Hent Kalmo, with reference to the seminal book by Robert Alexy, A Theory of Constitutional Rights (J Rivers translation) (Oxford, OUP, 2002). See, respectively, Madis Ernits, ‘Constitutional Review in the Age of Balancing’, in the 6th International Scientific Conference ‘Konstitucionālās vērtības mūsdienu tiesiskajā telpā I’ [‘Constitutional Values in the Modern Legal Space I’], Riga 10–11 November 2014 (Riga, University of Latvia Press, 2016) 127, 128; and Hent Kalmo, ‘Põhiseadus ja proportsionaalsus – kas pilvitu kooselu?’ [‘The Constitution and Proportionality – A Cloudless Partnership?’), (2013) Juridica 79, 81. 78 See Albi, Comparative Study (n 76). 79 Somek, Cosmopolitan (n 7) 10. 80 Albi/Bardutzky, ‘Revisiting’ (n 74) 13 ff. 81 Ibid.

Estonia  195 order was German law, including in the fields of administrative and criminal law.82 Moreover, the German Basic Law was one of the main examples for the Estonian Constitution during the travaux préparatoires.83 In general, the Estonian Supreme Court, similarly to the constitutional courts of most other Central and Eastern European EU Member States as well as eg the Portuguese Constitutional Court, adheres to the notably high German constitutional protection of the Rechtsstaat-tradition of the rule of law. This includes the extensive and rigorous approach to the definition and judicial protection of the formal and substantive requirements of the rule of law and the well-established subprinciples, such as access to justice, legal certainty and legitimate expectations, non-retroactivity and the principle of the social state.84 Indeed, these sub-principles, along with fundamental rights, are a frequent ground for constitutional review and annulment of legislation. This marks a profound departure from the Communist experience, where power was exercised in an arbitrary manner, and individuals’ rights and personal liberty were severely restricted, typically by executive decrees, often enforced retroactively, and with mechanical rubberstamping by courts. During Communist times, courts also had to apply the law in an anti-formalistic, teleological way, as will be further explored below. It seems important to point out that whereas in EU law discourses national constitutional cultures are assumed to represent something that is idiosyncratic and with connotations of backwardness and emotiveness – and which is partly why progress is seen to lie in a shift to autonomous EU law85 – the post-totalitarian and post-authoritarian constitutional cultures, in fact, represent very advanced achievements in the quality of constitutionalism, including judicial protection of the individual and carefully fine-tuned checks, balances and limits on the exercise of public power. Furthermore, rather than simply cultural issues, these represent the results of hard-won battles and sacrifices on the part of many individuals, which eventually won our freedoms.86 A central place in the Estonian Constitution is held by Article 3(1), which is interpreted to require the specific enactment of a statute for every specific exercise of state power.87 According to the Supreme Court, ‘[w]hat the legislator is … obliged to do under the Constitution cannot be delegated to the executive, not even temporarily and … on the condition of judicial review’.88 In general, the Constitution is central to the public discourse. In the aftermath of the post-Communist constitutional reforms, Estonia’s legal scholars, judges and state institutions, especially the successive holders of the post of the Chancellor of Justice, have consistently and proactively placed heightened emphasis – in media articles, scholarly research, at conferences and at other forums – on the importance of the principles of the rule of law and of conducting state affairs in a manner that complies with the Constitution. 82 Ernits et al, ‘Unexpected Challenges’ (n 27) 889 f. 83 Ernits, ‘Constitutional Review’ (n 77) 145; Paloma Krõõt Tupay, Verfassung und Verfassungsänderung in Estland. Eine Analyse zu Theorie und Praxis mit vergleichenden Anmerkungen zum deutschen Recht (Berlin, BWV Berliner Wissenschafts-Verlag, 2015) 35 ff. 84 For a detailed overview in English, see Ernits et al, ‘Unexpected Challenges’ (n 27) 907 ff. 85 See nn 7–8 above and the accompanying text. 86 In fact, with some hesitation, it may be worth recalling that this is not the first time that the Estonian legal system and legal community have experienced the effects of an ideational setting that sees progress in leaving behind the national legal culture. Tõnu Tannberg, Professor of Estonian history at the University of Tartu, has extensively researched how in the Soviet Union, the goal was to fight against nationalism, including in law and legal science, and to shift law and legal science discourses to the (Soviet) Union level, along with ideological muzzling. Tõnu Tannberg, presentation ‘Õigussüsteemi rakendamine poliitilise režiimi teenistusse Eestis 1950. aastatel’ [’The deployment of the legal system in the service of a political regime in Estonia in the 1950s’], at the Estonian Lawyers 35th Congress ‘Eesti Vabariik 100 – Kaasaegne riik’ [‘Estonian Republic 100 – Modern State’], 04.10.2018, www.uttv.ee/naita?id=27592. 87 Ernits et al, ‘Unexpected Challenges’ (n 27) 889 f. 88 For this and further cases, see ibidem, 889 f, 907 ff. Some revisions have been made by Albi to the unofficial translation.

196  Hent Kalmo and Anneli Albi In the light of the above-described constitutional culture, the amendment of the Constitution by a blanket clause in the Supplementary Act, as well as the near-abandonment of constitutional review in broad areas of law in the judgments of the Estonian Supreme Court, would seem to have an even more profound and disquieting dimension. Indeed, the author of this part of the chapter, Anneli Albi, would like to express her concern that the above-described post-totalitarian, Rechtsstaat-modelled type of constitutionalism has come under strain in the context of Estonia’s EU membership. Earlier, as the author of this part of the chapter has documented elsewhere, the Estonian courts, in order to ensure an EU-friendly approach, reduced the level of protection to property rights, the principles of legal certainty, legitimate expectations and non-retroactivity, and the rule that only published laws can be valid.89 This occurred as part of the ‘Sugar saga cases’ in several new Member States, where safeguards to the rule of law and constitutional rights had to be revised downwards, in order to comply with the CJEU’s autonomous, self-referential interpretation of these rights and general principles of law, whereby a notably wide margin of discretion is granted to EU institutions, with annulments of EU acts being very rare. In a nutshell, the Estonian sugar case concerned a 45 million EUR fine on excess sugar stocks, including for sugar that had been purchased by private individuals and households. The EU fine was imposed on the basis of an EU regulation that had been adopted one year after EU accession, and no EU act had ever mentioned sugar held in private homes. The Estonian Government’s arguments resembled the interpretation of the principle of legitimate expectations in the Estonian legal system, but were systematically rejected by the CJEU on teleological and market stability grounds, and the hefty fine was upheld.90 Regarding publication of laws, through the CJEU preliminary rulings in the above sugar stocks cases, the Estonian approach that an unpublished law is void and non-existent ab initio/ex tunc, which is also more widely applied in post-totalitarian constitutional systems, was replaced by the ‘valid but not enforceable against individuals’ approach that applies in France and Belgium.91 These and many other CJEU cases also do not sit easily with the requirement of vacatio legis, ie the time period that must be left between the publication and entry into force of a measure, in order to leave the addressees of the norms time to rearrange their activity, and, thus, to ensure legal certainty and to prevent the state from enacting new regulations overnight.92 Further surprising effects that different aspects of EU law have had on the well-established level of protection under Estonian constitutional law have in subsequent years increasingly come to the attention of media, scholarship, attorneys and some state institutions. By way of examples, in the field of the European Arrest Warrant, greater discussion was sparked by a documentary on the state television channel ETV, devoted to the case of Neeme Laurits who claimed his innocence and had compelling alibis but had been extradited to Finland without judicial review.93 After having sat in a Finnish prison for nine months, and losing his job 89 Anneli Albi, ‘From the Banana saga to a Sugar Saga and Beyond: Could the Post-communist Constitutional Courts Teach the EU a Lesson in the Rule of Law?’, (2010) 47 CML Rev 791. 90 Ibid, 797. 91 See also Michal Bobek, ‘Case C-345/06, Gottfried Heinrich, Judgment of the Court of Justice (Grand Chamber) of 10 March 2009’, (2009) 46 CML Rev 2086. 92 For CJEU judgments affecting this requirement in other areas, see Eve Fink, Õiguspärase ootuse kaitse põhimõtte eeldused ja piirid Euroopa Liidu õiguses [‘The Conditions and Limits of the Protection of the Principle of Legitimate Expectations in EU Law’] (Tartu, Tartu University Press, 2016) (with English summary). For the rule of vacatio legis in the judgments of the Supreme Court of Estonia, see Eesti Vabariigi Põhiseadus. Kommenteeritud Väljaanne [‘The Constitution of the Republic of Estonia. Commented Edition’], 3rd edn (Tallinn, Juura, 2012), 658, 674. 93 ETV programme Pealtnägija [‘Eyewitness’], summarised in R Kagge, ‘Narkoparuniks tembeldatud eestlane pisteti 9 kuuks Soome vanglasse’ [‘An Estonian who was Labelled a Drug Baron Sat in a Finnish Prison for 9 Months’], ERR [English language service of ETV] 01.02.2012 (http://uudised.err.ee/index.php?06244908).

Estonia  197 and his wife and seeing his health deteriorate, Mr Laurits was eventually found innocent. Notably, Mr Laurits had sent letters to numerous institutions asking for help because he was innocent, but he received no assistance and his case was met with no interest whatsoever. The state bodies he approached had referred to the obligation of mutual recognition and mutual trust of other Member States’ legal systems, and the duty of loyal cooperation. Whilst in the EU law discourse there has been some discussion about the exceptions permissible in relation to prison conditions, it may be worth remarking that given the high standards in Finland, the concern here was not about prison conditions. Instead, the concern is about the neglect of human dignity through arbitrary deprivation of liberty, without judicial review and with disregard to the constitutional duty for the courts and state institutions to protect fundamental rights. Indeed, in the TV programme cited above, the perplexed Mr Laurits drew parallels with the forced deportation of Estonians to Siberia during the Soviet regime. After Mr Laurits’ case was highlighted in the media, an Estonian attorney, Kaido Pihlakas, drew attention to further individuals who had been extradited and had later been found innocent.94 Attorney Pihlakas expressed concern that extradition decisions are ‘rubber-stamped’ by judges. He called for granting a hearing to the accused prior to extradition, in order to avoid needless suffering by innocent persons. Following this case and on the initiative of the Estonian Bar Association, the then Chancellor of Justice Indrek Teder commenced in 2013 abstract review proceedings (within the Office of the Chancellor of Justice; ie without referral to the Supreme Court) to assess the constitutionality of the extradition rules. He considered it necessary to explore problems raised in protecting human rights in practice and to learn from any mistakes. The Chancellor of Justice recorded several concerns with regard to defence rights.95 The proceedings were continued by subsequent Chancellor of Justice Ülle Madise, without eventually finding a conflict with the Constitution. Questions about access to justice have also been raised in the media by attorneys in relation to other fields to which the system of mutual recognition has been extended, for example as regards imposition of hefty and unfair administrative fines on individuals without practical possibilities for judicial review.96 The CJEU case law regarding the obligation of mutual trust for national courts has been comprehensively researched by a young Estonian scholar, Birgit Aasa, in her PhD thesis at the European University Institute in Florence.97 Aasa’s thesis concludes with the finding that whilst enforced as an increasingly constitutional principle of the EU and used in a way that instrumentally advances the utopian, community-building objectives, trust, in fact, is not at all suitable for governing legal relations and courtroom matters. Aasa explains that institutionalisation of distrust is a precondition of the architecture of liberal democratic constitutionalism, in order to create an overall system of trust. As part of this, there are certain professions which are expected to exercise suspicion and distrust as a professional duty, including judges; courts are expected to be neutral, objective, professional institutions settling societal disputes. The CJEU requirement of blind trust leads to a culture of naivety and dysfunctionality, and this is illustrated with numerous cases on the ground in the Member States where fundamental rights and important legal values have been sidelined to ensure the EU rules on mutual trust. Aasa’s thesis goes on to make the

94 ‘Advokaat: Eesti täidab Euroopa vahistamismäärusi liiga püüdlikult’ [‘An Attorney: Estonia Fulfils the European Arrest Warrant too Eagerly’], Postimees 29.02.2012. 95 Ernits et al, ‘Unexpected Challenges’ (n 27) 913 ff. 96 See eg the case of Aivo Piirsoo and other cases in Albi, ‘Erosion’ (n 4), (Part 1) 151. 97 Birgit Aasa, ‘The Principle of Mutual Trust in EU Law: What is in a Name’? PhD thesis defended at the EUI, Florence, on 26 February 2021.

198  Hent Kalmo and Anneli Albi observation that as the EU has legalised and institutionalised the obligation of trust, this seems an expression of an autocratic rather than democratic regime.98 That constitutional and rule of law safeguards in these and other fields tend to come second in the balancing in the CJEU case law, in favour of discretion in achieving policy objectives, seems to be part of a broader change through EU law from constitutional and rule of law constraints to the prioritisation of effectiveness, the market and functionalist and neofunctionalist considerations. Michal Bobek has observed with regard to the use of the doctrine of effet utile in EU law that it entitles the EU judges ‘to do pretty much anything’; ‘purposive reasoning is often reduced to one and only one purpose: the full effectiveness of Community law, which is turned into the crucial principle not allowing for any balancing or opposition’.99 Bobek notes with regard to the extensive use of teleological interpretation and effet utile in the CJEU’s case law that ‘[h]eretical though it may sound, there are some striking similarities between the communist/Marxist and Community approaches to legal reasoning’.100 The early Stalinist phase of Marxist law required that judges disregard the remnants of the old bourgeois legal system. They had to apply the law in an anti-formalistic, teleological way, directing their aim towards the victory of the working class and the Communist revolution. In EU law, the purpose also comes first, leading to reasoning that consequentially follows from the purpose, using open-ended clauses such as effet utile; these take precedence over a textual interpretation of the written law.101 This historical background also helps to explain the reservations that the author of this part of the chapter, as well as many judges from the Central and Eastern European Member States, have with regard to effectiveness and extensive teleological reasoning, especially when imposing criminal sanctions. For example, in Poland (before the recent illiberal changes), some older judges expressed concern that the teleological interpretation reminded some of them of communist times, with one judge noting, ‘I don’t like crossing the same river twice’.102 It seems pertinent to add that in Estonia, parallels with the law of Soviet times arose sharply in relation to the gradual disappearance of the distinction between administrative penalties and criminal punishments through implementation of EU measures, which require prioritising the effectiveness of sanctions. This led to heated discussions in the media, including an article entitled ‘An attempt to revive Soviet law’, where a prominent attorney, Leon Glikman, expressed concern that the new type of proceedings, introduced to implement EU directives in the field of data protection, financial and competition law, ‘turns the whole order of proceedings upside down’ and severely curtails classic procedural rights in the name of efficiency of convictions.103 A panel with a title referring to ‘seeking to fit the unsuitable in the Estonian legal order’ was held as part of the 2020 Estonian Lawyers Congress, with observations that the direction of travel is towards ‘criministrative’ law, an odd combination of criminal and administrative law, and that regrettably this process cannot be stopped due to the threat of European Commission’s costly enforcement proceedings in the event of incorrect implementation.104 98 Ibid, with references to further literature, including the sociological theory of trust by Sztompka, and the publications of Luhmann. 99 Michal Bobek, ‘On the Application of European Law in (Not Only) the Courts of the New Member States: “Don’t Do as I Say”?’, (2007) 10 Cambridge Yearbook of European Legal Studies 1, 10, 21, citing Radoslav Procházka, ‘Prekážka rozhodnutej veci – judikatúra Súdného dvora ES a jej dopad na konanie vnútroštátnych súdov’ [‘Res iudicata – the Case law of the Court of Justice and its Impact on the Procedure before National Courts’] (2007) 10 Justičná revue, 1240, 1248. 100 Ibid, 23. 101 Ibid, 23 f. 102 As cited in Urszula Jaremba, National Judges as EU Law Judges: The Polish Civil Law System (Leiden, Brill, 2014) 222. 103 Leon Glikman: Nõukogude õiguse taaselustamise katse’ [‘Leon Glikman: An Attempt to revive Soviet law’], Postimees, 11.06.2020. 104 Panel ‘Euroopa Liidu halduskaristuste sisseviimine Eesti õigusruumi: kas sobitamatu sobitamine sobivaimas võtmes?’ [’Introduction of European Union administrative penalties into the Estonian legal order: Seeking to fit the

Estonia  199 Returning to the theme of protection of legitimate expectations, which is a core rule of law principle in the Estonian legal order, a significant change through EU law has been documented in a recent doctoral thesis by Eve Fink, a practising lawyer in the field of real estate law in Estonia, who explored the CJEU’s case law and relevant literature regarding the principle of legitimate expectations.105 Fink’s thesis documents the circumstances that have led to the finding of a breach of the principle of legitimate expectations in the review of validity of EU measures (and with a spill-over to the national measures implementing EU law), finding that the protection of legitimate expectations is in fact statistically very rare and exceptional in the CJEU case law. Furthermore, the protection is subject to remarkably different and narrow conditions, which do not exist in the Estonian constitutional case law in terms of some key elements. Although the study by Fink overall finds that the CJEU’s interpretations are justified on the basis of EU public interest grounds such as competition and the functioning of the single market, her study nevertheless concludes that the CJEU’s rulings in this field set the ‘threshold of protection of legitimate expectations very high’ and meet the requirements of the rule of law ‘only at a minimal level’.106 A recent article by a Spanish scholar, Pablo Rodríquez, seems to indicate that the concern about the change in the protection of this principle is present in the wider Rechtsstaat-based constitutional area. Rodríquez expresses his concern – through the title of his article – that legal certainty is ‘A Missing Piece of European Emergency Law’.107 Rodríquez notes that in the euro crisis adjudication, ‘legal certainty and legitimate expectations have succeeded in … some Constitutional Courts and in the European Committee of Social Rights, but the European Court of Justice appears still to remain rather impervious to them’.108 One could add that the Portuguese Constitutional Court has come under fierce criticism for annulment of the EU and IMF-required drastic austerity measures, with the annulments having inter alia been on the grounds of the principle of legitimate expectations.109 What has by and large been unnoticed is that this type of protection is in fact applied by the constitutional courts of Germany, Latvia, Estonia, the Czech Republic, Slovakia, Romania, Slovenia, Portugal and many other countries. Thus, the EU requirements – rather than affecting an idiosyncratic feature of national constitutional identity – have put under strain a well-consolidated principle of Europe’s post-totalitarian and post-authoritarian constitutional systems. Regarding the standard of constitutional review, in Estonia, the Constitutional Review Chamber declared an act or a provision unconstitutional in 75.5 per cent of the cases (68 out of 90 cases) in the period 1993–2004.110 In 2010–13 the Estonian Supreme Court declared a norm or absence of a norm unconstitutional on 72 occasions (total number of challenges not available). In 2014, unconstitutionality was found in 23 out of 48 cases (48 per cent). In many cases, the unsuitable in the most suitable key?’] at the 36th Estonian Lawyers Congress, 08.10.2020. On ‘criministrative law’, see the presentation and slides of Markus Kärner, ‘Euroopa Liidu sanktsioonide sobitamine Eesti riigisisesesse õigusesse: kümme aastat diskussiooni’ [‘Fitting the European Union sanctions in Estonian national law: Ten years of discussions’] www. uttv.ee/naita?id=30538#. See also Anneli Soo, Alexander Lott, Andreas Kangur, ‘Võimalused Euroopa Liidu halduskaristuste ülevõtmiseks Eestis [Possibilities for Adoption of European Union Administrative Punishments in Estonia]’, Juridica 4/2020, 242 f, with references to ‘criministrative law’ at 243. 105 Fink, Õiguspärase (n 92). 106 Ibid, as translated by Albi. 107 Pablo Martín Rodríquez ‘A Missing Piece of European Emergency Law: Legal Certainty and Individuals’ Expectations in the EU Response to the Crisis’, (2016) 12 European Constitutional Law Review 265. 108 Ibid, 282. 109 See eg an article by a former judge responding to the harsh criticism of the Portuguese Constitutional Court: Ana Maria Guerra Martins, ‘Constitutional Judge, Social Rights and Public Debt Crisis: The Portuguese Constitutional Case Law, (2015) 22 Maastricht Journal of European and Comparative Law 678; and the report on Portugal in this volume by the same author. 110 Berit Aaviksoo, ‘Kohtulik activism põhiseaduslikkuse järelevalve funktsioonina. Kui aktivistlik on Eesti põhiseaduskohus?’ [‘Judicial Activism as a Function of Constitutional Review. How Activist is the Estonian Constitutional Court?’], Juridica (2005) 295, 296.

200  Hent Kalmo and Anneli Albi annulment of a law was based on fundamental principles of law, such as legitimate expectations, equal treatment and proportionality.111 This standard of constitutional review is typical to posttotalitarian and post-authoritarian constitutional courts both in the Western and Central-Eastern parts of Europe, especially in cases of abstract constitutional review.112 There is additionally the preventive effect, in that further instances of unconstitutionality are unlikely to arise once the Constitutional Court has decided on a matter. In the EU Courts, a study by Takis Tridimas and Gabriel Gari of the period 2001–05 shows that approximately 6.4 per cent of challenges against measures of general application (directives and regulations) led to annulment in the General Court.113 Whilst the success rate of 16.1 per cent in the Court of Justice seems higher, it is likely that the success rate of challenges brought by the Member States and private parties to the Court of Justice broadly remained at a similar level as in the General Court. The higher figure reflects the fact that EU institutions could bring a challenge directly to the Court of Justice where they had a high rate of success of 75 per cent, compared to the 26.4 per cent success rate of actions initiated by the Member States (it is assumed here that these two figures reflect challenges to all types of measures rather than to directives and regulations only). The study by Tridimas and Gari further showed that the annulments in most cases had been based on competence and procedural grounds rather than fundamental rights grounds.114 Indeed, the Data Retention Directive was only the second ever directive that was annulled on fundamental rights grounds; it was annulled in the second judicial challenge, after constitutional issues were raised in the constitutional courts of several Member States. It also seems pertinent to recall scholarly debates with regard to the CJEU approach of ‘anything goes’ in the judicial review of the ESM Treaty.115 It would further seem in order to briefly flag a broader concern about the profound effects on the legal education of the so-called Bologna Process, not strictly an EU initiative but a broader policy of global governance networks which was implemented in Estonia without any debate. In the framework of the Bologna Process, university level legal education was changed from the standard four-year curriculum to the so-called 3+2 model (three years of undergraduate, two years of postgraduate studies). The now prevalent three-year legal education has sparked widespread concerns that younger generations of lawyers lack a basic understanding of the fundamentals of law, and of a more integral and systemic understanding of the Estonian legal system and its core values and features. In an official report by the Supreme Court, and as discussed at a special conference in 2021, formal calls were put forward for reversing the Bologna Process reform and reinstating the four-year legal education in Estonia.116 Furthermore, the Supreme Court report concluded that market forces and competition have profoundly damaged rather than strengthened higher education – at least in the field of law and legal scholarship – contrary to the declared (neoliberal) aims of the Bologna Process reforms. The sheer extent to which there has been a change from binding constitutional rules to pragmatism in Estonia was in fact a driving factor for the author of this part of the present 111 The statistical information and grounds are from Ernits et al, ‘Unexpected Challenges’ (n 27) 936. 112 See Albi, Comparative Study (n 76). 113 Takis Tridimas and Gabriel Gari, ‘Winners and losers in Luxembourg: A statistical analysis of judicial review before the European Court of Justice and the Court of First Instance (2001–2005)’, 35 European Law Review (2010) 131, 170 ff. 114 Ibid, 170 f. 115 Cf n 59 above and the accompanying text. 116 Eesti õigushariduse ja juristikutse probleemidest ning võimalikest lahendustest’ [’Problems and potential solutions to Estonian legal education and the profession of lawyers’], Report by the Supreme Court, prepared by Villu Kõve, Ivo Pilving and Ene Andersen, 2021, available at www.riigikohus.ee. See also M Ernits, ‘Saja-aastane monopol – kas too big to fail?’ [‘Hundred-year monopoly – Is it too big to fail’?], Sirp, 28.05.2021, www.sirp.ee, and the conference’ Eesti Õigushariduse Tulevik’ [‘The Future of Estonian Legal Education’], 11.06.2021.

Estonia  201 chapter, Anneli Albi, for writing a longer paper about erosion of constitutional rights in EU law. That paper ended with a call for a move towards ‘substantive co-operative constitutionalism’ in the European Union.117 In that paper, concerns were collated from scholars who have written about the ‘erosion’ (Dieter Grimm) or ‘twilight’ (Petra Döbner and Martin Loughlin) of constitutionalism, and ‘the end of constitutionalism as we know it’ (Ming-Sung Kuo).118 Similarly to the concern in this paper about ‘pragmatisation’ of the Constitution in Estonia, in Italy a process of ‘flexibilisation’ of constitutions has been observed: the prevailing theories in Italy, as summarised by Andrea Simoncini, are that ‘the present economic crisis is clearly impairing the soundness of the guarantees and limits provided by constitutional law’, with concerns being raised about whether the ‘extra ordinem’ phase will be irreversible.119 Simoncini brings forth ample examples – from the changed practices regarding the sources of law to the reduced constitutional autonomy of the regions – of how the euro crisis measures have accelerated a ‘decline of European constitutionalism’ and are leading to the ‘progressive flexibilisation’ of constitutions and the creation of institutions that do not fall within constitutional legitimacy. Constitutions, as Simoncini notes, are ‘destined to be obsolete’ in ‘the present age [… that is] no longer the age of constitutions’.120 Additionally, Albi’s above-mentioned article collated examples of concerns expressed by scholars and judges about constitutional rights under strain in other Member States. Concerns about the decline of constitutionalism have become notably more heightened in the writings of scholars across Europe in the context of the EU’s euro crisis governance. In these writings, concerns have been expressed about the prolonged and perhaps even irreversible suspension of constitutionalism, of the Rechtsstaat and democracy, about courts bending the rules, as well as about authoritarian tendencies in the EU’s euro crisis management. These concerns have also been collated in the above-mentioned article. Indeed, such concerns have subsequently been documented in a more systematic way in the framework of a separate research project in the framework of which in-depth national reports on twenty-eight EU Member States were published.121 Based on systematic, European-wide comparative research, this co-author, Anneli Albi, in fact, joined a small but growing number of scholars who find that what has been underway is a gradual process of transition from the paradigm of constitutional and public law to the paradigm of autonomous governance that prevails in EU law.122 The paradigm of autonomous EU governance is predicated on very 117 Albi, ‘Erosion’ (n 4). 118 For references to relevant literature, see Albi, ‘Erosion’ (n 4), Part 2, 304 ff. 119 Andrea Simoncini, ‘Back to “Flexible” Constitutions? The Impact of Financial Crisis and the Decline of the European Constitutionalism’, (2013) 5 (2) Italian Journal of Public Law, 157 f. 120 Ibid, 158 f, 186. 121 The systematic study has been carried out in the framework of the research project ‘The Role and Future of National Constitutions in European and Global Governance’, funded by European Research Council grant No 284316 (the views cannot be attributed to the European Research Council). Updates are available at https://research.kent.ac.uk/roc/. The resulting national reports have been published in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law – National Reports (The Hague, TMC Asser Press, 2019), available open access at www.springer.com. On the basis of the national reports, a comparative monograph is forthcoming (Albi, Comparative Study (n 76)). 122 Some aspects of such a paradigm change from constitutional and public law to governance have been identified by the following scholars: Alexander Somek (in EU governance, constraints on public power have become economic (Somek, Cosmopolitan (n 7) 23 f); Agustín Menéndez (conceptual innovation in legal research, which includes a switch to ‘governance’ amongst other concepts) (Augustín Menéndez, ‘Editorial: A European Union in Constitutional Mutation?’, (2014) 20 European Law Journal 127, 140); Joana Mendes (with regard to transnational technical regulatory regimes with norms that acquire the status of EU legal acts and also acquire supremacy through incorporation into EU law without traditional avenues for participation (Joana Mendes, ‘Rule of Law and Participation: A Normative Analysis of Internationalized Rule-making as Composite Procedures’, (2014) 12 International Journal of Constitutional Law 370, 371)). Anneli Albi agreed with the findings about such an ongoing paradigm change, in Albi/Bardutzky, ‘Revisiting’ (n 74) 24 f; the broader implications of the paradigm change are developed further in Albi Comparative Study (n 76).

202  Hent Kalmo and Anneli Albi different foundational ideas for the exercise of public power, which include the neofunctionalist ideational setting that underlies much of the European integration and discourses, as well as other new foundational ideas for the exercise of public power, including neoliberalism.123 This research corroborates Somek’s finding that the EU’s euro crisis management represents a change from a constitutional law paradigm to a paradigm of governance, where formal legal constraints on public power have been replaced by economic and executive necessities; we have entered into ‘[t]he brave new world of exigencies’.124 Notably, the problems regarding fundamental rights, rule of law safeguards and decline of constitutionalism only arise if the normative point of reference lies in the national constitutions, especially of the post-totalitarian or post-authoritarian type; these problems do not arise if the normative point of reference shifts to EU Treaties or to the EU Charter of Fundamental Rights.

C.  A Note About the Recent Rule of Law Crisis in Central and Eastern Europe and Developments in Estonia Before concluding this section, a side remark seems in order given the widening systemic rule of law crisis in Central and Eastern Europe after Hungary instituted an illiberal regime in 2010 and Poland followed suit in 2015. A growing proportion of voters in Estonia have also turned to the far-right EKRE Party (Estonian Conservative People’s Party). This party, which was part of coalition governments in recent years and has undermined Estonia’s internationalist credentials, sought to introduce measures which raise rule of law issues of an altogether different magnitude. It is also alarming that through the rise of the EKRE Party, the protection of the Constitution came to often be used as a genuinely nationalistic tool of opposition to EU law measures. This co-author, Anneli Albi, would like to express her concern that whilst the voters’ turn to extremist parties tends to be attributed to nationalism, populism and problems within the individual Member States, some of the potential causes may also lie in the impossibility, in practice, to contest and change problematic EU measures, especially the very large financial liabilities under the ESM Treaty while the people have suffered real hardship. Some of the other more general concerns that have emerged amongst different groups of society about a loss of democracy due to EU requirements, but where there have been no realistic ways to address the issues, include the following. One is that, as Estonians are committed to organic agriculture, the way in which genetically modified foods have been approved in the EU has sparked disquiet. 123 These and other foundational ideas are identified and explored in greater detail in Albi, Comparative Study (n 76). That the new paradigm of governance is predicated on neofunctionalism, see Albi/Bardutzky, ‘Revisiting’ (n 74) 24 f, acknowledging in fn 6 at p 24 the draft chapters of Maris Moks’ PhD thesis ‘Guardianship of the Constitution versus the Expectations of the European Integration: Judicial Review of the Euro-crisis Management’ (Hertie School of Governance in Berlin). This work by Moks greatly helped Anneli Albi to formulate the broader observation about a change from the constitutional law mindset to a neofunctionalist mindset, and especially as regards the changing role of courts towards agents of integration through the neofunctionalist theory. That the EU’s most established and powerful meta-narrative and paradigm of ‘integration-through-law’ represents neofunctionalism, see Vauchez, ‘Integration’ (n 8). Generally on neofunctionalism and EU law, see Graínne De Búrca, ‘Rethinking Law in Neofunctionalist Theory’, (2005) 12 Journal of European Public Policy 310; and Alec Stone Sweet, Neofunctionalism and Supranational Governance (unabridged version), Faculty Scholarship Series, Paper 4628 (2012), http://digitalcommons.law.yale.edu. Regarding the extensive literature on neoliberalism in EU governance, see eg some of the publications referred to in different parts of this paper, eg Nicol, Capitalism (n 3); Manners/Whitman, ‘Another Theory’ (n 6) fn 137 and the accompanying text; Menéndez (n 131). More generally and critically on the neoliberal policies imposed by the IMF, the World Bank and other international financial institutions, see Joseph Stiglitz, Globalization and its Discontents (London, Penguin, 2002). 124 Somek, Cosmopolitan (n 7) 22 ff.

Estonia  203 On one occasion, in 2017, intense public debate in the country started only one day before a vote in Brussels, following a press release by the Green Movement. Another concern is that small organic farmers are going out of business. There were farmers’ protests in front of the Riigikogu in 2016. In a strongly worded press release (‘We can’t bear it any more’), the farmers expressed concern that the EU system of subsidies and sanctions favours large corporations, which means that small farmers cannot sell naturally, organically grown produce at a price level that would cover costs, and large numbers of small farms are facing the decision to close. They added that small farmers in Latvia and Lithuania are facing similar problems.125 One could link this issue to the disappearance of once vibrant rural communities126 as well as to a subsequent increase of emigration to Western European countries, which is often embarked upon unwillingly, driven by sheer economic necessity; this then in turn exacerbates public disquiet about the immigration levels in those countries. A further problematic EU rule that has been repeatedly raised, most recently by the newly established Biodiversity Party, is the unreasonableness of the fact that Estonian farmers have to compete with strawberries and cucumbers imported far away from Spain, also given the effects of ‘food miles’ and transport on climate change. There is also an acute and widely shared concern that the transition from the Estonian kroon to the euro has sharply increased the cost of living. There are genuine difficulties in raising and addressing policy issues at the EU level. For most Estonians, it would be difficult to travel to Brussels or Strasbourg to talk to decision-makers or to organise protests or form multi-country coalitions there, also given the cost, distance and language barriers. At the same time, the domestic authorities refer to the requirement of loyal implementation. The paradox is well captured by Gareth Davies. Davies observes that the national authorities have been infantilised, as their only choice is to correctly implement EU law in a wide range of areas. There is no room for discussion on ‘who are we and what do we want?’ or to have a ‘meaningful debate about society’.127 There would seem to be a case to consider such wider structural issues in the EU system when exploring the reasons for the recent rule of law crisis in Central and Eastern Europe, also with a view to preventing backsliding in further countries. Additionally, what has been very understudied is the human suffering and the long-term effects of the so-called ‘shock therapy’ administered by the IMF and the World Bank policies to Central and Eastern Europe in the years following the 1989 transformations. The neoliberal policies of privatisation of public services, financial and credit liberalisation, universal commodification (leading inter alia to high cost of housing, water etc), the conversion of private debt into public debt in financial crises and subsequent harsh austerity measures, as well as the general dismantling of the role of the state and especially of social protection, have been seen as a cause of the rise in poverty, inequality and non-emergence of a strong middle class and healthy civil society in Central and Eastern Europe.128 Yet these policies have subsequently been repeated and constitutionally entrenched as part of the European 125 ‘Sibiscare põllumajanduse meeleavaldus: Meil on kõrini’; pressiteade Eesti Rohelise Liikumise infolistis [’Sibiscare Agricultural Protest: We can’t bear it any more’; Press release in the Estonian Green Movement information list], 05.02.2016, on file with A Albi. 126 Liis Serk, ‘Optimismi õppetunnid ehk tühjad Eesti külad ja üksindus maal’ [‘Lessons of Optimism or empty Estonian villages and loneliness in the countryside’], Sirp, 20.05.2016, https://sirp.ee/s1-artiklid/arhitektuur/optimismioppetunnid-ehk/. 127 Gareth Davies, ‘Social Legitimacy and Purposive Power: The End, the Means and the Consent of the People’, in D Kochenov, G De Búrca and A Williams (eds), Europe’s Justice Deficit (Oxford, Hart, 2015) 272, 274. 128 See, in particular, the work of the Nobel Prize laureate economist Joseph Stiglitz, Globalization (n 123) 167, 141, 162, and 133 ff. For a summary of Stiglitz’ work in Estonian, see book review by Anneli Albi, ‘Stiglitz: Vaja on uut majandusmudelit ja finantssüsteemi’ [’Stiglitz: A New Economic Model and a Financial System Are Needed’], Sirp, 20.10.2011, www.sirp.ee/s1-artiklid/c9-sotsiaalia/stiglitz-vaja-on-uut-majandusmudelit-ja-finantssuesteemi/.

204  Hent Kalmo and Anneli Albi Commission and IMF management of the euro crisis. Furthermore, the direction of travel is towards further centralisation of power to the EU level, including through initiatives towards mutualisation of debt of the Member States and the creation of the EU’s own fiscal resources.129 In the view of this co-author, it is likely that the above direction of travel will cause further resentment amongst the people, with discontent likely to be expressed through voters’ turn to extreme parties. This co-author, Anneli Albi, would add that indeterminate debt which needs constant economic growth to service it, and further centralisation of powers to the EU level, backed up by EU enforcement proceedings, will also severely curtail prospects for restoring the European social, democratic constitutional Rechtsstaat as well as the possibilities for the Member States to switch to alternative, environmentally more sustainable and socially just economic ideologies.

VIII.  Concluding Remarks: A Case for a Joined-Up Discussion on the Impact of EU law on the Member States’ Constitutional Systems as Part of the Debate on the Future of Europe The chapter identified a broader shift from the binding rules of the post-totalitarian constitutional Rechtsstaat to pragmatism, which has occurred in the Estonian constitutional culture through adjustments to European integration. This broader shift was explored through three developments. First, Hent Kalmo explored the opening of the Constitution to the supremacy of EU law by virtue of the Supplementary Constitutional Act. It was seen that the judgments of the Supreme Court have given EU law by and large exceptionless primacy, and that the formerly strict, binding nature of the constitutional provisions has been replaced by a more flexible approach. This approach is closely related to the second broader shift explored by Kalmo – the near-abandonment of constitutional review on the part of the Supreme Court in relation to a very broad range of provisions ‘tied’ – even remotely – to EU law. Eventually, a hint that the fundamental principles of the Constitution may set limits to the applicability of EU law was offered by the ESM Treaty case in 2012, where the Supreme Court found that a potential new treaty may require legitimisation by a referendum. The ESM ruling also revealed deep divisions within the Court: nine dissenting justices voiced profound concerns about the constitutional impact of the very large financial liabilities entailed by the ESM Treaty. The ESM case was explored in the final section of this chapter on substantive constitutional law, written by Anneli Albi. Albi additionally outlined the third broader shift: one from rules to pragmatism as regards the stringent post-totalitarian, German Rechtsstaat-tradition-influenced constitutional culture, where a high standard of constitutional and judicial protection had earlier been provided to fundamental constitutional rights and the rule of law safeguards. A range of examples were given of a shift to a pragmatic approach that is based more on the considerations of effectiveness, teleological interpretation, the prioritisation 129 At the time of concluding the main text of this chapter in 2019, President Macron of France, as well as the European Commission, the European Parliament and many scholars, have been advocating the mutualisation of debt and creation of EU’s own fiscal resources – including through borrowing from markets – for which Member States would ultimately remain jointly liable. Notably, the recently established New Hanseatic League, which includes Nordic and Baltic states (including Estonia), has expressed opposition to ‘far-reaching transfer of competences to the European level’. For an overview, see Federico Fabbrini, A Fiscal Capacity for the Eurozone: Constitutional Perspectives, In-depth analysis requested by the European Parliament AFCO Committee, Policy Department for Citizens’ Rights and Constitutional Affairs, PE 608.862, February 2019. For the positions of the New Hanseatic League, see pp 14 f.

Estonia  205 of the market, and the functionalist and neofunctionalist orientation to law represented by the EU legal order. If the reader agrees with concerns about the scale of the pragmatisation – or, indeed, even displacement – of the core structural and substantive components of the Estonian constitutional order, this prompts the question whether the shift to pragmatism ought to become the new normal or, as in our view, ought to remain a temporary, short-term solution, with an eventual return to the binding rules of the constitutional Rechtsstaat. In the latter case, it should be borne in mind that Estonia or its legal community are not in a position to change the direction of travel, not only due to the constraints of the EU legal order but also due to geopolitical concerns. In Estonia as well as the other Baltic states, membership in the EU is part of a larger security strategy, and anything less than full compliance with its rules, albeit on solid constitutional grounds, are seen as playing with fire. In other words, a threat to sovereignty is taken to lie not in EU law but in a refusal to apply it (and the consequent diminished status of the country as a member). As one Estonian compatriot informally and anonymously summed up the situation, ‘Constitutional safeguards are a peacetime topic. When survival is at stake, these become a luxury.’ However, it is an alienating way to live, especially on a prolonged basis. Furthermore, the recent departure of the UK from the EU and the voters’ turn to extremist parties all over Europe are also weakening the broader geopolitical stability. It was briefly pointed out by Anneli Albi in this chapter that many of the constitutional issues that have been posed by EU law in Estonia have also arisen in other Member States, and that there are broader concerns amongst scholars in Europe about ‘twilight’ or obsolescence of national constitutions and a change to a paradigm of governance that is predicated on entirely different foundational ideas for the exercise of public power. Such concerns have never really been discussed in a joined-up manner. We would like to end this chapter by putting forward a suggestion that questions about the impact of EU law on the Member States’ constitutional systems ought to be formally included on the agenda of the ongoing debate on the future of Europe. Indeed, a growing number of scholars have in one way or another called for retaining the constitutional orders of the Member States. By way of some examples, Agustín Menéndez, who has extensively outlined concerns about the EU’s ‘constitutional mutation’ through euro crisis management, finds that it may be a suitable time to ‘reconstruct European constitutional law with the help of “classical” democratic constitutional theory, as developed for decades in national Social and Democratic Rechtsstaats’.130 Menéndez regards ‘the collective of national democratic constitutions’ in Europe as ‘[t]he deep constitution of the European Union, the ultimate normative foundation of the whole edifice of the Union’, rather than the EU Treaties. He finds that ‘[w]hen integration starts going against the key normative content of the national constitutions, it is time to start using such constitutions … as the ultimate source of the yardstick of European constitutionality.’131 Matej Avbelj has suggested that the dominant constitutional narrative, which equates the constitutionalisation of the EU with US-style federalism, has been transplanted in a misconceived way that is unsuited to the European constitutional landscape, and that ‘work must begin’ on the EU’s ‘own, genuine and authentic constitutional theory’ that would be oriented towards a pluralist legal entity with 28 autonomous legal orders.132 One of the co-authors of the present chapter, Anneli Albi, has in an above-mentioned article outlined the procedural, formal, thin nature of EU constitutionalism 130 Menéndez, ‘Editorial’ (n 122) 140. 131 Augustín Menéndez, ‘The Existential Crisis of the European Union’ (2013) 14 German Law Journal 453, 525 f. 132 Matej Avbelj, The Pitfalls of (Comparative) Constitutionalism for European Integration. Eric Stein Working Paper 1/2008, 4 f, 15 f, 23, 24.

206  Hent Kalmo and Anneli Albi where the keywords are supremacy, uniformity, direct effect, autonomy, effectiveness and trust. She has propounded the concept of ‘substantive co-operative constitutionalism’, in which the aim would be to uphold the established standard of protection of fundamental rights, the rule of law and other constitutional values, and also to retain the diversity of national constitutional orders instead of transition to autonomous, self-referential, uniform set of EU norms in all areas of law.133 In order that a joined-up discussion on problematic effects of EU law on the Member States’ constitutional systems could take place in practice, there is a need to bring awareness to, acknowledge and openly address some structural issues in the mainstream EU discourse on national constitutions. This is also important given that the EU has to a significant extent been a ‘juristic’ project – written by a ‘community’ of lawyers, clerks, scholars and judges, with legal expertise effectively being a precondition for participating in the policy debates.134 The structural issues in the EU discourse include, as mentioned in the opening section, the ‘disconnect’ between EU discourses and national constitutional discourses, which reinforces the problem that EU epistemic communities, in fact, often do not have a good understanding of national and comparative European constitutional law, and which may well also be one of the reasons why constitutional issues arising at the national level tend to be simplistically blurred into the notions of sovereignty and national constitutional identity. Another structural issue is the assumption that the path to human progress lies in a shift to autonomous, self-referential EU law and legal consciousness that is decoupled from national cultures, histories and national legal languages.135 Further structural issues include the need to find ways to rebalance the discourse, publication opportunities and research funding in the reality where the European Commission has extensively invested in specialised EU law journals and university centres, with the aim of shifting EU studies away from a comparative analysis of national laws (which would aim to identify common principles) to the study of the ‘specificity’ of EU law, focusing on the relationships between EU law (and the CJEU) and national law (and courts).136 In the field of EU political science, a study by Ian Manners and Richard Whitman on articles published in the last decade in the leading Journal of Common Market Studies finds that mainstream EU scholarship has operated on the premise that the EU is a neoliberal, state-like political system, where all 28 Member States’ systems have been dissolved as easily as the 27 administrative regions in French politics; in the mainstream EU studies journals dissonant voices have rarely been published, while publication in other journals leads to limited visibility and citations.137 Impressionistically, the picture seems very similar in the field of leading EU law journals; the articles on national constitutions that get published in leading, high visibility journals tend to be predominantly from the narrow, neofunctionalist perspectives of compliance or posing obstacles to integration.138 In such a reality, ideally, there seems to be a need for convening some sort of a European-wide forum of constitutional lawyers, with funding provided by all Member States, in order to identify

133 Albi, ‘Erosion’ (n 4), Part 2, Sect VIII ff. 134 Francis Snyder, ‘New Directions in European Community Law’, (1987) 14 Journal of Law and Society 167, cited in Jo Shaw, ‘European Union Legal Studies in Crisis? Towards a New Dynamic’, (1996) 16 Oxford Journal of Legal Studies 231, 234; Harm Schepel and Rein Wesseling, ‘The Legal Community: Judges, Lawyers, Officials and Clerks in the Writing of Europe’, (1997) 3 European Law Journal 165; Vauchez, ‘Transnational politics’ (n 7) 4 and 26 ff. 135 See for scholarly work where concern has been expressed about this direction of travel nn 7–8 and the accompanying text. See also note 86. 136 Vauchez, ‘Transnational politics’ (n 7) 20 f. 137 Manners/Whitman, ‘Another Theory’ (n 6) 3, 7, with reference to the work of A Kreppel regarding comparison with the French system. 138 See Albi, ‘Erosion’ (n 4), Part 2, 308–11.

Estonia  207 in a joined-up manner overlooked and undiscussed problems that have been posed by EU law to the constitutional systems of the Member States. Systematic collation and publication of such concerns would then, in turn, enable work on practical sets of proposals that would ground and attune the European Union better to comparative European constitutional law and landscape, which needs to be part of the debate on the future of Europe.

References B Aasa, ‘The Principle of Mutual Trust in EU Law: What is in a Name’? PhD Thesis defended at the European University Institute, Florence, on 26 February 2021. B Aaviksoo, ‘Kohtulik activism põhiseaduslikkuse järelevalve funktsioonina. Kui aktivistlik on Eesti põhiseaduskohus?’ [‘Judicial Activism as a Function of Constitutional Review. How Activist is the Estonian Constitutional Court?’], (2005) Juridica 295. A Albi, ‘From the Banana saga to a Sugar Saga and Beyond: Could the Post-communist Constitutional Courts Teach the EU a Lesson in the Rule of Law?’, (2010) 47 CML Rev 791. A Albi, ‘Erosion of constitutional rights in EU law: A call for “substantive co-operative constitutionalism”’, (2015) 9 Vienna Journal of International Constitutional Law No 2, 151 (Part 1) and No 3, 291 (Part 2), http://ssrn.com/author=1246144. A Albi, A Comparative Study on National Constitutions in EU and Global Governance (The Hague, TMC Asser Press and Springer, forthcoming 2022). A Albi and S Bardutzky, ‘Revisiting the Role and Future of National Constitutions in European and Global Governance: Introduction to the Research Project’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law – National Reports (The Hague, TMC Asser Press, 2019, open access at www.springer.com) 3. R Alexy, ‘Põhiõigused Eesti põhiseaduses’ [‘Fundamental Rights in the Estonian Constitution’], (2001) Juridica Special Edition, 5. R Alexy, A Theory of Constitutional Rights (J Rivers transl) (Oxford, OUP, 2002). J Antonova, M Pardla and I Markus, ‘Estonia’, in J Laffranque (ed), Reports of the XXV FIDE Congress Tallinn 2012, Vol 3: The Area of Freedom, Security and Justice, Including Information Society Issues (Tartu, Tartu University Press, 2012) 324. M Avbelj, The Pitfalls of (Comparative) Constitutionalism for European Integration, Eric Stein Working Paper 1/2008, www.ericsteinpapers.cz/. G Beck, The Legal Reasoning of the Court of Justice of the EU (Oxford, Hart, 2012). M Bobek, ‘On the Application of European Law in (Not Only) the Courts of the New Member States: “Don’t Do as I Say”?’, (2007) 10 Cambridge Yearbook of European Legal Studies 1. M Bobek, ‘Case C-345/06, Gottfried Heinrich, Judgment of the Court of Justice (Grand Chamber) of 10 March 2009’, (2009) 46 CML Rev 2086. M Bobek, ‘The Legal Reasoning of the Court of Justice of the EU’, (2014) 39 European Law Review 423. G Davies, ‘Social Legitimacy and Purposive Power: The End, the Means and the Consent of the People’, in D Kochenov, G De Búrca and A Williams (eds), Europe’s Justice Deficit (Oxford, Hart, 2015) 272. G De Búrca ‘Rethinking Law in Neofunctionalist Theory’, (2005) 12 Journal of European Public Policy 310. Eesti Vabariigi Põhiseadus. Kommenteeritud Väljaanne [The Constitution of the Republic of Estonia. Commented Edition], 3rd edn (Tallinn, Juura, 2012). M Ernits, ‘Constitutional Review in the Age of Balancing’, in the 6th International Scientific Conference ‘Konstitucionālās vērtības mūsdienu tiesiskajā telpā I’ [‘Constitutional Values in the Modern Legal Space I’], Riga 10–11 November 2014 (Riga, University of Latvia Press, 2016) 127. M Ernits, C Ginter, S Laos, M Allikmets, P K Tupay, R Värk and A Laurand, ‘The Constitution of Estonia: The Unexpected Challenges of Unlimited Primacy of EU Law’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law – National Reports (The Hague, TMC Asser Press, 2019) 887.

208  Hent Kalmo and Anneli Albi F Fabbrini, A Fiscal Capacity for the Eurozone: Constitutional Perspectives, In-depth analysis requested by the European Parliament AFCO Committee, Policy Department for Citizens’ Rights and Constitutional Affairs, PE 608.862, February 2019. E Fink, Õiguspärase ootuse kaitse põhimõtte eeldused ja piirid Euroopa Liidu õiguses [‘The Conditions and Limits of the Protection of the Principle of Legitimate Expectations in EU Law’] (Tartu, Tartu University Press, 2016). C Ginter, ‘Constitutionality of the European Stability Mechanism in Estonia: Applying Proportionality to Sovereignty’, (2013) 9 European Constitutional Law Review 335. S Glanert, ‘Speaking Language to Law: The Case of Europe’, (2008) 28 Legal Studies 163. AM Guerra Martins, ‘Constitutional Judge, Social Rights and Public Debt Crisis: The Portuguese Constitutional Case Law’, (2015) 22 Maastricht Journal of European and Comparative Law 678. U Jaremba, National Judges as EU Law Judges: The Polish Civil Law System (Leiden, Brill, 2014). H Kalmo, ‘Põhiseadus ja proportsionaalsus – kas pilvitu kooselu?’ [‘The Constitution and Proportionality – A Cloudless Partnership?’], (2013) Juridica 79. AL Kjaer, ‘Theoretical Aspects of Legal Translation in the EU: The Paradoxical Relationship between Language, Translation and the Autonomy of EU Law’, in S Sarcevic (ed), Language and Culture in EU Law. Multidisciplinary Perspectives (Burlington, Ashgate, 2015) 91. J Laffranque, ‘Constitution of the Republic of Estonia in the Light of Accession to the European Union’, (2001) 6 Juridica International, 207. P Leino-Sandberg and J Salminen, ‘Languages and EU law discourse: A view from a bilingual periphery’, Verfassungsblog 02.04.2014, https://verfassungsblog.de. L Mälksoo, ‘Eesti suveräänsus 1988–2008’ [‘Estonian Sovereignty 1988–2008’], in H Kalmo and M Luts-Sootak (eds), Iganenud või igavene? Tekste kaasaegsest suveräänsusest [‘Old-fashioned or Permanent? Texts on Contemporary Sovereignty’] (Tartu, Tartu Ülikooli kirjastus, 2010) 147. I Manners and R Whitman, ‘Another Theory is Possible: Dissident Voices in Theorising Europe’, (2016) 54 Journal of Common Market Studies 3. J Mendes, ‘Rule of Law and Participation: A Normative Analysis of Internationalized Rule-making as Composite Procedures’, (2014) 12 International Journal of Constitutional Law 370. AJ Menéndez, ‘The Existential Crisis of the European Union’, (2013) 14 German Law Journal 453. AJ Menéndez, ‘Editorial: A European Union in Constitutional Mutation?’, (2014) 20 European Law Journal 127. D Nicol, The Constitutional Protection of Capitalism (Oxford, Hart, 2010). C Pinelli, ‘The Formation of a Constitutional Tradition in Continental Europe since World War II’, (2016) 22 European Public Law 257. PM Rodríquez ‘A Missing Piece of European Emergency Law: Legal Certainty and Individuals’ Expectations in the EU Response to the Crisis’, (2016) 12 European Constitutional Law Review 265. M Rosentau, ‘Suveräänsus Euroopa Liidus’ [‘Sovereignty in the European Union’], Riigikogu Toimetised 2003/8, 11. H Schepel and R Wesseling, ‘The Legal Community: Judges, Lawyers, Officials and Clerks in the Writing of Europe’, (1997) 3 European Law Journal 165. J Shaw, ‘European Union Legal Studies in Crisis? Towards a New Dynamic’, (1996) 16 Oxford Journal of Legal Studies 231. A Simoncini, ‘Back to “Flexible” Constitutions? The Impact of Financial Crisis and the Decline of the European Constitutionalism’, (2013) 5 (2) Italian Journal of Public Law 157. A Somek, The Cosmopolitan Constitution (Oxford, OUP, 2014). A Soo, A Lott, A Kangur, ‘Võimalused Euroopa Liidu halduskaristuste ülevõtmiseks Eestis [Possibilities for Adoption of European Union Administrative Punishments in Estonia]’, Juridica 4/2020, 242. J Stiglitz, Globalization and its Discontents (London, Penguin, 2002) A Stone Sweet, Neofunctionalism and Supranational Governance (unabridged version), Faculty Scholarship Series, Paper 4628 (2012). D Thym, ‘The Solitude of European Law Made in Germany’, Verfassungsblog 29.05.2014, https:// verfassungsblog.de.

Estonia  209 T Tridimas and G Gari, ‘Winners and losers in Luxembourg: A statistical analysis of judicial review before the European Court of Justice and the Court of First Instance (2001–2005)’, (2010) 35 European Law Review 131. PK Tupay, Verfassung und Verfassungsänderung in Estland. Eine Analyse zu Theorie und Praxis mit vergleichenden Anmerkungen zum deutschen Recht (Berlin, BWV Berliner Wissenschafts-Verlag, 2015). A Vauchez, ‘The transnational politics of judicialization. Van Gend en Loos and the making of the EU polity’, (2009) 16 European Law Journal 4. A Vauchez, ‘“Integration-through-Law” Contribution to a Socio-history of EU Political Commonsense’, EUI Working Paper RSCAS 2008/10, http://cadmus.eui.eu. EW Walker, Dissolution. Sovereignty and the Breakup of the Soviet Union (Lanham MD, Rowman & Littlefield, 2003).

210

7 Ireland GAVIN BARRETT

I.  Main Characteristics of the National Constitutional System A. Introduction Ireland is a unitary state and a constitutional democracy with a parliamentary system of ­government. It gained its current constitutional status gradually, attaining effective independence as the Irish Free State in 1922 under the Anglo-Irish Treaty.1 (This followed the proclamation of an (unrecognised) Republic in the unsuccessful 1916 rebellion, the execution of its leaders, abstention from Westminster by the overwhelming majority of Irish MPs elected in 1919 and a period of violent guerrilla warfare.) The new state was a (reluctant) constitutional monarchy, and a dominion within the British Commonwealth. It was compelled to retain strong links with the UK (eg, granting British forces use of so-called ‘Treaty ports’, having its parliamentarians swearing an oath of fidelity to the British Crown, allowing appeals from its courts to the Privy Council, and having a Governor-General). Technically, on the date of its establishment,2 the Free State consisted of the entire island of Ireland, but Northern Ireland (comprising the six north-eastern counties of the island) immediately opted out of the newly founded state, and has remained to this day a constituent part of the UK. The new Free State had its own Constitution, enacted by the Irish revolutionary parliament, the Dáil (sitting as a constituent assembly) in October 1922.3 In accordance with its own Article 83, the 1922 Constitution came into operation on the issue of a Proclamation by King George V on 6 December 1922. It endured a mere 15 years, but many of its principal aspects were re-adopted in its 1937 successor, the present lrish Constitution. Controversy over the 1922 compromise led to an 11-month-long civil war, causing more loss of life and property than the earlier war of independence and laying the foundations for the most enduring political rivalry in Irish political life – that between the Fianna Fáil and Fine Gael parties – which would dominate Irish politics for the next ninety years. The pro-Treaty faction gained the upper hand both electorally and militarily, thereby guaranteeing the survival of the new state. However, their republican opponents, led by Eamon de Valera, then converted to parliamentary methods, winning a series of elections from 1932 onwards. Fianna Fáil-led 1 The so-called ‘Articles of Agreement for a Treaty between Great Britain and Ireland’. 2 06.12.1922. 3 As the Constitution of the Irish Free State (Saorstát Eireann) Act 1922. It was given effect in UK law by the Irish Free State Constitution Act 1922.

212  Gavin Barrett governments severed link after link with the British Crown (abolishing the oath,4 the post of the Governor-General5 and appeals to the Privy Council,6 and negotiating the return of the Treaty ports, thereby rendering possible Irish neutrality in World War II).7 By 1936, the sole practical function left to the Crown was accrediting Irish diplomatic representatives.8 In 1937, a new Constitution was adopted, dropping all references to the Crown, Empire, UK and the 1921 Treaty. It has remained in force ever since. Its Article 4 declared the name of the State to be ‘Éire, or in the English language, Ireland’. Article 5 asserted ‘Ireland is a sovereign, independent, democratic state’. After parliamentary approval, the new Constitution was approved by referendum on 1 July 1937. Ireland remained in the Commonwealth yet stayed neutral in World War II, then became a republic under the Republic of Ireland Act 1948, leaving the Commonwealth. Section 2 of the 1948 Act declared ‘that the description of the State shall be the Republic of Ireland’.9 Section 3 provided that ‘the President, on the authority and on the advice of the Government, may exercise the executive power or any executive function of the State in or in connection with its external relations’, in effect upgrading the President to a head of state. As Kelly has summarised the royal and British theme descended between 1922 and 1948 in a chromatic scale until it ran off the constitutional keyboard; and the full republican chord, which might have been struck together in 1916 if the Easter Rising had succeeded, sounded (for twenty-six counties) in an arpeggio gradually formed over a quarter of a century.10

Constitutional ambiguity concerning the extent of the national territory was more enduring. Enforced partition in 1922 in order to accommodate the wishes of the Unionist majority in Northern Ireland caused deep resentment. Article 2 of the 1937 Constitution asserted that ‘the national territory consists of the whole island of Ireland, its islands and the territorial seas’.11 In May 1998, consequent on the Good Friday Agreement formalising the ending of nearly 70 years of political violence in Northern Ireland, this territorial claim to Northern Ireland was abandoned and replaced by an assertion merely of the entitlement of every person born in the island of Ireland to be ‘part of the Irish Nation’ and recognising that ‘a united Ireland shall be brought about only by peaceful means with the consent of a majority of the people, democratically expressed’. Ireland has a bicameral parliament known as the Oireachtas.12 The lower House (Dáil Éireann) has 166 members, directly elected at least every five years. The voting system is proportional representation by single transferable vote. The upper House or Senate (Seanad Éireann) has 60 members. Forty-three of these are elected by fewer than 1,000 elected local authority councillors, from broad panels of candidates supposedly representing culture, agriculture and fisheries, organised labour, industry and commerce, and public administration. Six are elected by the graduates of some Irish universities. Another  11  are nominated by the Taoiseach (prime minister). This combination of a 4 Under the Constitution (Removal of Oath) Act, 1933. 5 Under the Constitution (Amendment No 27) Act, 1936. 6 Under the Constitution (Amendment No 22) Act, 1933. The legality of this step was upheld by the Judicial Committee of the Privy Council itself in Moore v Attorney-General of the Irish Free State [1935] AC 484, by reference to the 1931 Statute of Westminster. 7 See the British-Irish Tripartite Agreement on Trade, Finance and Defence, agreed at London on 25.04.1938. 8 Under the Executive Authority (External Relations) Act, 1936, repealed by the Republic of Ireland Act 1948. 9 This did not however alter Art 4 of the Constitution, according to which the name of the State is Éire, or in the English language Ireland. Such a change would have required amending legislation approved by referendum. 10 Preface to the first edition of John M Kelly, The Irish Constitution, reproduced in the G Hogan and G Whyte, JM Kelly: The Irish Constitution, 4th edn (Haywards Heath, Tottel, 2006). 11 Although, under Art 3, ‘pending the re-integration of the national territory’, laws enacted by the Oireachtas were to have the same application as those of the Free State. 12 Art 15 of the Constitution.

Ireland  213 (legislatively) narrowly defined Seanad electorate dominated by political parties, and the large number of Seanad members nominated by the Taoiseach, gives the Dáil the upper hand in Parliament. The Dáil itself, however, is dominated by the executive via exercised party political control. Government plans to abolish the Seanad failed in October 2013, when a referendum proposal was rejected. A constituent element of the Oireachtas is Ireland’s directly elected President. Lacking any executive/policy role, he exercises his or her functions on the advice of the Government. There are a small number of instances where the President may exercise his or her own discretion, the most prominent being the power to refer a Bill to the Supreme Court for a ruling on its constitutionality under Article 26, and the discretion to refuse to dissolve the Dáil on the advice of a Taoiseach who has ceased to retain a Dáil majority.13

B.  Judicial System Ireland retained the common law legal system introduced by its former British rulers, with a court system employing an adversarial approach. The Court system is provided for in Article 34 of the Constitution, largely mirroring that set up under the 1922 Constitution. There is a unified court structure, no separate system of administrative courts and no separate Constitutional Court. The Superior Courts consist of a High Court, with full original jurisdiction, a Court of Appeal (the creation of which was authorised in a referendum held in October 201314) and a Supreme Court. Constitutional review of laws is permitted to the Superior Courts.15 There is no procedure analogous to the German Verfassungsbeschwerde, but a law’s constitutionality can be challenged in normal review proceedings brought by a litigant who can show locus standi (with a particularly liberal approach having been adopted by the Courts to standing in challenges concerning European treaties).16 The President can also refer the question of a Bill’s constitutionality to the Supreme Court under Article  26. Courts of local and limited jurisdiction are also provided for by Article 34, including the non-jury District Court and the (higher-ranking) Circuit Court. Both try both criminal and civil cases, are geographically limited in their jurisdiction and have an upper financial limit on the claims they may hear. There are also some specialist Tribunals such as the Labour Court and the Equality Tribunal.

II.  Constitutional Culture A. Introduction A Constitution providing for judicial review is perhaps the most prominent feature of the Irish legal system distinguishing it from that of the United Kingdom. The Constitution is a ‘strong’ one in that it both makes explicit provision for fundamental rights and, since the 1960s, has been used by the Irish courts as the foundation of a rich vein of implied or ‘unenumerated’ constitutional rights.17 Extensive case law exists regarding both, and 13 Art 13.2.2. Additional functions can, however, be conferred on the President by law. 14 Prior to this, there was already a statute-based Court of Criminal Appeal. 15 See Art 34.3.2° of the Constitution. 16 See Irish Supreme Court (hereinafter IESC) 09.04.1987, Case 1986 No 12036P, Crotty v An Taoiseach [1987] 713 at 766. 17 Beginning with IESC 03.07.1964, Case 1962 No 913 P, Ryan v Attorney General [1965] 2 IR 294.

214  Gavin Barrett executive actions and statutes alike may be declared invalid for having infringed the Constitution in litigation brought by private parties. The Constitution is relatively rigid in that since a short transition period ended, it can be amended only by referendum. Including the 1937 referendum on the Constitution itself, Ireland has had 42 constitutional referendums, 31 of which have succeeded. However, such successful votes have not been evenly spread out over the 84 years of the Constitution’s life. Almost 35 years elapsed after the adoption of the Constitution in 1937 and the first successful referendum in 1972. The Constitution has thus seen two periods of several decades, the first involving no successful amending referendums, the second seeing a very large number. Seven of the 30 successful referendums since 1972 have been linked to Irish membership of the now-EU – the initial 1972 membership referendum, and referendums on the SEA, the Treaty of Maastricht, the Treaty of Amsterdam, the Nice Treaty, the Lisbon Treaty and most recently the Fiscal Compact (not an EU Treaty, but closely linked to the EU and the eurozone). Eleven proposed constitutional amendments have failed to win acceptance in referendum since 1937. Two of these – the 2001 and 2008 proposals facilitating respectively the ratification of the Nice and Lisbon Treaties – were EU-related. Irish ratification of EU Treaties could thenceforth no longer automatically be assumed, a point of considerable interest, given that amendments to EU Treaties must be ratified by all Member States.18 Both referendum rejections were subsequently reversed. The Nice Treaty amendment was passed on the second occasion when the Government increased Oireachtas involvement in European affairs, and provided constitutional protection of Irish neutrality in the EU context. The Lisbon Treaty amendment succeeded in a second referendum after Ireland was given (i) Danish-style reassurances on the meaning of certain Lisbon Treaty provisions, and (ii) a political commitment from the European Council not to reduce the size of the Commission.

B.  The Future A Constitutional Convention was established by resolutions of both Houses after the 2011 general election. Instituted in response to a widespread feeling that existing constitutional structures had contributed to Ireland’s economic crisis of 2008 onwards, the Convention was nonetheless conservatively constituted both in its composition (including as it did a high proportion of politicians) and in the narrow scope of the issues it could consider. It had 100 members, a Chairman, 29 Oireachtas members, four Northern Ireland Assembly members and 66 randomly selected Irish citizens. The Convention made several recommendations requiring constitutional referendums, including reducing the minimum age for candidacy for the presidency from 35 to 21, reducing the voting age from 18 to 16, amendments regarding Article 41.2.1 (concerning the role of women in the home), legalising same-sex marriage, giving non-resident citizens the right to vote in presidential elections, removing the Article 40.6.1˚ duty to criminalise blasphemy and an increased constitutional role for economic, social and cultural rights.19 The resolution establishing the Convention had avoided obliging the Government to hold referendums on foot of its recommendations, however, and the 2011–16 Government held only two (both on the same day in May 201520): one to permit same-sex marriage,21 which succeeded; the other to reduce 18 See Art 48 TEU. 19 For a more extensive list, see Harry McGee, ‘Only Two Proposals for Constitution Referendum’, Irish Times 26.01.2015. 20 Ibid. 21 Thirty-fourth Amendment of the Constitution (Marriage Equality) Act 2015, approved in referendum 22.05.2015 and signed into law on 29.08.2015.

Ireland  215 the age of eligibility for candidacy in presidential elections, which failed.22 All of the Convention proposals bar that concerning economic, social and cultural rights returned to the agenda after the election of a new Government in 2016. The ‘Programme for a Partnership Government’ which followed the February 2016 general election contained a commitment to establish a Citizens’ Assembly (this time without politicians participating). The creation of a 100-person Assembly was subsequently approved by resolution of both Houses with a mandate to consider several issues, including, inter alia, the Eighth Amendment of the Constitution (concerning abortion), and how referendums are held. The Assembly published a final report in June 2017 recommending replacement of the Eighth Amendment with a provision allowing the Oireachtas to decide how to legislate on terminations. A joint Oireachtas committee then similarly recommended removal of the Eighth Amendment. The referendum, held on 25 May 2018, passed by a majority of 66.4 per cent. As regards how referendums are conducted, the Assembly published its final report in June 2018, recommending inter alia (a) the establishment of a permanent electoral commission; (b) retention of the current prohibition on the Government spending public money to advocate on one side of a referendum campaign; (c) effective spending limits in referendum campaigns for political parties, campaign groups and individuals; and (d) the introduction of citizens’ initiatives putting both constitutional referendum proposals and legislative changes to the people, and also putting items on the agenda for decision by the Oireachtas.23 Adoption of such proposals regarding citizens’ initiatives would be a radical departure for Irish democracy. Earlier, in September 2017, the (minority) Government had already agreed an indicative timetable for seven referendums arising from the work of the Citizens’ Assembly, the (earlier) Convention on the Constitution, and the Programme for a Partnership Government: a referendum on ending the Constitutional prohibition on abortion (subsequently successfully held in May 2018); a referendum repealing the constitutional provision on blasphemy (successfully held in October 2018),24 a referendum on a ‘woman’s life within the home’25 (still forthcoming); plebiscites on directly elected executive mayors (held in three cities in May 2019 but successful only in one26); a referendum on liberalising divorce conditions (successfully held in

22 Thirty-fifth Amendment of the Constitution (Age of Eligibility for Election to the Office of President) Bill 2015. 23 See the Citizen’s Assembly, Report and Recommendations of the Citizens’ Assembly on the Fourth and Fifth Topics: The Manner in Which Referenda Are Held & Fixed Term Parliaments (21.06.2018), ch 1, www.citizensassembly.ie/. As regards the recommendation of the creation of a permanent electoral commission, the General Scheme of an Electoral Reform Bill was approved by the Government and circulated to the Joint Oireachtas Committee on Housing, Local Government and Heritage in January 2021. The Committee, having heard the views of stakeholders and experts, published a report in August 2021. An Electoral Reform Bill is now being drafted. (See written Answer provided to a Dáil Question by the Minister of State at the Department of Housing, Local Government and Heritage on 11 November 2021.) 24 A referendum to remove the then requirement of the criminalisation of blasphemy in Article  40.6.1° of the Constitution was held on 27.11.2018. On a relatively poor turnout of only 43.9 per cent, probably indicative of some apathy on an issue which was of little practical relevance, 64.9 per cent cast their ballots in favour of repeal. 25 The 2012–14 Constitutional Convention favoured amending Article 41.2.1° concerning the role of women in the home. Controversy over whether this paragraph should simply be deleted or amended to acknowledge in more gender-neutral language the importance of care within the home, combined with the (linked) insistence of an Oireachtas committee on pre-legislative scrutiny, derailed the Government’s intention to hold a referendum in October 2018 on this provision. The referendum was postponed to allow for further scrutiny. (C Gleeson and P Logue, ‘Referendum on “Sexist” Reference to Women’s Place in the Home Postponed’, Irish Times 05.09.2018 and Anon, ‘Why is the “Women’s Place” Referendum So Controversial?’, Irish Times 05.09.2018). In February 2020, a newly constituted Citizens’ Assembly on Gender Equality chaired by former EC Secretary-General Catherine Day, and tasked, inter alia, with considering the topic of the family in the Constitution and the law, agreed with the Constitutional Convention’s conclusion that Art 41.2.1°should be amended but divided on what should replace the 1937 formulation (R McGreevy, ‘Citizens’ Assembly backs call for “women in the home” clause to go’, Irish Times 16.02.2020). An eventual constitutional referendum on the provision of some shape on Article 41.2.1° seems inevitable, however. 26 Namely Limerick, Cork and Waterford. Only in Limerick was it successful.

216  Gavin Barrett June 2019),27 a vote on extending the franchise at presidential elections to Irish citizens resident outside the state (proposed but stalled),28 and a vote on reducing the voting age to 16 (initially slated for June 2019 but then postponed by the then Government to an unspecified future date29). The reaction to the June 2018 recommendations of the Citizens’ Assembly concerning the conduct of referendums remains to be seen. From January 2020 to April, 2021 the Citizens’ Assembly deliberated on Gender Equality. Among its recommendations were recommendations on the Constitution, namely the insertion of a new clause into Article 40 to refer explicitly to gender equality and non-discrimination, the replacement of the text of Article 41.2 with language that is not gender-specific and obliges the state to take reasonable measures to support care within the home and wider community, and finally the amendment of Article  41 so that it would protect private and family life, with the protection afforded to the family not limited to the marital family.30

III.  Constitutional Foundations of EU Membership A.  Specific Provisions on EU Membership From an Irish law perspective, the EU acquis finds its place in the national legal order by passing through the narrow Constitutional door found in Article 29.4 of the Constitution. Article 29.4.3° was inserted into the Constitution by section 1 of the Third Amendment of the Constitution Act 1972,31 making Ireland the first state to amend its Constitution to facilitate accession to the Communities.32 Originally Article 29.4.3° consisted of one paragraph: The State may become a member of the European Coal and Steel Community …, the European Economic Community … and the European Atomic Energy Community … No provision of this Constitution invalidates laws enacted, acts done or measures adopted by the State necessitated by the obligations of membership of the European Communities or prevents laws enacted, acts done or measures adopted by the Communities, or institutions thereof, from having the force of law in the State.

The insertion of this clause followed the recommendations of a 1962 inter-departmental committee chaired by the Attorney General set up to examine the legal problems connected with 27 Deleting from Article 41.3.2°.i of the Constitution, as has been seen in the text above, the requirement of a four-year period of estrangement as a condition for a divorce to be granted. 28 Although the Thirty-ninth Amendment of the Constitution (Presidential Elections) Bill 2019 was introduced in the Dáil by the Government in September 2019 in the initial anticipation of facilitating a November 2019 referendum, strong cross-party misgivings regarding both the content of the proposal and its timing rapidly forced a rapid and long overdue rethink by the minority administration. The proposed Dáil vote was shelved, supposedly in favour of proceeding at a more opportune time, most unlikely to be anytime soon. Formally, the Bill lapsed with the dissolution of the Houses prior to the February 2020 general election. 29 The Assembly recommendations had been the subject of an earlier reneged-upon promise by the previous Government to hold a referendum. The commitment to an eventual such referendum survived the 2016 election. Sponsored by independent and Sinn Féin members, the Electoral (Amendment) (Voting at 16) Bill 2016 was introduced in the Seanad as a private members’ bill in July 2016 but was rejected by the Government (with Opposition Fianna Fáil support) as premature and subsequently lapsed on the dissolution of the Dáil and Seanad in February 2020. The Bill was restored to the Order Paper in March 2021 as the Electoral (Amendment) (Voting at 16) Bill 2021, but has not progressed beyond its second stage. In May 2021, the Thirty-ninth Amendment of the Constitution (Right to vote at 16) Bill 2021 was introduced in the Dáil, but has similarly not progressed beyond its second stage. 30 See Assembly website. 31 The Third Amendment of the Constitution Act 1972, after being passed by both Houses, was approved by the ­electorate in referendum on 10.05.1972 and signed into law 08.06.1972. 32 Leonard Besselink, Monica Claes, Šejla Imamović, Jan Hermann Reestman, National constitutional avenues for further EU integration (Luxembourg, Publications Office of the European Union, 2014).

Ireland  217 accession to the Communities, including its constitutional implications.33 The new paragraph had two aspects. First, it specifically authorised acceding to the three then-extant Communities. Secondly, it created a shield of immunity protecting laws, acts or measures by the state ‘necessitated’ by the obligations of membership and preventing acts or measures by the Communities or their institutions from being deprived of legal force within the state. Overall, Article 29.4.3° obviated what would otherwise have been the need to amend or adapt numerous constitutional provisions: it was a one-stop shop for eradicating Constitutional difficulties otherwise consequent on membership. Article 29.4.3° did not effect Irish entry to the Communities, but rather merely made it constitutionally possible. Irish entry was made effective domestically by depositing the instrument of Irish ratification of the Accession Treaty with the Italian Government, and the Treaty’s subsequent entry into force. Nor did Article 29.4.3° incorporate the law of the then Communities into Irish law. That task was left to section 2 of the European Communities Act 1972.34 Five decades later, the ‘European clause’ once found in Article 29.4.3° has been amended six times and nine referendums have been held on it. The ‘European clauses’ in the Irish Constitution have expanded to eight separate subsections (now numbered Article 29.4. 3° to Article 29.4.10°), which read as follows: 3 4 5 6

The State may become a member of the European Atomic Energy Community … Ireland affirms its commitment to the European Union within which the member states of that Union work together to promote peace, shared values and the well-being of their peoples. The State may ratify the Treaty of Lisbon amending the Treaty on European Union and the Treaty establishing the European Community, signed at Lisbon on the 13th day of December 2007 (‘Treaty of Lisbon’), and may be a member of the European Union established by virtue of that Treaty. No provision of this Constitution invalidates laws enacted, acts done or measures adopted by the State, before, on or after the entry into force of the Treaty of Lisbon, that are necessitated by the obligations of membership of the European Union referred to in subsection 5° of this section or of the European Atomic Energy Community, or prevents laws enacted, acts done or measures adopted by i. the said European Union or the European Atomic Energy Community, or institutions thereof, ii. the European Communities or European Union existing immediately before the entry into force of the Treaty of Lisbon, or institutions thereof, or iii. bodies competent under the treaties referred to in this section, from having the force of law in the State.

7

The State may exercise the options or discretions i. to which Article 20 of the Treaty on European Union relating to enhanced cooperation applies, ii. under Protocol No 19 on the Schengen acquis integrated into the framework of the European Union annexed to that treaty and to the Treaty on the Functioning of the European Union (formerly known as the Treaty establishing the European Community), and iii. under Protocol No 21 on the position of the United Kingdom and Ireland in respect of the area of freedom, security and justice, so annexed, including the option that the said Protocol No 21 shall, in whole or in part, cease to apply to the State, but any such exercise shall be subject to the prior approval of both Houses of the Oireachtas.

8

The State may agree to the decisions, regulations or other acts i. under the Treaty on European Union and the Treaty on the Functioning of the European Union authorising the Council of the European Union to act other than by unanimity,

33 Involving the Departments of External Affairs and Justice, and the Office of the Attorney General, see Dáil Debates Vol 197 No 7 (14.11.1962). 34 Which provided that ‘from the 1st day of January, 1973, the treaties governing the European Communities and the existing and future acts adopted by the institutions of those Communities shall be binding on the State and shall be part of the domestic law thereof under the conditions laid down in those treaties’.

218  Gavin Barrett ii. under those treaties authorising the adoption of the ordinary legislative procedure, and iii. under subparagraph (d) of Article 82.2, the third subparagraph of Article 83.1 and paragraphs 1 and 4 of Article 86 of the Treaty on the Functioning of the European Union, relating to the area of freedom, security and justice, but the agreement to any such decision, regulation or act shall be subject to the prior approval of both Houses of the Oireachtas. 9

The State shall not adopt a decision taken by the European Council to establish a common defence pursuant to Article 42 of the Treaty on European Union where that common defence would include the State. 10 The State may ratify the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union … No provision of this Constitution invalidates laws enacted, acts done or measures adopted by the State that are necessitated by the obligations of the State under that Treaty or prevents laws enacted, acts done or measures adopted by bodies competent under that Treaty from having the force of law in the State.

These provisions (i) made Irish accession to the EU and to Euratom plus Irish ratification of certain EU treaties constitutionally permissible; (ii) immunise EU secondary laws and acts and certain ‘necessitated’ Irish measures from constitutional attack; (iii) affirm Irish commitment to the EU;35 (iv) permit the state to participate in certain optional steps at EU level (subject to prior approval of both Houses); (v) preclude Irish participation in certain decisions establishing a common European defence; and (vi) both rendered Irish ratification of the Fiscal Compact constitutionally permissible and immunise measures and acts by bodies competent under that Treaty and certain ‘necessitated’ Irish measures from constitutional attack. These provisions can be dubbed Version 2.1 of the Article 29.4 licence: they are the replacement of the original repeatedly amended EU-related clauses.36 The replacement occurred at the time of the Lisbon Treaty amendment in 2009, when the opportunity was taken to set out in a more user-friendly form Ireland’s constitutional arrangements regarding the EU. An amendment was effected to facilitate ratification and domestic implementation of the Fiscal Compact.

1. Accession The original version of the Article  29.4.3° accession clause (providing, as it did, that the state might become a member of the European Coal and Steel Community, the European Economic Community and the European Atomic Energy Community) was removed from the Constitution in 2009, but its successors are the present Articles  29.4.3° and 29.4.5°, which now provide, r­ espectively, that 3

The State may become a member of the European Atomic Energy Community …

and that 5

The State may ratify the Treaty of Lisbon amending the Treaty on European Union and the Treaty establishing the European Community, … (‘Treaty of Lisbon’), and may be a member of the European Union established by virtue of that Treaty.

35 A provision inserted by the Twenty-Eighth Amendment of the Constitution (Treaty of Lisbon) Act 2009. Its legal effect (if any) remains unclear. See further Gavin Barrett, ‘The Evolving Door to Europe: Reflections on an Eventful Forty Years for Article 29.4 of the Irish Constitution’, 48 Irish Jurist (2012) 132 Section 3. 36 Once-amended by the Thirtieth Amendment of the Constitution (TSCG) Bill 2012 (inserting Art 29.4.10° into the Constitution, facilitating ratification of the Fiscal Compact).

Ireland  219 These subsections – the function of which is similar to that which their predecessor (the original Article 29.4.3°) fulfilled regarding the original treaties – facilitated rather than effected Irish entry to the new EU created by the Lisbon Treaty.37 The most significant ruling regarding these Articles  29.4.3° and 29.4.5°38 accession clauses was the 1987 Supreme Court judgment in Crotty v An Taoiseach.39 The Supreme Court held (in the constitutionally required single judgment format) regarding the constitutionality of legislation designed to incorporate the provisions of the Single European Act (SEA) into Irish law40 that there was no unconstitutionality involved because, inter alia, the amendments effected by the SEA did not go beyond the essential scope or objectives of the original treaties. Finlay CJ (who delivered the ruling of the court) formulated the applicable test in these terms: [I]t is the opinion of the Court that the first sentence in Article 29, s.4, sub-s.3 of the Constitution must be construed as an authorisation given to the State not only to join the Communities as they stood in 1973, but also to join in amendments of the Treaties so long as such amendments do not alter the essential scope or objectives of the Communities. To hold that the first sentence of Article 29, s.4, sub-s.3 does not authorise any form of amendment to the Treaties after 1973 without a further amendment of the Constitution would be too narrow a construction; to construe it as an open-ended authority to agree, without further amendment of the Constitution, to any amendment of the Treaties would be too broad.41

2.  Treaty Amendments The above pronouncement in the Crotty case concerning the meaning of the accession clause42 doubles, of course, as a pronouncement on the extent of the discretion available to the Irish state to agree to amendments to the Treaties without thereby infringing the Irish Constitution. The Crotty approach turns largely on the question of whether a particular amending treaty alters the essential scope or objectives of the existing treaties. If it does, then Crotty establishes that the existing constitutional authorisations to join Euratom and the Union43 (and the Article  29.4.10° authorisation to ratify the Fiscal Compact) will not permit the ratification of the relevant amending Treaty. A constitutional amendment will then be necessary if the treaty’s ratification involves some unconstitutionality. If the ‘essential scope or objectives’ test in Crotty is not the most permissive one imaginable, nonetheless the court at least applied it reasonably liberally.44 Hence Finlay CJ observed on behalf of the court (regarding amendments found in the SEA) that neither the proposed changes from unanimity to qualified majority, nor the identification of topics which while now separately stated, are within the original aims and objectives of the EEC, bring these proposed amendments outside the scope of the authorisation contained in Article 29, s.4, sub-s.3 of the Constitution.45 37 Once again, actual accession to the new Union was effected by the process of ratification, culminating in the depositing of the instrument of Irish ratification of the Treaty of Lisbon with the Italian Government on 23.10.2009, and the entry into force of the Treaty on 01.12.2009. 38 Or, more precisely, their predecessor accession clause, the then Art 29.4.3°. 39 IESC, Crotty v An Taoiseach (n 16). 40 The European Communities (Amendment) Act 1986. A single judgment was required by Art  34.4.5° of the Constitution. 41 IESC, Crotty v An Taoiseach (n 16) 767. Emphasis added. 42 Now clauses. 43 Now located in Arts 29.4.3° and 5°. 44 Note also that with the expansion in the scope and objectives of the Treaties over time, the Crotty test should allow more scope for Treaty amendment than was the case in 1987. 45 IESC, Crotty v An Taoiseach (n 16) 770. The Chief Justice also warned that ‘it does not follow that all other decisions of the Council which now require unanimity could, without a further amendment of the Constitution, be changed to decisions requiring less than unanimity’.

220  Gavin Barrett The authorisation by the SEA of the attachment of a Court of First Instance to the European Court of Justice was also deemed to come within the existing constitutional authorisation, as it did ‘not affect in any material way the extent to which the judicial power has already been ceded to the European Court’. Aliter the power to adopt health and safety measures by QMV, ‘since the existing Treaty contains various provisions dealing with the approximation of laws in general, with freedom for the provision of services in the Member States, with working conditions and with the prevention of occupational accidents and diseases’.46 A test as broadly worded as the ‘essential scope or objectives’ test naturally causes uncertainty on each subsequent occasion that a new amending treaty is agreed. In the event, the application of Crotty has now given rise to seven referendums on five different European treaties.47 The reasons why a referendum is deemed by an Attorney General to be necessary are never made public. However, a referendum seemed clearly necessary under the Crotty test on the Maastricht Treaty, given its establishment of an EU with competences in areas which previously had very little European involvement, eg, justice and home affairs, a common foreign and security policy and economic and monetary union. A referendum also seems to have been necessary for the Lisbon Treaty, since this Treaty involved, eg, the establishment of a new EU with legal personality replacing both the earlier European Community and the EU. It also involved new powers being conferred on national parliaments at European level by a Protocol, giving the Dáil and Seanad a role in foreign policy (constitutionally a function of government). It also involved giving to the Charter of Fundamental Rights of the EU the same legal value as the Treaties.48 Hogan and Whyte argue Crotty also made a referendum necessary regarding the Amsterdam Treaty because of the ‘communitarisation’ of certain competences in the justice and home affairs area. This argument carries force given the importance the Supreme Court in Crotty attached to the cession of judicial power.49 They are less clear that Crotty made a referendum necessary on the Nice Treaty (which involved adjustments to the institutions so as to facilitate the enlargements of the EU in 2004 and 2007).50 The degree of uncertainty given rise to by the difficulty of applying the Crotty test for the applicability of Article 29.4.3° and 5° constitutional licences for acceding to new EU treaties probably partly underlay the announcement by the then Minister for Justice, Alan Shatter, in July 2012 that the Government would consider a referendum on allowing51 it to refer to the Supreme Court international agreements – including those relating to EU treaties – for advice on their constitutionality. That step would remove any possible doubt from the issue of whether a referendum on a given amending treaty (or indeed an entirely new Treaty in or outside an EU context) is constitutionally required.52 However, Minister Shatter resigned from his office in May 2014 for unrelated reasons and the matter subsequently faded from the agenda. It did not feature as an issue in the June 2018 Citizens’ Assembly report on the manner in which referendums are held.

46 IESC, Crotty v An Taoiseach (n 16) 770. 47 The referendum on the TSCG in 2012 was not caused by the application of the essential scope or objectives test. Being a non-EU treaty, Art 29.4 had no application. 48 See Paul Gallagher SC, ‘A Referendum for the ESM?’, IIEA/Bar Council lecture, 29.06.2011, www.youtube.com/ watch?v=5nR1Alhb6PA. Mr Gallagher was the Attorney General who advised that it was necessary to hold a referendum on the Lisbon Treaty. 49 See previous para. 50 See Hogan and Whyte, Kelly (n 10) 520. The authors speculate that the need to renumber references in Art 29.4 to Treaty provisions involving options or discretions may have created the need for a referendum on this occasion. 51 Among other matters. 52 See Carol Coulter, ‘Referendum to pave way for shake-up of courts system’, Irish Times 18.07.2012.

Ireland  221

3.  Judicial Control Rescuing various provisions of Community law from being attacked on constitutional grounds in Irish courts formed one of two major foci of the 1972 amendment to the Constitution. According to what is now Article 29.4.6° (which is a lengthier version of the second sentence in the original Article 29.4.3°), No provision of this Constitution invalidates laws enacted, acts done or measures adopted by the State, before, on or after the entry into force of the Treaty of Lisbon, that are necessitated by the obligations of membership of the European Union referred to in subsection 5° of this section or of the European Atomic Energy Community, or prevents laws enacted, acts done or measures adopted by i. the said European Union or the European Atomic Energy Community, or institutions thereof, ii. the European Communities or European Union existing immediately before the entry into force of the Treaty of Lisbon, or institutions thereof, or iii. bodies competent under the treaties referred to in this section, from having the force of law in the State.

At the time of Ireland’s accession to the Treaty of Rome, such a provision was needed to avoid numerous Articles in the Constitution colliding meteor-like into the new European legal order (mainly those ascribing and dividing governmental powers between the various Irish institutions). Among Articles in the Constitution persuasively identified as having presented potential difficulties were Articles  6 (confining governmental powers to the organs of the Irish state); 15.2.1° (vesting sole and exclusive law-making power in the Oireachtas); 28.2 (providing that the executive power of the state be exercised by or on the authority of the Government); 34.1 (requiring justice to be administered by judges appointed in the manner provided by the Constitution); 34.3.1° (providing for a High Court with full original jurisdiction); and 34.3.2° (expressly including in the jurisdiction of the High Court jurisdiction to question the constitutional validity of any law).53 Ominously, Henchy J (writing extrajudicially in 1977) opined that the approximation and harmonious development of economic activities envisaged by the EEC Treaty was also ‘incompatible with the sovereignty of the Irish people which was so explicitly stated in the Constitution’.54 He returned fatefully to this implicit linking of Article 5 of the Constitution to European affairs in his judgment in the Supreme Court majority in Crotty v An Taoiseach,55 where he held Part III of the SEA unconstitutional for infringing this very Article. Henchy J also threw Article 1 of the Constitution into the equation, opining that the right of the Irish people to control the institutions of government and to develop the life of the nation along the lines of ‘its own genius and traditions’, which was declared by the Constitution to be inalienable, has now in fact been alienated, at least in part and, according to one theory, irreversibly.56

The opening clause of what is now Article 29.4.6° constitutes an acknowledgment that it was not only the new legal order itself which (at least from the domestic Irish legal perspective) needed to be rescued from constitutional missiles, but also certain Irish laws, acts and measures – namely, those which membership made necessary. 53 See eg Bryan McMahon and Finbarr Murphy, European Community Law in Ireland (Dublin, Butterworth, 1989) 264 f; Hogan and Whyte, Kelly (n 10) 515. 54 Séamus Henchy, ‘The Irish Constitution and the EEC’, (1977) 1 Dublin University Law Journal 20, 21. Note also the observation of Fennelly J in Maher v Minister for Agriculture and Food, that the legislative capacity of the Council, and the EP as co-legislator ‘seriously encroaches on the legislative sovereignty of the State’ (IESC 30.03.2001, Case 340/2000, Maher v Minister for Agriculture and Food [2001] 2 IR 139, 248) (emphasis added). 55 IESC, Crotty v An Taoiseach (n 16). 56 Henchy, ‘Irish Constitution’ (n 54) 25.

222  Gavin Barrett The extensive shield of immunity conferred by the two clauses of what is now Article 29.4.6° constituted a major limitation on the Constitution’s reach – and one which has increased in breadth as the EU has spread its activities in the last two and a half decades into wider policy areas. Article 29.4.6° (or rather its similarly worded predecessor) was most broadly interpreted by Walsh J in Campus Oil v Minister for Industry and Energy,57 when he interpreted it as making Article 177 of the EEC Treaty58 part of Irish law, qualifying and altering the interpretation to be given to Article 34 of the Irish Constitution – thus effectively scheduling the treaties like some kind of gigantic new protocol to the Constitution’s text. This has been (arguably correctly) criticised as an excessively broad interpretation of Article 29.4.6°.59 Subsequently, however, the CJEU would formulate its principle of harmonious interpretation, which ‘requires the national court to consider national law as a whole in order to assess to what extent it may be applied so as not to produce a result contrary to that sought by EU law’.60 Should this be applied to a national provision at constitutional level, then the result yielded, ironically, may not be unlike that envisaged by Walsh J in Campus Oil.61 As has already been seen, the first limb of Article 29.4.6° provides a shield of immunity for laws, acts or measures ‘necessitated by the obligations of membership of the European Union’. It is noteworthy that the obligations of membership are referred to, not the obligations of EU law. Phelan – arguably correctly – considers this latter issue ‘partly a political question’.62 Academics dispute whether the question of whether a law, act or measure is necessitated by Union membership is a question of Irish law63 or of European law. It seems appropriate, and consistent with the approach of the Supreme Court in Meagher v Minister for Agriculture,64 to regard the concept of ‘necessitated by the obligations of membership of the European Union’ as being (a) a test determined by Irish law, (b) which is broader than but nonetheless to be expected to include anything deemed by the CJEU to be a requirement of EU law. Temple Lang’s belief65 that the original ‘necessitated’ clause was a renvoi from the Constitution of Ireland to the constitutional law of the Community (now Union)66 arguably has some truth to it.67 Irish law should indeed arguably be expected to deem anything required by EU law to be something necessitated by the obligations of membership. However, this arguably does not encapsulate either the breadth of Article 29.4.6° 57 IESC 17.05.1983, Case 1982 No 9256 P, Campus Oil v Minister for Industry and Energy [1983] IR 82. 58 Now Art 267 TFEU. 59 Hogan and Whyte, Kelly (n 10) 515. See also David O’Keeffe, ‘Preliminary Deference: The Supreme Court and Community Law’, (1983) 5 Dublin University Law Journal 286. Fennelly has (correctly, it is submitted) taken the view that the subsequent Supreme Court decision in Data Protection Commissioner v Facebook Ireland Ltd. and others [2019] 3 IR 255 (see especially) the ruling of Clarke CJ, with whom all the other judges agreed,) at paras. 55 to 69 thereof) reasoned the substantive result of Campus Oil out of existence. (See especially the ruling of Clarke CJ, with whom all of the other judges agreed, at paras. 55 to 69 thereof and see Nial Fennelly, ‘Appealing the Decision to Refer a Question to the CJEU: Is it Allowed?’ in G Berrett, J-P Rageade, D wallis and H. weil, The Future of Legal Europe: Will We Trust in it?, p. 247 at p. 266). The Facebook ruling did not address the aspect of Campus 0:1 referred to in the text above, however. 60 CJEU 16.12.2010 C-239/09 (Seydaland Vereinigte Agrarbetriebe) ECLI:EU:C:2010:778, 50. See also ECJ 16.07.2009 C-12/08 (Mono Car Styling) ECLI:EU:C:2009:466, 60–62. 61 See now in this regard, the ruling of the European Court of Justice (sitting as a Grand Chamber) in Case C-497/20 Randstad Italia SpA v Umana SpA and others ECLI:EU:C:2021:1037 at para. 54 thereof. 62 See Diarmuid Phelan, Revolt or Revolution. The Constitutional Boundaries of the European Community (Dublin, Round Hall, 1997) 345. 63 A view defended by Phelan, ibidem, 338–49. 64 IESC 18.11.1993, Case 127/1993, Meagher v Minister for Agriculture [1994] 1 IR 329. 65 See also Gerard Hogan and Anthony Whelan, Ireland and the European Union: Constitutional and Statutory Texts and Commentary (London, Sweet and Maxwell, 1995) 30. 66 See John Temple Lang, ‘The Widening Scope of Constitutional Law’, in D Curtin and D O’Keeffe (eds), Constitutional Adjudication in European Community and National Law (Dublin, Butterworth, 1992) 229, 231. See also Paul Gallagher, ‘The Constitution and the Community’, (1993) 2 Irish Journal of European Law 129, 130–32. 67 It has also garnered considerable judicial support (see below).

Ireland  223 (which could be greater than merely what European law requires) or ultimately its nature, which is Irish, not EU law, even if EU law may offer conclusive evidence of some of what Irish law (in the shape of the ‘necessitated’ clause) ought to be deemed to require. The question of when laws, acts or measures will be regarded as ‘necessitated’ is obviously a significant one. The more stringent the approach taken by the Irish courts to the concept, the more tightly drawn is this choke chain on the facilitation of EU Law’s reception into the Irish legal system. Notwithstanding the importance of the issue, the Irish courts have experienced difficulty in producing an entirely satisfactory test for determining what is ‘necessitated’,68 and there has been a lack of consistency, in particular in the Supreme Court’s approach to this issue.69 At the interlocutory High Court stage in Crotty v An Taoiseach,70 Barrington J described the significance of the ‘necessitated’ clause as being that ‘the Constitution could not now be invoked to invalidate any measure which the state was directed by the institutions of the EEC to take arising out of the exercise of their powers’71 – seemingly drawing a link between the requirements of EU law and the applicability of the ‘necessitated’ clause. The High Court plenary ruling dwelt somewhat longer on what is now the Article  29.4.6° immunity clause (of which, as we have seen, the ‘necessitated’ clause forms the first branch). According to Barrington J, who delivered the three-man Court’s ruling, the immunity conferred by the second sentence of the Third Amendment would appear to apply to legislative and administrative measures taken in the day-to-day running of the Community … Put another way, there are some acts of the institutions of the Community which are directly enforceable in all the Member States whereas others require legislative or administrative action by the Member States to procure their enforcement. It is these matters which are referred to in the [immunity clause] … It is these matters alone which are given immunity from constitutional challenge by the [immunity clause]. But such of these matters as are acts of the institutions of the Communities derive their status in domestic law from the European Communities Act, 1972. If the [immunity clause] is the canopy over their heads, the Act of 1972 is the perch on which they stand.72

On appeal, the Supreme Court took a similar view of the non-application of the immunity clause to treaties, and found the ‘necessitated’ clause to the SEA – a treaty which amended the original EEC Treaty. Finlay CJ (who gave the single judgment of the Court in this respect) held that ‘it is clear and was not otherwise contended by the defendants that the ratification by the State of the SEA … would not constitute “an act necessitated by the obligations of membership of the Communities”’. For the Supreme Court and the High Court alike then, in so far as the constitutionality of the SEA rested on the Article 29.4 accession clause, not the immunity clause.

68 Gallagher, ‘Constitution and Community’ (n 65) 131. 69 See Barry Doherty, ‘Land, Milk and Freedom – Implementing Community Law in Ireland’, (2004) 11 Irish Journal of European Law 141, 171. 70 IESC, Crotty v An Taoiseach (n 16). 71 Ibid, 727 f. 72 Ibid, 758 (emphasis added). The thinking underlying this rejection of the idea of applying the immunity clause to treaties was explained by Barrington J thus: ‘Had the Oireachtas not passed the European Communities Act 1972, Ireland might still have been a member of the Community in international law but it would have been in breach of its obligations in international law under the Treaty of Rome and under the Treaty of Accession. This however would not have been a matter regarding which the domestic courts of this country would have had any competence because the Treaty would not have been part of the domestic law. The immunity from constitutional challenge conferred by [the second limb of the immunity clause] on laws enacted, acts done, or measures adopted by the Community or its institutions would therefore have been meaningless as these laws, acts or measures would not have been part of the domestic law of this country. To make them part of the domestic law of this country the European Communities Act, 1972, was necessary. This Act cannot therefore have been passed by virtue of [the immunity clause] but by virtue of the licence to join the European Community contained in the first sentence of the Third Amendment.’

224  Gavin Barrett In three High Court cases decided shortly after Crotty, more focus was brought to bear on the exact nature of the licence for action provided by the ‘necessitated’ clause. In Lawlor v Minister for Agriculture,73 Murphy J stated: [I]t seems to me that the word ‘necessitated’ in this context must extend to and include acts or measures which are consequent upon membership of the Community and are in general fulfilment of the obligations of such membership and even where there may be a choice of degree or discretion vested in the State as to the particular manner in which it would meet the general spirit of its obligations of membership.74

Murphy J himself clearly regarded such an approach as having its limits. Six months after his Lawlor ruling, the same judge observed in Greene v Minister for Agriculture that ‘there must be a point at which the discretion exercised by the State or the national authority is so far-reaching or so detached from the result to be achieved by the directive that it cannot be said to have been necessitated by it’.75 Overall, this may be regarded as having involved taking a moderately liberal approach to ‘necessitated’, viewing it as encompassing laws, acts or measures in a penumbra going beyond – but not too far beyond – the strict requirements of EU law. An alternative (and broader-seeming) reasonableness-based version of the same kind of ‘penumbra’ approach can be seen in the ruling of Lynch J in Condon v Minister for Agriculture, where that judge, speaking of domestic measures implementing a scheme envisaged by a European-level regulation, held that insofar as such details of implementation are reasonable they must be regarded as necessitated by the obligations of membership of the Communities and cannot therefore be unconstitutional. If however the details of the implementation were unreasonable or unfair then they could hardly be said to be necessitated by the obligations of membership of the Communities and they would be open to constitutional challenge.76

Although the decisions have attracted some criticism,77 they are far from being without merit: they left the Government the kind of room for manoeuvre in implementing European law the creation of which, one suspects, motivated the drafting of the immunity clause in the first place. In subsequent cases on the immunity clauses, focus shifted to the issue of the form of domestic law instrument used to implement EU law measures, and whether the choice of form could itself be immunised.78

B.  Continued Relevance of General Provisions on International Law The Irish legal system is dualist (at least outside the scope of EU law). This dualism is written into Article 29.6, which provides that no international agreement shall be part of the domestic law of the state save as may be determined by the Oireachtas. The usual formalities regarding treaty ratification are adhered to in the case of EU treaties. If, above and beyond this, it is feared that implementing an international treaty will involve unconstitutionalities, the normal approach has been to include a provision authorising ratification



73 IEHC

02.10.1987, Case 1986 No 11837 P, Lawlor v Minister for Agriculture [1990] 1 IR 356; [1988] ILRM 400. 1 IR 356, 377. 75 IEHC 04.05.1989, 1989 WJSC-HC 1577, Greene v Minister for Agriculture [1990] 2 IR 17, 25. 76 IEHC 12.10.1990, 1990 WJSC-HC 2557, Condon v Minister for Agriculture [1993] IJEL 151, 156 (emphasis added). 77 See eg Hogan and Whyte, Kelly (n 10) 522–24. 78 This case law is returned to below. 74 Ibid,

Ireland  225 of the Treaty in question in the Constitution (an approach which requires a referendum). This approach was taken regarding the Lisbon Treaty.79 But it has also been done regarding other non-EU treaties. Sometimes all that is necessary is a bare permission to ratify (Article 29.9 simply announces ‘the State may ratify the Rome Statute of the International Criminal Court done at Rome on the 17th day of July, 1998.’). Sometimes, the amendment goes further. Article 29.4.6° (regarding EU treaties) is an example of this. Another example is 29.4.10° which contains its own immunity clause, modelled on what is now Article 29.4.6°, and provides: The State may ratify the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union done at Brussels on the 2nd day of March 2012. No provision of this Constitution invalidates laws enacted, acts done or measures adopted by the State that are necessitated by the obligations of the State under that Treaty or prevents laws enacted, acts done or measures adopted by bodies competent under that Treaty from having the force of law in the State.80

Article  29.5 of the Constitution also imposes procedural requirements. Under Article  29.5.1°, every international agreement to which the state becomes a party must be laid before Dáil Éireann. Under Article 29.5.2° the state shall not be bound by any international agreement involving a charge upon public funds unless the terms of the agreement shall have been approved by Dáil Éireann.81 Such provisions are adhered to even in the case of EU treaties.

C.  Practice and Doctrinal Debate on These Provisions Much of the debate in the case law has focused on immunity and more specifically on the topic of what laws, acts or measures done by the state will be regarded as necessitated.

IV.  Constitutional Limits to EU Integration A.  Limits to the (Further) Transfer of Powers to the EU Through Treaty Amendments One of the main limits to the further transfer of powers to the EU through treaty amendments has already been seen. According to the 1987 Supreme Court ruling in Crotty v An Taoiseach the accession clauses now found in Articles 29.4.3° and 29.4.5° of the Constitution ‘must be construed as an authorisation given to the State not only to join the Communities as they stood in 1973, but also to join in amendments of the Treaties so long as such amendments do not alter the essential scope or objectives of the Communities.’82 Where this limit is transgressed (as it was in the case of Crotty itself, by Title V of the SEA), all of the remaining provisions of the Irish Constitution retain their relevance, which is also the case where a treaty is not actually an EU treaty at all (eg, the Fiscal Compact). With Crotty having been to date the only case to have been decided on the point, it can be very difficult to know in advance when an amending Treaty goes beyond the essential scope or 79 Under Art 29.4.5° ‘the State may ratify the Treaty of Lisbon …, and may be a member of the European Union established by virtue of that Treaty.’ Note also, concerning the Euratom Treaty, Article 29.4.3°, ‘the State may become a member of the European Atomic Energy Community.’ 80 See also Art 29.7 related to the British-Irish Agreement of 10 April, 1998 (‘Good Friday Agreement’). 81 Although under Art 29.5.3° neither requirement applies to agreements or conventions of a technical and administrative character. 82 Ruling of Finlay CJ (who delivered the ruling of the Court), IESC, Crotty v An Taoiseach (n 16) 767 (emphasis added).

226  Gavin Barrett objectives of the existing Treaties. Certainty would normally only be capable of being provided by (a) a reference by the Irish President to the Supreme Court of any legislative Bill purporting to incorporate the terms of the Treaty into Irish law for an opinion on its Constitutionality,83 or else by (b) a challenge by a private party to the constitutionality of incorporating legislation and/or to any attempt by the Government to ratify the Treaty or any part thereof without a referendum. In practice, every Irish Government ratifying a Treaty since Crotty has found itself under immense pressure to hold a referendum even when there is merely a reasonable probability of Treaties going beyond the essential scope/objectives of existing constitutive Treaties. From a political perspective, the Crotty case has also contributed to an expectation in parts of the Irish public that every major European treaty will be accompanied by a referendum – and thus to political pressure to proceed in this manner. Crotty led to the deployment in Ireland of direct democracy in the place of representative democracy to an unprecedented extent. To the extent that the Constitution is deemed to mandate a referendum, the role of the legislative and executive branches is reduced to one deciding matters such as the form of the referendum proposal and the timing of the referendum.84

B.  Procedural Constitutional Limits on Further EU Integration Two procedural constitutional limits to further EU integration merit special mention. First, Articles 29.4.7° and 29.4.8° introduce constitutional protection – but also a requirement of prior approval of both Houses of the Oireachtas – for an enumerated set of decisions, which are envisaged in EU Treaties, but which can not in any way be described as necessitated by the obligations of membership and thus are not covered by the Article 29.4.6° immunity clause. These choices are returned to later. Secondly, it should be recalled that the holding of a referendum itself (including a referendum on a European Treaty) triggers the application of certain strict procedural rules held implicitly required by the Supreme Court. Hence, in McKenna v An Taoiseach (No 2),85 the Supreme Court granted a declaration that the Government, in expending any public monies whatsoever in the promotion of a particular result in a referendum, was acting unconstitutionally. In Coughlan v Broadcasting Complaints Commission and RTÉ,86 the Supreme Court effectively conferred an equal entitlement to broadcasting time on both sides in referendum campaigns completely regardless of their electoral representativity. Such rules are not without merit, but do mean that winning a referendum is rendered far more difficult for any Government than would otherwise be the case. As has been seen, in its 2018 report on the manner in which referendums are held, the Citizens’ Assembly recommended retaining the current exclusion of the Government spending public money to advocate for any particular referendum outcome. It also made the welcome, if practically challenging, recommendation that effective spending limits be introduced in referendum campaigns for political parties, campaign groups and individuals. Such limits (which, surprisingly, were not introduced immediately after the McKenna ruling) might restrict the possibility 83 Such references can be made under Art 26 entirely at the President’s discretion (subject to an obligation to consult the Council of State first). They are thus outside the control of the Government. Note that Art 26, according to its own terms, has no application to a Bill expressed to be a Bill containing a proposal to amend the Constitution. 84 Although the need to have the Treaty ratified within a reasonable time frame by all Member States will in practice impose limits even on the timing of a referendum. 85 IESC 17.11.1995, Cases 361 & 366/1995, McKenna v An Taoiseach (No 2) [1995] 2 IR 10. 86 IESC 26.01.2000, Cases 142 & 148/1998, Coughlan v Broadcasting Complaints Commission and RTÉ [2000] 3 IR 1.

Ireland  227 of referendum results being unduly influenced by the injection of large sums of privately sourced money. The Assembly’s suggestion of a permanent electoral commission might also help ensure the proper conduct of referendums, and is envisaged by the Government sponsored Electoral Reform Bill in the process of being drafted at the time of writing.

1.  Core Competences The Constitution has not been deemed to exempt particular rights or competences from its own amendment procedures, even if some of its language might be argued to suggest that this is the case – eg, the description in Article 41 of the family as ‘a moral institution possessing inalienable and imprescriptible rights, antecedent and superior to all positive law’, and the description in Article 43 of the Constitution of the right to private ownership of property as ‘a natural right, antecedent to positive law’.87 There is no theory of unamendable rights in Irish constitutional jurisprudence. To the contrary, in Re Article  26 and the Regulation of Information (Services Outside the State for Termination of Pregnancies) Bill 199588 and in some subsequent rulings, the Supreme Court has held that there could be no question of a constitutional amendment properly placed before the people and approved by them being held unconstitutional.89 This would include an amendment inserted as a result of a referendum on an EU treaty. Thus ‘no constitutional provision is immune from amendment, and a constitutional amendment cannot itself be unconstitutional’.90 There is a specifically European aspect to this in that the immunity clause in the Irish Constitution makes provision for no exceptions in its declaration that ‘no provision of this Constitution’ invalidates acts of EU institutions or necessitated acts of the Irish state.91

2.  Constitutional Identity No explicit doctrine of ‘constitutional identity’ exists in Irish law, requiring decisions regarding particular issues to remain with the national Parliament, executive or electorate.92 87 While it is thus true that no explicit doctrine of inalienable core constitutional competences exists in Irish law, it is worth noting that the rather different question of institutional competences formed a key element of the reasoning in the second part of the Crotty judgment (which largely centred around the notion of sovereignty). Central here was the idea that, as Hederman J (one of the three-judge majority in that ruling) put it, ‘the State’s organs cannot contract to exercise in a particular procedure their policy-making roles or in any way to fetter powers bestowed unfettered by the Constitution. They are the guardians of these powers – not the disposers of them.’ (IESC, Crotty v An Taoiseach [n 16] 794.) Along similar lines, Walsh J, also in the majority, observed that ‘in enacting the Constitution the people conferred full freedom of action upon the Government to decide matters of foreign policy [inter alia] … In my view, this freedom does not carry with it the power [inter alia] to abdicate that freedom.’ (Ibid 783). Any such transfer of competence will thus be an unconstitutional infringement of the distribution of powers envisaged in the Constitution. However, any such unconstitutionality can of course be remedied by referendum. This non-transferability of competences conferred by the Constitution on the branches of government likely played a strong role in the decision to hold a referendum regarding the Fiscal Compact, given the limitations on national executive, ­legislative and judicial power which this treaty involved. 88 IESC 12.05.1995, Case 87/1995, Re Article 26 of the Constitution and the Regulation of Information (Services outside the State for the Termination of Pregnancies) Bill 1995 [1995] 1 IR 1. 89 See Fiona de Londras and David Gwynn Morgan, ‘Constitutional Amendment in Ireland’, in X Contiades, Engineering Constitutional Change. A Comparative Perspective on Europe, Canada and the USA (Abingdon, Routledge, 2013) 179. 90 Ibid. 91 Contrast the corresponding provision of the German Grundgesetz, which subjects German cooperation to the principles of democracy, the rule of law, social and federal principles, the principle of subsidiarity and comparable protection of fundamental rights – and also specifically asserts the applicability of the German Constitution’s Ewigkeitsklausel which precludes constitutional amendments affecting Germany’s federal structure, human dignity or democracy, and see Maria Cahill, ‘Constitutional Exclusion Clauses, Article 29.4.6 and the Constitutional Reception of European Law’, (2011) 34 Dublin University Law Journal 74, 77 f. 92 Cf Julian Sterck, ‘The Constitutional Identity of Member States and the Primacy of European Union Law. A Comparative Study of Ireland and France’ (unpublished PhD thesis, University College Dublin, 2013).

228  Gavin Barrett Two sources might arguably have suggested themselves a basis for a doctrine of unamendable rights.93 First, the strong tradition of reliance on natural law reflected in fundamental rights jurisprudence94 and in the Constitution’s text.95 Previously, strong indications were given (both judicially and extra-judicially) that such rights were not amendable. In McGee v Attorney General, Walsh J cited the Constitution’s text as indicating that natural rights, or human rights, were not created by law but had their existence confirmed and were given protection by the Constitution.96 Walsh J asserted extra-judicially that ‘quite clearly there are fundamental rights which may not be trampled upon even with the support of an overwhelming majority of the people.’97 However, the Supreme Court’s ruling in Re Article 26 and the Information (Termination of Pregnancy) Bill 199598 signalled a move away from reliance on natural law reasoning. The Constitution’s Preamble (with its explicitly Christian references) was suggested as another arguable source of unamendable rights,99 but modern judicial support for this seems unlikely. Notably, Kearns P refused permission to two litigants to bring petitions challenging the outcome of the May 2015 same-sex marriage referendum on grounds including the alleged conflict of the referendum result with the Constitution’s Christian ethos – an area Kearns P reportedly stated the Court had no power to address.100 Notwithstanding the absence of an express constitutional identity doctrine, there are nevertheless issues which in the past have raised particular sensitivities in the Courts – although never to the extent of a finding that amendments concerning these issues would be in any way invalid. The issue of Irish military neutrality is one of these. Ironically, this is not provided for in the Irish 1937 Constitution, but following the decision of the then Taoiseach Eamon de Valera to make Ireland the sole Commonwealth non-belligerent in World War II, military neutrality attained near-canonical political status. The rather harmless provisions in the SEA concerning cooperation in the field of foreign policy became the reason seized on by the Supreme Court in the Crotty case for holding the ratification of that treaty unconstitutional (a legal position reversed in the subsequent referendum). In order to secure electoral approval for a constitutional amendment to facilitate ratification of the Nice Treaty, the Twenty-Sixth Amendment of the Constitution Act 2002 also made provision for the insertion of a new subsection into Article 29.4. In its present (subsequently renumbered) form, this subsection now provides that 9° The State shall not adopt a decision taken by the European Council to establish a common defence pursuant to Article 42 of the Treaty on European Union where that common defence would include the State. 93 See Phelan, Revolt or Revolution (n 61) 361–67. 94 Note in particular the Supreme Court rulings in IESC, Ryan v Attorney General (n 17), IESC 22.07.1966, State (Nicolaou) v An Bord Uchtála [1966] IR 567 and IESC, 19.12.1973, Case 1971 No 2314 P, McGee v Attorney General [1974] 284. In McGee, Walsh J observed (at 318) that ‘Articles 41, 42 and 43 emphatically reject the theory that there are no rights without laws, no rights contrary to the law and no rights anterior to the law. They indicate that justice is placed above the law and acknowledge that natural rights, or human rights, are not created by law but that the Constitution confirms their existence and gives them protection.’ 95 See Arts 41, 42 and 43 of the Constitution, the former two articles concerning the rights of the family (which is described in Art 41 as ‘the natural primary and fundamental unit group of Society, and as a moral institution possessing inalienable and imprescriptible rights, antecedent and superior to all positive law’), the latter article concerning private property (‘a natural right, antecedent to positive law). 96 IESC, Article 26 (n 87) 310. 97 Foreword to James Casey, Constitutional Law in Ireland (London, Sweet & Maxwell, 1987) 9. Note that in his dissenting judgment in State (Ryan) v Lennon, Kennedy CJ asserted the existence of natural law limits to the power to amend the 1922 Constitution (IESC, 19.12.1934 State (Ryan) v Lennon [1935] IR 170, 204). 98 IESC, Article 26 (n 87). 99 The Christian and democratic nature of the state was relied on by Kenny J as a foundation for constitutional fundamental rights in IESC, Ryan v Attorney General (n 17) 312. 100 See ‘High Court Rejects Two Challenges to Yes Vote’, Irish Times 06./07.06.2015. Appeals to the Court of Appeal were unsuccessful and further appeals to the Supreme Court refused. See Mary Carolan, ‘Marriage Referendum Appeals Rejected by Supreme Court’, Irish Times 17.09.2015.

Ireland  229 Article 29.4.9° does not provide any kind of absolute guarantee of Irish neutrality since it does not prevent some other method being used to engage in a military alliance (such as an international treaty), whether at European level or otherwise. It does, however, block the main route to such an alliance provided for under the treaties as they presently stand. Another ‘hot button’ issue was formerly that of abortion. However, with the decline in influence of the Catholic Church, majority public opinion in Ireland has moved into line with other EU Member States and, as has been seen, the Constitutional prohibition on abortion was deleted in the wake of a referendum vote in May 2018.101 Previously, in Society for the Protection of Unborn Children (Ireland) Ltd v Grogan,102 involving an apparent opposition between the right to life (under Article 40.3.3° of the Constitution) and the right under EEC law to supply services (or, more exactly, to provide information in order to facilitate the provision of services), the Supreme Court had indicated a willingness to disregard the supremacy principle, either (a) by means of balancing different constitutional provisions (in this case the European clause (then Article 29.4.3°) and Article 40.3.3°), and deciding in favour of Article 40.3.3° because the latter constitutional provision was adopted later in time; or (b) by using in effect an ultra vires approach and declaring that it could not be an objective of the European Communities that a member state should be obliged to permit activities which were ‘clearly designed to set at nought the constitutional guarantees for the protection within the State of a fundamental human right’.103

3.  ‘Sovereign’ State Article 5 of the Irish Constitution provides that Ireland is a ‘sovereign, independent, democratic state’. The reference to Ireland being a ‘sovereign’ state became a key plank in the Crotty judgment. An unpersuasive super-rigorous application of it led the Supreme Court by a 3:2 majority to find ratification of Title III of the SEA unconstitutional.104 Thus eg, Henchy J asserted that the State’s right to conduct its external relations is part of what is inalienable and indefeasible in what is described in Article 5 as ‘a sovereign, independent, democratic State’. It follows … that any attempt by the government to make a binding commitment to alienate in whole or in part to other States the conduct of foreign relations would be inconsistent with the government’s duty to conduct those relations in accordance with the Constitution.105

The requirements of sovereignty, which the Supreme Court asserted would be transgressed by the SEA’s ratification, were so demandingly framed by the Court that continued adherence to them

101 This was effected by the Thirty-sixth Amendment of the Constitution Act 2018 (approved in referendum on 25 May 2018). The Health (Regulation of Termination of Pregnancy) Act 2018 was subsequently enacted in order to provide for and regulate terminations of pregnancy. 102 IESC 06.03.1997, Case 317/92, Society for the Protection of the Unborn Child v Grogan [1989] IR 753. 103 See in both respects, judgment of Walsh J, ibid, 768 f. 104 Which was regarded by the IESC as lacking any protection under the accession clause of the then Art 29.4.3° since it went beyond the essential scope or objectives of the existing treaties. Notwithstanding the Supreme Court ruling in Crotty, the SEA vies with the Nice Treaty as the treaty in relation to which the constitutional need for a referendum vote has been the most questionable. For some concerns regarding the Crotty case, see Gavin Barrett, ‘Building a Swiss Chalet in an Irish Legal Landscape? Referendums on European Union Treaties in Ireland and the Impact of Supreme Court Jurisprudence’, 5 European Constitutional Law Review (2009) 32 and Gavin Barret, A Road Less Travelled. Reflections on the Supreme Court Rulings in Crotty, Coughlan and McKenna (No. 2) (Dublin, IIEA, 2011). See also on this part of the Crotty case, David Fennelly, ‘Crotty’s Long Shadow: the European Union, the United Nations and the Changing Framework of Ireland’s International Relations’, in E Carolan (ed), The Constitution of Ireland. Perspectives and Prospects (London, Bloomsbury, 2012) chapter 33. conference paper, The Irish Constitution: Past, Present and Future, Honourable Society of the Kings Inns, Dublin, June 28–29, 2012. For a more favourable reading of it, compare Hogan and Whelan, Ireland and the EU (n 64) 42–49. 105 IESC, Crotty v An Taoiseach (n 16) 787.

230  Gavin Barrett would have raised doubts even as to the compatibility of Ireland’s membership of the UN.106 The Supreme Court itself arguably validated such criticisms in the 1990 case of McGimpsey v Ireland107 where it distinguished this aspect of Crotty – rather unconvincingly – rather than apply it regarding the 1985 Anglo-Irish Agreement.108 Subsequently, in the Pringle case, change came: a reinterpretation of Crotty was applied by the Supreme Court.109 The judgment of Clarke J (as he then was) may be regarded as the clearest expression of the majority approach.110 Clarke. favoured a restrictive reading of the earlier case, observing that, on a narrow reading of some of the passages cited it might be said that this court, in Crotty, came to the conclusion that the overall architecture of the Irish Constitution does not permit the Government, in exercise of its power to conduct the foreign policy of the State in accordance with Art 29.4.1° of the Constitution, to enter into binding arrangements with other countries which would, in any way, have the effect of circumscribing Ireland’s freedom of action in the foreign policy field … It does not seem to me that such a conclusion can be found in the judgments of this court in Crotty.111

Clarke J went on to formulate what in effect was a new, more liberal test of what was permitted to the Government: the limit on the discretion which the Government holds arises where the relevant treaty involves Ireland in committing itself to undefined policies not specified in the treaty and in circumstances where those policies, which Ireland will be required to support, are to be determined not by the Government but by institutions or bodies specified in the treaty. It is an abdication, alienation or subordination of policy formation and adoption which is not permitted. A transference of the means of implementing a policy agreed by the Government, and specified in the treaty concerned, to an appropriate implementation institution or body may be permitted provided that it does not go so far as to amount, in substance, to an abdication, alienation or subordination of the role of government under the Constitution.112

For Clarke J, it seemed that the European Stability Mechanism Treaty challenged in Pringle was thus an exercise in sovereignty, rather than an abdication or transfer of it.113 Thus no unconstitutionality was involved in the Treaty’s ratification (and therefore no referendum to facilitate it was needed).

106 See Gerard Hogan, ‘The Supreme Court and the Single European Act’, (1987) 22 Irish Jurist (ns) 55, 69; also Anthony Whelan and Liz Heffernan, ‘Ireland the United Nations and the Gulf Conflict: Legal Aspects’, (1991) Irish Studies in International Affairs 115, 140–45. 107 IESC 01.03.1990, Case 314/88, McGimpsey v Ireland [1990] 1 IR 110, [1991] ILRM 400. 108 The main – apparently significant – distinction to which reference was made by the Court in McGimpsey was the fact that the Anglo-Irish Agreement was an agreement reached between two sovereign governments rather than 12 (ibid 121 f). Finlay CJ who delivered a judgment agreed to by four of the five judges also noted that ‘the Government of Ireland at any time carrying out the functions which have been agreed under the Anglo-Irish Agreement is entirely free to do so in the manner in which it, and it alone, thinks most conducive to the achieving of the aims to which it is committed. A procedure which is likely to lead to peaceable and friendly cooperation at any given time must surely be consistent with the constitutional position of a state that affirms its devotion not only to the ideal of peace and friendly cooperation but to that ideal founded on international justice and morality’ (ibid 121). Where, however, the Court saw any distinction in this regard between the SEA and the Anglo-Irish Agreement is not apparent. 109 IESC 19.10.2012, Case 339/2012, Thomas Pringle v The Government of Ireland, Ireland and the Attorney General [2012] IESC 47. 110 The sole dissenting voice in the Supreme Court ruling by Denham CJ, Murray J, Hardiman J, Fennelly J, O’Donnell J, McKechnie J and Clarke Jwas provided by Hardiman J Clarke J’s analysis was accepted by both O’Donnell J and McKechnie J and is compatible with the analysis of Denham CJ. 111 IESC, Thomas Pringle (n 108), para 4.16 of the ruling of Clarke J. 112 Ibid, para. 4.25 (emphasis added). 113 Ibid, para. 8.13.

Ireland  231 The ruling in Pringle has already led to a reduction in the number of referendums in Ireland in that, because of it, no referendum needed to be held on the ESM Treaty itself. The extent of the reduction in the number of referendums that would otherwise have been caused by the Crotty ruling as originally understood (and as understood in Hardiman J’s dissenting opinion in Pringle) remains to be seen. Whether a referendum must be held on any given treaty will now depend on various factors, including whether undefined policies are envisaged in the treaty in question; whether the decision to support these policies is taken out of the hands of the Government by that treaty; and whether that Treaty is capable of attracting immunity from constitutional attack under the ‘European clause’ in the Constitution in Article 29.4 thereof.

4.  Respect for Certain Constitutional Features at European Level (Standard of Fundamental Rights Protection, Democracy, Rule of Law, Social Cohesion, etc) The standard of human rights protection has perhaps shown the most potential for creating judicial disagreement between the national and European levels, either via the operation of an ultra vires approach or otherwise. Hence, a little remarked-upon element of the Crotty ruling was the observation of the Court (in that part of the ruling delivered by Finlay CJ) that ‘[certain provisions of the SEA did not] alter the essential character of the Communities. Nor has it been shown that they create a threat to fundamental constitutional rights.’ The latter point is significant since it seems to imply that that the accession clause in Article 29.4.3° alone (a) would not extend to anything going beyond the essential scope or objectives of the existing treaties, which would be ultra vires the then accession clause, it would also (b) not extend to authorising threats to fundamental constitutional rights. This point was subsequently given very little attention however, with all of the main focus of both commentary and case law having been on the essential scope or objectives test, and this nascent Bundesverfassungsgericht-like approach has not subsequently been built upon. Another case in which human rights considerations were more directly concerned has already been remarked upon, and as noted, in Society for the Protection of Unborn Children (Ireland) Ltd v Grogan,114 the Supreme Court indicated its willingness to disregard the supremacy principle in order to uphold a fundamental right, although in the event, this did not prove necessary. Potential for conflict over human rights has thus not been entirely absent but has never to date generated more than occasional judicial straws in the wind.

C.  Scrutiny of Secondary Legislation (eg, via an Ultra Vires Doctrine) 1.  Against the Yardstick of National Constitutional Requirements The idea of ultra vires scrutiny of EU law generally has surfaced only rarely in the jurisprudence of the Irish courts, in cases such as Crotty and Grogan, both of which involved provisions of founding or amending EU Treaties. Such an approach could conceivably also be deployed against secondary EU legislation although this has not happened in practice, perhaps because of the very wide scope of the Article 29.4.6 immunity clause, formulated as it is, without restrictions or exceptions.

114 IESC,

SPUC v Grogan (n 101).

232  Gavin Barrett

2.  Against the Yardstick of EU Law Irish courts do not scrutinise EU secondary legislation against the yardstick of EU law.115 Instead, and in accordance with the Foto-frost doctrine,116 the question of the validity of EU secondary legislation under EU law will normally be made the subject of a preliminary ruling to the European Court of Justice. An example of this happening was seen in the well-known case of Digital Rights Ireland Ltd.,117 in which the Court of Justice held the 2006 Data Retention Directive to be invalid.118

D.  Limits to European Integration Outside the EU Legal Order (eg the ESM, the Fiscal Compact) The full panoply of constitutional provisions may be deployed against non-EU treaties in any legal challenge. As already noted, if it is feared that an international treaty will involve any unconstitutionality, the normal approach has been to insert a provision into the Constitution authorising the Treaty’s ratification (an approach which requires a referendum), either via a bare permission to ratify, or with additional provisions such as an immunity clause (as in the case of EU Treaties and the Fiscal Compact). If use of a Fiscal Compact-type instrument – meaning a non-EU treaty capable of coming into force upon ratification by less than the full complement of all member states of the EU – is made again, the European clauses of Article 29.4 in their present state will provide no constitutional protection for accession to any such treaty (as by definition it will not be an amending treaty capable of falling within the essential scope or objectives of existing treaties, to use the Crotty formula). The future will then see further attempts to amend Article 29.4 in a manner similar to that seen regarding the Fiscal Stability Treaty in what is now Art.29.4.10°.

V.  Constitutional Rules and/or Practice Concerning the Implementation of EU Law A.  Binding (Parliamentary or Governmental) Resolutions for the Ministers in the Council? Notwithstanding the recommendations of two separate Oireachtas Sub-Committees,119 recent Irish governments have not sought to empower the Oireachtas in a manner similar to that of 115 One case which might have offered some perspective of this happening was IESC, Campus Oil (n 57). Had its approach been adhered to, it would have offered obvious potential for such scrutiny. 116 See Case 314/85 Firma Foto-frost v Hauptzollant Lübeck-Ost, ECLI:EU:C:1987:452. 117 Case C-293/12 Digital Rights Ireland Ltd. v Minister for Communications, Marine and Natural Resources and others ECLI:EU:C:2014:238. 118 See also Case C-311/18 Data Protection Commissioner v Facebook Ireland Ltd. and Schrems ECLI:EU:C:2020:559 in which the Court of Justice held Commission Implementing Decision (EU) 2016/1250 of 12 July 2016 to be invalid. 119 See Report of the Sub-Committee on Review of the Role of the Oireachtas in European Affairs, July 7, 2010 and Chapter 4 of Ireland’s Future in the European Union: Challenges, Issues and Options (the report of Oireachtas Sub-Committee on Ireland’s Future in the European Union) November 2008. The Joint Committee on EU Affairs, in its 2014 report Assessment of Current Structures for Oireachtas Scrutiny of EU Affairs, also criticised the failure to introduce a scrutiny reserve (see paras 6 f thereof).

Ireland  233 other Member State parliaments in the field of European affairs, via the introduction either of a mandate system (following the model of the Danish Parliament) or a scrutiny reserve system (similar to that used in the UK Parliament). A limited mandate system exists under Articles 29.4.7° and 8° of the Constitution in that Ireland may agree to certain EU-related decisions only if this is approved by both Houses.120 The Article 29.4 immunity clause originally envisaged only laws, acts and measures ‘necessitated’ by the obligations of membership. With the emergence at Amsterdam of the possibility of closer cooperation between some Member States, and the possibility that integration within the EU might sometimes be a voluntary choice rather than a necessity (once agreed upon) for all Member States, the constitutional facilitation of such participation in the relevant policy areas became necessary. A desire to provide extra democratic legitimacy regarding such enhanced integration beyond that provided by amending the Constitution also emerged during the Amsterdam Treaty ratification process.121 Five options or discretions were thus made subject to a parliamentary prior approval requirement122 (concerning justice and home affairs cooperation; establishing closer cooperation between some Member States; the field of visas, asylum, immigration and other policies related to free movement of persons; the Schengen Protocol; and the Protocol on the position of the UK and Ireland).123 Subsequently, the Twenty-Sixth Amendment of the Constitution Act 2002 (facilitating the Nice Treaty’s ratification124) made the exercise of six further options or discretions subject to prior approval of both Houses (all relating to the Nice Treaty provisions concerning enhanced cooperation).125 Further reforms effected by the Lisbon Treaty superseded certain options or discretions in the field of free movement and justice and home affairs. Consequently, section 1 of the Twenty-Eighth Amendment of the Constitution (Treaty of Lisbon) Act 2009126 amended Article 29.4, reducing to three the number of options or discretions requiring approval of the Houses of the Oireachtas. Article 29.4.7° now provides: The State may exercise the options or discretions i. to which Article 20 of the Treaty on European Union relating to enhanced cooperation applies, ii. under Protocol No 19 on the Schengen acquis integrated into the framework of the EU annexed to that treaty and to the Treaty on the Functioning of the European Union (formerly known as the Treaty establishing the European Community), and iii. under Protocol No 21 on the position of the United Kingdom and Ireland in respect of the area of freedom, security and justice, so annexed, including the option that the said Protocol No 21 shall, in whole or in part, cease to apply to the State, but any such exercise shall be subject to the prior approval of both Houses of the Oireachtas.

120 Arts 29.4.7° and 29.4.8° were cited, although somewhat peripherally, in argument in IEHC 06.06.2012, Case 2012 481JR, Doherty v The Referendum Commission [2012] IEHC 211 (High Court) without, however, the court finding it necessary to express a definitive view on the argument made regarding them (see paras 62 f of the ruling of Hogan J). 121 See more generally, Hogan and Whyte, Kelly (n 10) 518, 529 f. 122 See Eighteenth Amendment of the Constitution Act 1999, inserting a new Art 29.4.5° and 6° in the Constitution (subsequently replaced by the Lisbon Treaty amendment). 123 The options or discretions provided by or under Arts 1.11, 2.5 and 2.15 of the Amsterdam Treaty. 124 This Act inserted a new Art 29.4.7°, 8 and 9° in the Constitution (which were subsequently replaced by the Lisbon Treaty amendment). Passed by both Houses of the Oireachtas, it was approved by referendum on 19 October 2002 and signed into law on 7 November 2002. 125 Namely the options or discretions provided by or under Arts 1.6, 1.9, 1.11, 1.12, 1.13 and 2.1 of the Nice Treaty. 126 Passed by both Houses of the Oireachtas, approved by referendum on 2 October 2009 and signed into law on 15 October 2009.

234  Gavin Barrett However, at the same moment, the adoption of certain measures of secondary EU legislation was made subject to this requirement. Article 29.4. 8° provides: The State may agree to the decisions, regulations or other acts i. under the Treaty on European Union and the Treaty on the Functioning of the European Union authorising the Council of the European Union to act other than by unanimity, ii. under those treaties authorising the adoption of the ordinary legislative procedure, and iii. under subparagraph (d) of Article 82.2, the third subparagraph of Article 83.1 and paragraphs 1 and 4 of Article 86 of the Treaty on the Functioning of the European Union, relating to the area of freedom, security and justice, but the agreement to any such decision, regulation or act shall be subject to the prior approval of both Houses of the Oireachtas.

In Iqbal v Minister for Justice,127 the Supreme Court interpreted broadly the licence provided to the Houses of the Oireachtas under these provisions, holding that an opt-in to the adoption of a proposed measure would be invalid only if there had subsequently been ‘such significant departure from the approved text as to warrant the conclusion that the constitutionally necessary prior approval had not been given’.128 In practice, the Article  29.4.7° (iii.) procedure is triggered whenever the Department of Justice, following consultations with the Office of the Attorney General, indicates that an opt-in motion under Title V (pursuant to Ireland’s ‘opt-out with an opt-in’ arrangement at EU level as regards justice and home affairs provisions) should be considered by the Joint Committee on Justice.129 Motions are then – nominally – considered by the full Houses: although normally, they are adopted without any further debate taking place.130 The various Constitutional rules in Article 29.4.7° and 8° requiring ex ante approval by both Houses to Ireland’s taking of various steps at EU level should not be seen in isolation. Statutory provisions similarly seek to exact government accountability regarding European policy-related decisions – although their impact cannot be said to be very great. Section 2(5) of the European Union (Scrutiny) Act 2002 – a statute which formed part of the Government’s response to the defeat of a referendum facilitating the Nice Treaty’s ratification – requires that every minister make a report to each House of the Oireachtas not less than twice yearly regarding measures, proposed measures and other developments concerning the EU in relation to which he or she performs functions.131 Under section 5 of the European Communities Act 1972 (as substituted by section 4 of the same 2002 Act) the Government has been obliged to make a report to each House of the Oireachtas on developments in the EU each year since 2003.132 127 IESC 06.05.2008, Cases 277, 237, and 238/2007, Iqbal v Minister for Justice [2008] 4 IR 362. 128 Ibid, 377. Clear disregard by the Oireachtas of its constitutional duties would thus be needed in order to see its opt-in invalidated by the courts. In Iqbal, approval of the draft Framework Decision on the European Arrest Warrant was held to include any subsequent reasonable and usual drafting changes. 129 On these occasions, briefings were prepared for the Joint Committee by the Library and Research Services Committee team (Houses of the Oireachtas Commission, Annual Report of the 2010, 37). 130 See eg, the motion approved without debate by the Dáil on 14.06.2012 to take part in the adoption and application of the on the freezing and confiscation of instrumentalities and proceeds of crime in the European Union (Directive 2014/42/EU of the European Parliament and of the Council of 3 April 2014, [2014] OJ L 127/39–50), and eg the motion regarding Directive 2013/40/EU of the European Parliament and of the Council of 12 August 2013 on attacks against information systems and replacing Council Framework Decision 2005/222/JHA, [2013] OJ L 218/8–14, adopted in Seanad without debate. 131 In this writer’s experience, the reports’ value is limited. They have tended to be delivered by officials rather than by Ministers; to be received by Committees rather than by the full Houses (limiting their impact); to be largely retrospective and to arrive several months late, reducing their value as anything more than a historical record. 132 The usefulness of this step is questionable. The 2010 Report of the Joint Sub-Committee on the Review of the Role of the Oireachtas in European Affairs expressed the view that ‘the annual report is historical and therefore its purpose is limited’

Ireland  235 At least one ‘soft law’ rule also exists, imposing government accountability to the Dáil in EU matters: namely that deriving from the 2002 Seville Declaration, under which the Government declared that decisions on whether Irish troops should participate in EU tasks would require the ‘triple lock’ of UN endorsement, Government decision and Dáil approval. Again, this commitment formed part of the successful Government strategy to secure the amendment of Article 29.4 to facilitate ratification of the Nice Treaty after initial defeat in referendum. Although the 2011–16 Government’s programme undertook that ‘all Ministers will be obliged to appear before their respective Committees or before the Committee on European Affairs prior to travelling to Brussels for meetings of the Council where decisions are made’, this only ever happened to a very limited extent, then or since.133 Where an ex ante ministerial appearance does occur, a written statement is provided to committee members. The degree of preparation by committee members or indeed interest in close scrutiny of the Government’s negotiating position varies. Inadequate ministerial attendance seems related to the heavy workload of Oireachtas committees in terms of domestic issues and the busy schedules of government ministers. Increased scrutiny of EU affairs therefore suits the interests of neither party. The Government does not report on Council working group or COREPER meetings either to committees or to the plenary session of the Dáil.

B.  Participation of the National Parliament in Secondary Law-making (and in Relation to the Protocol on Subsidiarity) The outcomes of Council meetings are occasionally discussed in the Dáil and the Seanad. When this happens, it may take the form of (a) responses to one of the various forms of parliamentary question or of (b) one of the several kinds of debate, as EU affairs are frequently ventilated by both means. In an attempt to improve the EU-related performance of the Oireachtas, May 2013 saw parliamentary Standing Orders amended to set up a Parliamentary Steering Group on EU Affairs. Re-established in 2017, the Group’s role was to ‘give parliamentary and cohesive direction in European Affairs matters, to enhance the role of the Dáil in EU affairs through engagement with the Government and through Dáil Éireann and its Select Committees’.134 Among the Group’s tasks was to ‘prioritise for debate in the Dáil motions in regard to reports from the Select Committees on EU affairs or on other related EU matters’.135 This was presumably to put an end to the situation in which committee reports on EU-related affairs had languished undebated by either plenary chamber. The effort was subsequently abandoned. The Steering Group disappeared

and went on to recommend that there should no longer be a requirement to prepare an annual report (see para 21 of the report’s conclusions and recommendation 5 of the Sub-Committee made in the Report.). Cf Gavin Barrett, The Evolving Role of National Parliaments in the European Union: Ireland as a Case Study (Manchester, Manchester University Press, 2018) 290. 133 See ibid, 191, 207–10, 231 f and see generally, G Barrett, ‘The Irish Parliament and Brexit’, in T Christiansen and D  Fromage (eds), Brexit and Democracy – The Role of Parliaments in the UK and the European Union (Houndmills, Palgrave Macmillan, 2019). 134 Standing Order 111(1)(a) (now repealed). The Steering Group was chaired by the Ceann Comhairle (Dáil Speaker) and also comprises the Chair of the Working Group of Committee Chairs, the Chair of the Select Committee on European Union Affairs and the Chair and Deputy Chair of other Dáil Select Committees. Order 111 specifically acknowledged the role of national parliaments under the TEU and the TFEU and itself echoed the wording of Art 12 TFEU. 135 Ibid.

236  Gavin Barrett from the Standing Orders with the election of the 33rd Dáil in 2020. Instead, Dáil Standing Order 124(1)(c) now decrees less ambitiously that having regard to the EC Annual Work Programme and the Government priorities for the EU Presidency for the time being as adopted, it shall be a function of the Select Committee on European Union Affairs to present an annual EU work programme for adoption by Dáil Éireann by way of motion which shall set out its own priorities and those of other Committees.

1.  The Role of Committees At the time of writing there were 21 joint Oireachtas Committees, with members from both Houses.136 These committees are reconstituted after each election and can be restructured then (or at other times). The role of each committee is provided for in its Orders of Reference, which are adopted by both chambers. This generally involves shadowing the work of one or more government departments, and thus considering matters of policy, administrative affairs, policy on state bodies, draft heads of bills published by ministers, statutory instruments and strategy statements. Sectoral committees may also consider any EU-related issue that comes within their remit, and they spend varying amounts of time in doing so.137 The outcome of such consideration could be a motion or report to either or both chambers, which may then adopt a resolution although, as is noted above, this is infrequent in practice. An opinion on a particular EU initiative can also be forwarded by a committee directly to the EU institutions as part of the Political Dialogue under the Barroso initiative, although this too is infrequent and only a handful of such opinions have been produced by Oireachtas committees to date. Joint Oireachtas Committee meetings are normally open to the public, although when the need arises it is possible for them to sit in private session. A transcript of all debates is published online within a few days. Committee sessions are web-streamed live when they occur.

2.  The Role of the EU Affairs Committee The Joint Oireachtas EUAC is the only committee to focus exclusively on EU matters. In common with all Oireachtas committees, a newly constituted committee is constituted after each election and it is possible for such committees to be renamed or restructured at that point. The present Committee was established in July 2020. Technically, each Joint Oireachtas Committee is composed of members of a Dáil Select Committee and a Seanad Select Committee. In reality, however, the role of Select Committees is frequently confined to conducting the committee stage of legislation (a rare role for the EUAC as most EU-related legislation is dealt with by sectoral committees). Nine Dáil members out of 166 (5 per cent) and five Seanad members out of 60 (8 per cent) are members of the Joint EUAC. Irish parliamentarians are frequently members of more than one committee. On constitutional grounds, MEPs are not members of the Joint EUAC. However, MEPs elected from Irish 136 (Situation as on 15 March 2022). This does not include committees drawn from one House only such as the Business Committee, the Committee of Public Accounts, the Committee of Selection (Seanad Éireann), Committee on Budgetary Oversight, the Committee on Members’ Interests of Dáil Éireann, the Committee on Members’ Interests of Seanad Éireann, the Committee on Parliamentary Privileges and Oversight (Dáil Éireann), the Committee on Parliamentary Privileges and Oversight (Seanad Éireann) or the Committee on Standing Orders and Dáil Reform. Nor did it include the Sub-Committee on Mental Health and the Working Group of Committee Cathaoirligh (Chairs). A further eight committees no longer exist, having either expired or been renamed. 137 Subject to the normally broadly drafted Orders of the chamber that set them up. CFSP issues fall within the remit of the Joint Oireachtas Committee on Foreign Affairs and Defence.

Ireland  237 constituencies and, at the invitation of the EUAC, other MEPs, may attend its meetings and take part in its proceedings, without any right to vote or to move motions and amendments.138 It is rare for MEPs to avail of this right. The legal basis of the Joint EUAC rests on the respective Orders of the House rather than the Constitution or in statutes. The Joint Committee’s assigned role is to consider: (a) European Commission strategic planning documents including the Commission Work Programme; (b)  cross-sectoral policy developments at EU level; (c) matters listed for consideration on the agenda of meetings of the General Affairs Council and the outcome of such meetings; (d) such ministerial regulations necessitated by EU membership as the Committee may select; (e)  notifications referred by the Dáil of proposals to use the passarelle clauses on qualified majority voting provided for in Article  48.7 TEU; (f) notifications of proposals for the amendment of the treaties received from the European Council pursuant to Article 48.2 TEU; (g) notifications of applications for EU membership received from the European Council pursuant to Article 49 TEU; and (h) such other matters as may be referred to it by the Dáil or Seanad from time to time.139 The Joint EUAC is required to report to both Houses on the operation of the European Union (Scrutiny) Act 2002, the Irish statute governing the scrutiny of EU legislative proposals. It also initiates the yellow card procedure, drafting any relevant reasoned opinion and tabling motions in both plenaries, where the chambers complete the procedure by adopting resolutions approving the motions. The Joint EUAC initiates the response of the Houses of the Oireachtas to the Article  48.7 TEU generalised passerelle procedures and the subsidiarity review procedures. It also has the power to make recommendations to the Minister for Foreign Affairs and Trade (or Minister of State) on EU matters. It represents both chambers at COSAC meetings and reports back. The two Select Committees which provide the members of the Joint EUAC are empowered to consider matters arising from Irish membership of the EU and Irish adherence to the Treaty on European Union (TEU) and the Treaty on the Functioning of the European Union (TFEU), as they wish, which are not referred to any other committee. As already noted, the Joint EUAC is briefed by the Minister of State for European Affairs before or after some General Affairs Councils. Even if it is true that parliamentary committees do not normally play a role regarding European Councils (this being left to Dáil plenary sessions), it is possible for this general rule to be departed from, since the Joint EUAC sets its own agenda. Historically, Joint EUACs have created temporary sub-committees. One such sub-committee invited actors from civil society to speak about the Fiscal Compact and reported on it. Others have examined Ireland’s future in the EU, and the role of the Oireachtas in EU affairs. Perhaps surprisingly, no Brexit sub-committee was ever established.140

3.  The Role of Sectoral Committees In 2011, the system for scrutiny of EU legislative proposals, previously the prerogative of a separate European Scrutiny Committee, was mainstreamed to sectoral committees.141 In practice, 138 See Dáil motion establishing a Joint Committee on European Affairs (995 Dáil Debates No 5 (23 July 2020)) and the similar Seanad motion (271 Seanad Debates No 2 (18 September 2020)). Members of the Irish delegation to the Parliamentary Assembly of the Council of Europe enjoy similar rights in this regard to Irish MEPs. 139 Ibid. 140 That much said, a number of other parliamentary committees did make Brexit-related reports. See generally on the reaction of the Oireachtas to Brexit, G Barrett, ‘Irish Parliament’ (n 129). 141 Note Order 95(8)(d), Dáil Éireann Standing Orders Relative to Public Business 2020.

238  Gavin Barrett however, little time has tended to be spent by the sectoral committees on such scrutiny, presumably because it is of little political value or interest, and sectoral committees already have a large workload related to domestic matters, which tends to be given priority – and must itself compete for the attention of Deputies with more electorally valuable constituency work. The mainstreaming of European affairs to various sectoral committees, if arguably part of a necessary process of maturation of the Oireachtas committee structure, has also arguably defanged the Joint Committee on European Union affairs by preventing it from exercising leadership regarding various substantive areas affected by Brexit.142

4.  The Subsidiarity Early Warning Mechanism and Ex Post Control of Subsidiarity a.  Ex Ante Control The rules implementing the Early Warning Mechanism are found in the European Union Act 2009, and in Dáil and Seanad Standing Orders.143 Section  7(3) of the 2009 Act provides that either House of the Oireachtas may, not later than eight weeks after the transmission of a draft legislative act, send a reasoned opinion to the Presidents of the European Parliament, the Council and the European Commission if the House concerned passes a resolution authorising this. The first reasoned opinion sent by the Oireachtas concerned the proposal for a Common Consolidated Corporate Tax Base.144 Under revised Standing Orders, reasoned opinions are now drafted by the relevant sectoral committee, but in practice reasoned opinions are provided by Joint Oireachtas committees (which are composed of the members of Select Committees of both Houses) rather than by Dáil or Seanad committees alone. Where the committee is of the opinion that a draft legislative act does not comply with the principle of subsidiarity, it is required to submit a reasoned opinion by way of a report which must be laid before the relevant Oireachtas chamber. Where a chamber agrees the motion, the Ceann Comhairle in the case of the Dáil or the Cathaoirleach in the case of the Seanad sends a copy of the resolution, together with a copy of the Committee’s report to the Presidents of the European Parliament, the Council and the European Commission. b.  Ex Post Control Court action in order to enforce the principle of subsidiarity is envisaged under section 7(4) of the European Union Act 2009: where either House of the Oireachtas is of the opinion that an act of an institution of the European Union infringes the principle of subsidiarity provided for in the treaties governing the European Union and wishes that proceedings seeking a review of the act concerned be brought in the Court of Justice of the European Union … it shall so notify the Minister in writing for the purposes of Article 8 of Protocol No 2 to that treaty and the Treaty on European Union and the Minister shall, as soon as may be after being so notified, arrange for such proceedings to be brought.

142 The Joint EUAC has apparently seen a reduction in its relative prestige: thus for example, political parties showed relatively little interest in providing a chair for it in July 2016, under the d’Hondt system (Oireachtas official interviewed 29 June 2018). 143 See Dáil Standing Orders 133 and 95(8)(a) and Seanad Standing Order 116 and 71(4)(a). 144 COM (2016) 685: Proposal for a Council Directive on a Common Corporate Tax Base

Ireland  239 Either chamber can empower a select committee to consider whether any Act of an EU institution infringes the subsidiarity principle. If it feels this is so and that proceedings should be brought, the committee must lay a report before the relevant plenary. If the motion is approved by a simple majority in the relevant chamber, the Ceann Comhairle or Cathaoirleach must send a copy of the resolution to the relevant minister.145 The Oireachtas has never yet threatened to bring an EU legislative act to the European Court of Justice on the grounds of concerns regarding subsidiarity.

5.  The Activities of the Irish Parliament at the EU Level: The Political Dialogue No formal rules exist concerning the Political Dialogue with the European Commission. Commission records of submissions,146 looked at in conjunction with the records of submissions on committee websites, indicate that the degree of Irish engagement remains low compared with many other EU Member States. The mainstreaming process has led to sectoral committees being expected to bear more of the load regarding the Political Dialogue, and in practice, all submissions made in recent years under the Political Dialogue process have been made by sectoral Joint Oireachtas Committees.

6.  The Oireachtas Permanent Representative in Brussels Ireland has had a permanent representative in Brussels since its 2004 Presidency of the Council. Much of what the permanent representative does consists of monitoring and reporting, observing what is going on in the European Parliament and reporting back to the Oireachtas regarding legislative developments. The representative also reports on other parliaments’ thinking on legislation and the yellow cards and orange cards, and liaises with other national parliaments to promote common understanding.

7.  European Council and Eurozone Meetings Since 2011, the Taoiseach has briefed the Oireachtas prior to attending European Council meetings.147 Such ex ante briefings are scripted and generally heard without interruption. The Taoiseach normally outlines the agenda for the forthcoming European Council and gives a very broad statement on the Government position. Opposition parties cannot put questions, but can follow it with their own statements. After the opposition party leaders have spoken, sometimes with interruptions, the Taoiseach makes some concluding remarks. The Minister for Foreign Affairs may perform this task if the Taoiseach is unable to. Ex-post control of European Councils is a lengthier affair. A longer scripted statement than his pre-summit statement is made by the Taoiseach, and this is provided to opposition leaders in advance in order to facilitate debate. This consists of a summary of the outcome of the European Council, framed to show the Government in the best possible light, and is generally

145 See Dáil Standing Order 135 and Seanad Standing Order 118. 146 Available at https://ec.europa.eu/dgs/secretariat_general/relations/relations_other/npo/index_en.htm. According to this, in the last five years in respect of which records have been published, the Houses of the Oireachtas have made 6 submissions (2017), 8 submissions (2018), 0 submissions (2019), 1 submission (2020) and 3 submissions (2021). 147 See Dáil Standing Order 124(1)(b), which does not however make the practice of ex ante briefings compulsory.

240  Gavin Barrett heard without interruption. It is followed by statements by opposition leaders, and/or the relevant party’s European affairs spokesperson as well as by independent Deputies, all usually framed to challenge or question the achievements claimed by the Government. Statements are followed by 20 minutes of questions. Exchanges can be lively, but the debates tend to be rather sparsely attended. There is no mechanism for controlling the result of Eurozone summits. However, because these coincide with European Councils, in practice these are also controlled by the plenary session of the Dáil – at least when the outcome is sufficiently politically interesting. It can happen that the main focus of both statements and questions is the Eurozone summit rather than the European Council. Various mechanisms are used by the opposition to hold the Government to account. Parliamentary questions of various kinds are used to raise all manner of EU-related issues. The limits of such an approach are significant, however, given that there is no system to compel a satisfactory answer from a member of the Government who is unwilling to provide one. Committees, including the Joint EUAC, tend to be consensus-driven, limiting the extent to which ministers are held to account politically rather than simply being required to state what happened at a Council meeting.

C.  Techniques for Implementing Secondary Legislation Most EU directives are transposed into Irish law using ministerial regulations, although primary legislation is occasionally enacted.148 In the sample year of 2015, the Oireachtas adopted 21 public statutes (32  per  cent of the total) significantly connected with Ireland’s EU membership, and 379 similarly connected statutory instruments (59 per cent of the total) were also adopted. The EU-related proportion of statutory instruments was thus considerably higher than the EU-related proportion of statutes. Qualitatively, the impact seems to be greater too: the primary objective of large numbers of statutory instruments was implementing EU law. There is constitutional case law here. From about 1994, the focus in Irish jurisprudence concerning the constitutional immunity clauses found in Article 29.4.6°, began to focus on the issue of the form of domestic law instrument used to implement EU law measures, and whether this choice of form itself could be immunised. Meagher v Minister for Agriculture149 involved a challenge both to (i) section 3 of the European Communities Act 1972 (which permitted statutory instruments to alter primary legislation so as to implement Community (now EU) law) and (ii) certain statutory instruments which had been adopted thereunder.150 In the High Court, Johnson J held that conferring on a minister a power such as that given in section 3(2) to make regulations for the purposes of amending or repealing primary legislation violated Article 15 of the Constitution. This effectively outlawed the primary means used to implement EC law in Ireland. On appeal the Supreme Court reversed this finding of unconstitutionality. According to the Court (the single judgment of which in this regard was delivered by Finlay CJ): the power of regulation-making … contained in s.3 is prima facie a power which is part of the necessary machinery which became a duty of the State upon its joining the Community and therefore necessitated by that membership. 148 See eg the Counterfeiting Act 2021, enacted to give effect to Directive 2014/62/EU of the European Parliament and of the Council of 15 May 2014 on the protection of the euro and other currencies against counterfeiting by criminal law. 149 IESC, Meagher v Minister for Agriculture (n 63). 150 The statutory instruments created a power of search allegedly not required by the EC instrument they implemented and altered a statutorily set limitation period for prosecution of offences.

Ireland  241 The Court is satisfied that, having regard to the number of Community laws, acts done and measures adopted which either have to be facilitated in their direct application to the law of the State or have to be implemented by appropriate action into the law of the State, the obligation of membership would necessitate facilitating of these activities, in some instances, at least, and possibly in a great majority of instances, by the making of ministerial regulation rather than legislation of the Oireachtas.151

This part of Meagher is significant as an authority that the ‘necessitated’ test does not involve a mere renvoi to the substance of EU law without more. As far as the Supreme Court was concerned, the obligations of membership included factors that went beyond the substantive obligations of European law: questions of practical possibility entered the equation, and the question of ‘necessitated’ was also related to the means by which European-level regulations were implemented. Regarding the separate issue of whether the statutory instruments at issue in this case were valid, the Supreme Court delivered two judgments both of which, although in some respects contradictory, were agreed to by all three remaining Supreme Court judges. The result is that each may be taken to represent the opinion of four out of the five judges of the court. The most interesting aspect of Blayney J’s judgment was his observation that, in implementing European directives, the state was obliged to choose a method of implementation and, provided the method it chooses is appropriate for the purpose of satisfying the obligation of the State and the measures it incorporates do not go beyond what is required to implement the directive, it is correctly categorised as being necessitated by the directive. In any instance where the method was not appropriate, or its measures went beyond what was required, it would not be necessitated and would be open to constitutional challenge.152

Thus for Blayney J the question whether the use of a statutory instrument was necessitated depended on its appropriateness.153 According to Denham J, too, ‘the term “necessitated” is relevant to the choice of method’ for implementing European-level directives.154 However, Denham J’s test for the validity of statutory instruments differed from that of Blayney J According to Denham J, in choosing how to implement directives, ‘the Minister must balance the relevant Articles of the Constitution, which in this case are Article 15, s.2 and [the “necessitated” clause]’.155 Applying the same test used to determine whether delegated legislation was permissible in an entirely domestic law scenario,156 Denham J reasoned that, the test is whether the ministerial regulations under s.3 of the Act of 1972 are more than the mere giving effect to principles and policies of the said Act and the directives which are part of domestic law as to the result to be achieved. If the regulations contained material exceeding the policies and principles of the directives then they are not authorised by the directives and would not be valid under s.3 unless the material was incidental, supplementary or consequential.157

Meagher was followed by the High and Supreme Court rulings in Maher v Minister for Agriculture and Food.158 At issue here was a ‘use it or lose it’ rule that had been introduced regarding milk quotas by a statutory instrument implementing Community regulations – a statutory instrument now claimed by the applicants to have violated the Constitution, its use not having been

151 IESC,

Meagher v Minister for Agriculture (n 63) 352. 360 (emphasis added). 153 Sterck, Constitutional Identity (n 91) has described this as equating ‘necessitated’ with ‘appropriate’. 154 IESC, Meagher v Minister for Agriculture (n 63) 365. 155 Ibid, 364. 156 Namely, the test developed in IESC, Cityview Press v An Chomhairle Oiliúna [1980] IR 381. 157 IESC, Meagher v Minister for Agriculture (n 63) 366. 158 IESC, Maher v Minister for Agriculture and Food (n 54) 147. 152 Ibid,

242  Gavin Barrett necessitated. The applicants’ challenge to the regulations was rejected by Carroll J in the High Court. On appeal, the Supreme Court adopted a different approach to that taken by it in Meagher. Keane CJ mapped out the general approach taken in all of the judgments. In addressing the issue of whether proceeding by way of a statutory instrument ‘was in conflict with the exclusive legislative role of the Oireachtas under Article [15.2.1°] and was not necessitated by the obligations of membership’,159 Keane CJ observed there were two routes by which a conclusion could be reached on this issue: one can initially decide whether the making of the regulation in the form of a statutory instrument rather than an Act of the Oireachtas was ‘necessitated’ by the obligations of membership. If it was, then it is clearly unnecessary to consider whether it is in conflict with Article 15.2 or, for that matter the Articles guaranteeing the private property rights of the applicants. Alternatively, one can determine first whether it violates either Article 15.1 or the private property rights or both of them. If the latter course were adopted, and the conclusion were reached that no breach of the Constitution had been established, it would be unnecessary to consider whether enactment in the form of a regulation rather than by an Act was necessitated by the obligations of membership.160

Although the second of the two routes seems the more intellectually arduous of the two, it was the route taken by all of the Supreme Court judges in Maher. Hogan and Whyte cite Maher as authority for the proposition that the method of transposition of Community (now Union) legislation could never be ‘necessitated’ for the purposes of what is now Article 29.4.6°.161 The shift in Maher away from the ‘necessitated’ clause (as compared to the ruling in Meagher) and towards a reading of Article 15.2.1° in apparent isolation from the former provision is of interest, since it means that some of the constitutional burden of facilitating Ireland’s membership of the EU is now no longer carried by amendments to Article 29.4 specifically designed to facilitate that membership, but rather is carried exclusively by a modified construction of Article 15.2.1°, which applies, even where the statutory instrument at issue is not regarded as necessitated. Building on the views of the court in Meagher v Minister for Agriculture, and in particular on the ruling of Denham J in that case, Fennelly J identified the test of whether recourse to a statutory instrument would be precluded as being ‘whether the scope of the discretion conferred by Community law in regulations which become part of national law was so independent of principles and policies laid down by those Community regulations, as to place the State in conflict with Article 15.2.1° of the Constitution’.162 The subsequent High Court ruling in Sam McCauley Chemists v The Pharmaceutical Society of Ireland163 showed a welcome continued judicial resistance to the idea of excessively restricting executive freedom to implement EU law with an overly narrow approach to the principles and policies test. Unlike the Supreme Court in Maher, McCracken J (ultimately upheld on appeal by the Supreme Court164) viewed the ‘necessitated’ clause as being necessary to his decision in this case (which was to uphold the use of the 1991 statutory instrument to amend statute law).165 He specifically referred to the rulings of both Blayney J and Denham J in Meagher.166 Thus the

159 Ibid, 181. 160 Ibid. 161 Hogan and Whyte, Kelly (n 10) 527. 162 IESC, Maher v Minister for Agriculture and Food (n 54) 254. 163 IEHC, Sam McCauley Chemists v The Pharmaceutical Society of Ireland [2008] 1 IR 16. 164 The plaintiffs’ appeal was dismissed and the order of the High Court affirmed by order of the Supreme Court (Murray CJ, Denham, Hardiman, Fennelly and Lavan JJ) dated October 13, 2006. 165 IEHC, Sam McCauley Chemists (n 159) 24. 166 Ibid, 21 f.

Ireland  243 irrelevance of the ‘necessitated’ clause to methods of transposition of EU law had clearly not yet gained universal judicial acceptance. A subsequent important footnote to the landmark decisions in Meagher and Maher, at least insofar as concerns the ‘necessitated’ issue, was provided by the Supreme Court in Browne v Attorney General.167 Keane CJ’s judgment here seemed to indicate that whether or not the substance of a measure is necessitated, full compliance with all Irish constitutional procedural rules will be required – a conclusion that was not necessarily clear from the Maher ruling (in particular Fennelly J’s judgment in that case, which stressed the non-necessitated nature of the substance of the statutory instrument at issue there168).

D.  Enforcement through the Courts including the Charter of Fundamental Rights In general, the Irish courts have adopted a markedly integration-friendly approach to EU law. A caveat to this is that they can however be strict in looking at laws which are not fully in force, particularly unratified Treaties, in which (as seen in Crotty) constitutional provisions have been interpreted with extraordinary strictness. One prominent example of the generally positive attitude to European law – perhaps its apotheosis – was the ruling by Costello J in his High Court judgment in Pigs and Bacon Commission v McCarron169 that Community law ‘takes effect in the Irish legal system in the manner in which it itself provides’. This effectively ushered the doctrines of supremacy and direct effect painlessly into the Irish legal system. The decision of the Irish Supreme Court in Campus Oil v Minister for Industry and Energy170 was another high point. This decided that Article 177 EEC (now 267 TFEU) precluded any appeal to the Supreme Court against a preliminary reference made by the High Court, a perspective, which in the light of the less demanding perspective of the European Court of Justice at this time concerning Article 177, was described by one commentator as being ‘plus catholique que le pape’.171 Another example of a facilitative approach to EU law was seen in Meagher v Minister for Agriculture,172 in which a broad approach was taken to the notion of measures ‘necessitated’ by the obligations of membership and therefore constitutionally immune. Another somewhat random example has also been their positive attitude to the interpretive obligation under EU law. Thus eg, in Nathan v Bailey Gibson, the Irish Supreme Court happily interpreted Irish employment law broadly – more specifically, interpreting less favourable treatment of a worker ‘by reason of his sex’ as including indirect discrimination – in order to comply with the obligations of the Irish legal system under the doctrine of indirect effect.173

167 IESC 16.07.2003, Case 139/2002, Browne v Attorney General [2003] 3 IR 205. 168 IESC, Maher v Minister for Agriculture and Food (n 54) at 253, 254 f. 169 IEHC, 30.06.1978, 1978 WJSC-HC 3332, Pigs and Bacon Commission v McCarron (1978) JISEL 77. 170 IESC, Campus Oil (n 57). 171 O’Keeffe, ‘Preliminary Deference’ (n 59). See for an example of similar zeal, the interesting case IESC 18.02.2010, Case 91/2008, Abbeydrive Developments Ltd. v Kildare County Council (No 2) [2010] IESC 8; [2010] 2 IR 397, in which the Court reopened a case in which it had just delivered judgment and was deciding only the issue of costs in order to correct an error of European Union law it had made (Garrett Simons, ‘Is There a Reluctance on the Part of Irish Courts to Accept the Primacy of EU Law?’ unpublished paper delivered at the conference The Irish Constitution: Past, Present and Future, Dublin, 28–29 June 2012). 172 IESC, Meagher v Minister for Agriculture (n 63). 173 Discussed in Simons, ‘Reluctance’ (n 167).

244  Gavin Barrett This largely cooperative approach has not been without occasional exceptions, where highly sensitive issues have been raised in Irish courts. Simons has pointed out a small number of cases in which Irish superior courts have shown a surprising, if occasional reluctance (i) to apply the doctrine of direct effect in the absence of the proper transposition of EU law by the Irish legislature,174 (ii) to apply the doctrine of indirect effect sufficiently rigorously175 or (iii) to t­ olerate the disapplication by a court lower than the High Court of a national law in violation of a rule of EU law.176 By 2020 (the latest year in respect of which statistics were available at the time of writing), 130 preliminary references had been made by Irish courts under what is now Article 267 TFEU from the time of Irish entry to the EEC in 1973.177 Notwithstanding the fact that there has been a sizeable increase in the number of references in recent years, this is still a relatively low number compared to similarly-sized countries.178 The courts of Denmark, which also joined in 1973, have made 202 references – over one-and-a-half times as many as the Irish courts. Finland, which joined the EU in 1995, has seen its courts make 135 references (slightly more than Ireland in about half the period of membership). The low Irish total may have reflected judicial concerns about adding costs and delays on litigants beyond those imposed by an already slow and expensive judicial system. The number of references has increased since 2010, however, to an all-time high of 12 references in 2017, a number reached again in 2018. This may possibly have been related to the entry into force of the Lisbon Treaty, which broadened the jurisdiction of the Court in justice and home affairs, Ireland having not accepted the Court’s jurisdiction under the old Article 35 EU before this.179 Since then, the number of references has reduced, only five references were made in 2020.

VI.  The Resulting Relationship between European Law and National Law The issues of direct effect and supremacy are not expressly provided for by the provisions of the Constitution. Section 2 of the European Communities Act 1972 provides that from 1 January 1973, ‘the treaties governing the European Communities and the existing and future acts adopted by the institutions of those Communities shall be binding on the State and shall be part of the domestic law thereof under the conditions laid down in those treaties’. Costello J in Pigs and Bacon Commission v McCarron180 viewed this provision as importing the doctrines of supremacy 174 IEHC, 31.07.1997, Case 201/96, McBride v Galway Corporation [1998] 1 IR 485 (High Court decision of Quirke J); IEHC, 21.12.2010, Case 2010 756 JR, Lackagh v Galway City Council [2010] IEHC 479 (High Court decision of Irvine J) 175 IESC 02.05.2008, Case 1323/2005 JR, Harding v Cork County Council [2008] 2 ILRM 251 (although this, in fairness, concerned only one member of the Court – Kearns J – who was nonetheless, the only member of the Court to deal with the European Union law directive at all.) 176 IEHC 17.02.2009, Case 2008 793 JR, Minister of Justice, Equality and Law Reform v Equality Tribunal [2009] IEHC 72 (High Court decision of Charleton J); IEHC 04.06.2010, Case 2008 183, An Taoiseach v Information Commissioner [2010] IEHC 241; [2011] 1 ILRM 508. 177 41 of these had been by the Supreme Court, 47 by the High Court and 42 by other courts or tribunals (CJEU, Annual Report: Judicial Activity 2020, 229–230). 178 In 10 of the 48 recorded years of references from Ireland, no references were made at all (the last such year being 2009). 179 See Elaine Fahey, Practice and Procedure in Preliminary References to Europe: 30 Years of Article 234 EC Case Law from the Irish Courts (Firstlaw, Dublin, 2007) and by the same author, ‘Being a Third Pillar Guardian of Fundamental Rights? The Irish Supreme Court and the European Arrest Warrant’, 33 European Law Reporter (2008) 563. 180 IESC, Pigs and Bacon (n 165).

Ireland  245 and direct effect painlessly into the Irish legal system, even if on a different logical basis to that set out in Costa v ENEL and Van Gend en Loos.181 Although the provision of a statute may well regulate other statutes or statutory instruments, it cannot determine the relationship between the Irish Constitution and EU law. This may be regarded as determined by the Article 29.4.6° immunity clause which did not embed the doctrine of the supremacy of EU law in the Irish Constitution. Rather than resolve the issue of which legal order would prevail in the event of a collision between the national and the supranational, it simply made sure that, in so far as national constitutional norms were concerned, such collisions would not take place in the first place. An important postscript on a discussion on supremacy in Irish law is that under the so-called Maastricht Protocol (now reproduced word-for-word as Lisbon Protocol (No 35) on Article 40.3.3 of the Irish Constitution), and agreed so as to avoid difficulties arising concerning in the referendum facilitating ratification of the Maastricht Treaty, it is provided that ‘nothing in the Treaties, or in the Treaty establishing the European Atomic Energy Community, or in the Treaties or Acts modifying or supplementing those Treaties, shall affect the application in Ireland of Article 40.3.3 of the Constitution of Ireland.’182 This commitment was later expanded upon in order to ensure victory in the (second) 2009 referendum on the Lisbon Treaty. Thus, in fulfilment of a commitment made by the member states at the June 2009 European Council, Article 1 of the Protocol on the concerns of the Irish people on the Treaty of Lisbon signed at Brussels on 13 June 2012 provided that nothing in the Treaty of Lisbon attributing legal status to the Charter of Fundamental Rights of the European Union, or in the provisions of that Treaty in the area of Freedom, Security and Justice affects in any way the scope and applicability of the protection of the right to life in Article 40.3.1, 40.3.2 and 40.3.3, the protection of the family in Article 41 and the protection of the rights in respect of education in Articles 42 and 44.2.4 and 44.2.5 provided by the Constitution of Ireland. 183

Article 40.3.3 was replaced as a result on the entry into force of the Thirty-sixth Amendment of the Constitution Act 2018 (approved in referendum in May 2018) with a text providing that ‘provision may be made by law for the regulation of termination of pregnancy.’ Effectively, this made Lisbon Protocol (No 35) superfluous. However, both it and the later 2012 Protocol remain in force, exempting the above-mentioned constitutional provisions either from the impact of the Treaties or certain provisions of the Lisbon Treaty as the case may be.

References G Barrett, ‘Building a Swiss Chalet in an Irish Legal Landscape? Referendums on European Union Treaties in Ireland and the Impact of Supreme Court Jurisprudence’, (2009) 5 European Constitutional Law Review 32. G Barrett, A Road Less Travelled. Reflections on the Supreme Court Rulings in Crotty, Coughlan and McKenna (No 2) (Dublin, IIEA, 2011).

181 ECJ 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66; ECJ 07.03.1985 32/84 (Van Gend & Loos v Inspecteur der Invoerrechten en Accijnzen) ECLI:EU:C:1985:104. See Cahill, ‘Constitutional Exclusion’ (n 90) 76, n 14 for a series of arguments that the terms of the relationship between Irish and EU law ought not to be regarded as authoritatively set at statutory level, in particular by s 2 of the 1973 Act, but rather should be regarded as constitutionally determined. 182 At the time, Art 40.3.3, inserted in the Irish Constitution by the Eighth Amendment of the Constitution Act, 1983 acknowledged the right to life of the unborn, with due regard to the equal right to life of the mother. 183 See OJ L 060 (2.03.2013), pp. 129–139.

246  Gavin Barrett G Barrett, The Oireachtas and the European Union (Dublin, Houses of the Oireachtas, 2012). G Barrett, ‘The Evolving Door to Europe: Reflections on an Eventful Forty Years for Article 29.4 of the Irish Constitution’, (2012) 48 Irish Jurist 132. G Barrett, The Evolving Role of National Parliaments in the European Union: Ireland as a Case Study (Manchester, Manchester University Press, 2018). G Barrett, ‘The Irish Parliament and Brexit’, in T Christiansen and D Fromage (eds), Brexit and Democracy – The Role of Parliaments in the UK and the European Union (Houndmills, Palgrave Macmillan, 2019). L Besselink, M Claes, Š Imamović, JH Reestman, National constitutional avenues for further EU integration (Luxembourg, Publications Office of the European Union, 2014). M Cahill, ‘Constitutional Exclusion Clauses, Article 29.4.6 and the Constitutional Reception of European Law’, (2011) 34 Dublin University Law Journal 74. J Casey, Constitutional Law in Ireland (London, Sweet & Maxwell, 1987). F de Londras and D Gwynn Morgan, ‘Constitutional Amendment in Ireland’, in X Contiades, Engineering Constitutional Change. A Comparative Perspective on Europe, Canada and the USA (Abingdon, Routledge, 2013) 179. B Doherty, ‘Land, Milk and Freedom – Implementing Community Law in Ireland’, (2004) 11 Irish Journal of European Law 141. E Fahey, Practice and Procedure in Preliminary References to Europe: 30 Years of Article 234 EC Case Law from the Irish Courts (Dublin, Firstlaw, 2007). E Fahey, ‘Being a Third Pillar Guardian of Fundamental Rights? The Irish Supreme Court and the European Arrest Warrant’, (2008) 33 European Law Reporter 563. D Fennelly, ‘Crotty’s Long Shadow: the European Union, the United Nations and the Changing Framework of Ireland’s International Relations’, in E Carolan (ed), The Constitution of Ireland. Perspectives and Prospects (London, Bloomsbury, 2012) ch 33. P Gallagher, ‘The Constitution and the Community’, (1993) 2 Irish Journal of European Law 129. S Henchy, ‘The Irish Constitution and the EEC’, (1977) 1 Dublin University Law Journal 20. G Hogan, ‘The Supreme Court and the Single European Act’, (1987) 22 Irish Jurist (ns) 55. G Hogan and A Whelan, Ireland and the European Union: Constitutional and Statutory Texts and Commentary (London, Sweet and Maxwell, 1995). JM Kelly, The Irish Constitution, reproduced in the G. Hogan and G. Whyte, JM Kelly: The Irish Constitution, 4th edn (Haywards Heath, Tottel, 2006). B McMahon and F Murphy, European Community Law in Ireland (Dublin, Butterworth, 1989). S McNeice, ‘Government approves proposed abortion legislation including “safe zones” around hospitals’, Newstalk, 10.07.2018. D O’Keeffe, ‘Preliminary Deference: The Supreme Court and Community Law’, (1983) 5 Dublin University Law Journal 286. D Phelan, Revolt or Revolution. The Constitutional Boundaries of the European Community (Dublin, Round Hall, 1997). G Simons, ‘Is There a Reluctance on the Part of Irish Courts to Accept the Primacy of EU Law?’ unpublished paper delivered at the conference The Irish Constitution: Past, Present and Future, Dublin, 28–29 June 2012. J Sterck, ‘The Constitutional Identity of Member States and the Primacy of European Union Law. A Comparative Study of Ireland and France’ (unpublished PhD thesis, University College Dublin, 2013). J Temple Lang, ‘The Widening Scope of Constitutional Law’, in D Curtin and D O’Keeffe (eds), Constitutional Adjudication in European Community and National Law (Dublin, Butterworth, 1992) 229. A Whelan and L Heffernan, ‘Ireland the United Nations and the Gulf Conflict: Legal Aspects’, (1991) Irish Studies in International Affairs 115.

8 Greece LINA PAPADOPOULOU

I.  Main Characteristics of the National Constitutional System A.  Brief History and Description of the Hellenic Constitution The Modern Greek state – officially the Hellenic Republic – was established under the London Protocol of 1830, following the war of independence from the Ottoman Empire.1 The first Hellenic constitutions were adopted by revolutionary National Assemblies: the first as early as in 1822 in Epidaurus and in 1823 the second one proclaimed in Astros, being more concise and technically succinct. The Third National Assembly (Troizena, 1827) endorsed the ‘Political Constitution of Greece’, the most important constitution of the revolutionary period, in which the principle of popular sovereignty was proclaimed. All three revolutionary constitutions owned a highly democratic and liberal character, compared to their contemporary counterparts. After a period of absolute monarchy (1832–43) under King Otto, a popular uprising led to a new Constitution (1844). The constitutional monarchy was abolished in 1864, when the next constitution introduced the system of crowned democracy (1864–1909). Although a kingdom for the greatest part of its history, Greece ultimately became a republic after the referendum of 7 December 1974 (69.18 per cent voted against retaining monarchy), which took place after the fall of a seven-year dictatorship (24 July 1974).2 The 1975 Constitution (a rather prolix d ­ ocument of 23,846 words) has since been amended four times: in 1985 (limiting the President’s ­competences), in 2001 (enhancing human rights, through insertion of alternative military service for conscientious objectors, the right to information and participation to the information society, protection of health, genetic identity, and personal data, as well as transparency and enhancement of independent authorities, in 2008 (with minor politically insignificant changes) and in 2019 (mainly election of the President and citizens’ initiative). This Constitution of 1975/1981/2001/2008/2019 is now in force (hereinafter GC). Under this Constitution, Greece is a unitary, parliamentary mono-cameral democracy, headed by the President of the Republic who is elected by Parliament with a qualified majority,

1 For a concise historical account, see Philippos Spyropoulos and Theodore Fortsakis, Constitutional Law in Greece (Alphen aan den Rijn, Kluwer Law International, 2009) 41 ff. 2 For a more nuanced analysis see George Katrougalos, The Constitutional History of Greece, in the Balkan Context, http://www.cecl.gr/RigasNetwork/databank/reports/r1/GR_1_Katrougalos.htm.

248  Lina Papadopoulou whereas real political power is vested with the Prime Minister. With 300 parliamentary seats and an electoral system of reinforced proportional representation, the political system was polarised until 2011, with each Government being supported by one party only. Since 2011, due to the collapse of the political system caused by Greece’s sovereign debt crisis, coalition governments have been formed. The Constitution provides the legal basis for the country to be part of the international legal order.3 The Hellenic Republic is a founding member of the United Nations, and among others a member of the Council of Europe (CoE), NATO, OECD, OSCE, the WTO. It became the 10th member of the European Union (EU) on 1 January 1981 and of the eurozone in 2001.

B.  Division and Balance of Powers 1.  The Judiciary Justice is rendered by civil, criminal, and administrative courts (Article 93§1 GC). Most of these are ordinary courts, adjudicating all sorts of cases except those few which fall under the jurisdiction of the special courts. Accordingly, following a French paradigm, there is a separate system of administrative courts, while penal and civil law cases are adjudicated by the same judges. The superior courts consist of (a) the Areios Pagos (henceforth AP), or Court of Cassation, which is the supreme civil and penal court, (b) the Symvoulio tis Epikrateias (Council of State, henceforth CoS, Article 95 GC), which is the supreme administrative court4 and (c) the Court of Audit (henceforth CoA, Article 98 GC), which functions both as a court and the major auditing authority of the administration. There is also a ‘Supreme Special Court’ (Article 100 GC) with very specific competences and a Special Court with the competence to try alleged criminal acts committed by members of the Government and the President of the Republic.5 The Constitution also provides for extraordinary courts in a state of emergency (Article 48 GC).6 Unlike many European states, Greece has no Constitutional Court. Judicial review of constitutionality is diffuse, meaning that it is exercised by all Greek courts, in all jurisdictions and instances, the only exception being the chambers of the three highest courts (CoS, AP and CoA), which are obliged (Article 100§5 GC) to refer the matter to the Plenum in cases where statutory provisions are found to be unconstitutional. As in Ireland, there is no specific procedure to bring such cases to the court (cf amparo in Spain or Verfassungsbeschwerde in Germany), but the unconstitutionality of legislation relevant in a case can be questioned in any procedure by a litigant with locus standi. Thus constitutional review is also incidental. Moreover, courts judge – at least as an ideal type – in concreto, not in abstracto. Formally, judgments have only an inter partes effect. The constitutional question is posed in the course of

3 Art 2 GC stipulates that ‘Greece, adhering to the generally recognised rules of international law, pursues the ­strengthening of peace and of justice, and the fostering of friendly relations between peoples and States’. 4 See Feloktemon Arnaoutoglou, ‘Grèce: Le Conseil d’Etat’ in J Iliopoulos-Strangas (ed), Cours suprêmes nationales et cours européennes: concurrence ou collaboration? (Athens, Sakkoulas, 2007) 179. 5 Also of a specialised nature are the juvenile courts (Art 96 §3 GC), the military, naval and air-force courts and the prize courts (Art 96§4 GC). There is also a special court for suits against judicial functionaries for ‘faulty wrongful judgment’ (Art 99 GC), which – in an enhanced composition – also functions as a judges’ salary court. The latter, like some of the other special courts, is staffed not only by judges but also by other functionaries (eg law professors). 6 See Spyropoulos and Fortsakis, Constitutional Law (n 1) 194 ff.

Greece  249 ordinary litigation, since there is no legal means of targeting specifically and directly the unconstitutionality of a norm. Last but not least, when courts declare laws unconstitutional, they set them aside and do not apply them in the case before them, but have no competence to annul them (Article 93§47 in combination with Article 100§1e GC). Notwithstanding this diffused, incidental, concrete, and declaratory character of the ­constitutionality review,8 there are procedures that lead to a concentration of constitutional control in the highest courts, especially the Council of State.9 This is a mainly judicial organ with the competence to adjudicate on applications for the annulment of individual or regulatory administrative acts, on petitions for the reversal of judgments issued by administrative courts in substantive disputes and appeals in annulment disputes decided by ordinary appeal courts, as well as other cases specified by the law.10 The Council of State also holds the competence to constitutionally review ex ante normative presidential decrees. Yet even this Court can only declare the unconstitutionality (or unconventionality, ie violation of a treaty) of a norm, and not annul it. Its judgments, however, have a wider effect than this rule suggests, especially when the Council of State annuls an administrative Act of general content which is deemed to be based on an unconstitutional statutory provision. The legal means of ‘application for annulment’ of an administrative act is thus seen – and rightly so – as a concentration mechanism in a diffused system of constitutionality review. In practice, the statutory provision found to be unconstitutional is no longer applicable, although not formally abolished. Thus, the ‘application of annulment’ introduces a deviation from the diffused system of constitutional review and seems closer to a concentrated one. In Decision 1125/2016 (plenary session) the Council of State ruled that the administration, in order to comply with a decision delivered by itself, is obliged not only to consider as invalid and non-existent to the legal world a statutory provision that has been found to be unconstitutional or an administrative act that has been annulled, but also to take the appropriate measures in order to reform the legal situation arisen from these same acts (recital 15). The gradual concentration of constitutional control in the Council of State was further enhanced by the institution of ‘pilot trials’, introduced in the Greek legal order with Article 1 L 3900/2010. According to it, any legal remedies or appeal pending before an ordinary administrative court can be brought before the Council of State when it concerns a matter of general interest having consequences for a wider range of persons. In the ‘pilot trial’ the constitutional review is neither incidental nor concrete, since the Court is seised with a case, in which the main issue is the constitutionality of a statutory provision concerning the interests of a group of persons. However, the rule is that courts finding a statutory norm to be unconstitutional only leave it aside in concreto, but the norm itself remains formally valid. There is an exception to this rule: a norm may be abolished as unconstitutional only by the ‘Supreme Special Court’ (Anotato Eidiko Dikastirio). The case may arrive before that Court only under very strict conditions, specifically contradictory judgments on the (un-)constitutionality

7 ‘The courts shall be bound not to apply a statute whose content is contrary to the Constitution’. 8 Cf Vassilios Skouris, ‘Constitutional Disputes and Judicial Review in Greece’, in C Landfried (ed), Constitutional Review and Legislation (Baden-Baden, Nomos, 1988) 177. 9 Cf Arnaoutoglou, ‘Grèce’ (n 4) 179 ff. 10 Spyropoulos and Fortsakis, Constitutional Law (n 1) 197.

250  Lina Papadopoulou of a specific norm issued by two of the supreme courts (CoS, AP, CoA). Consequently, and ­exceptionally, the Supreme Special Court’s judgments are the only ones that have an erga omnes effect and this is why they are published in the Government Gazette (GG), in the same way as regulatory acts.

2.  Balance of Powers between the Courts and the Parliament The Parliament in such a system seems to have more power than parliaments in systems with a Constitutional Court, since the courts – formally speaking – cannot abolish statutes voted upon by the democratically elected representatives of the people. If judged on the basis of Mark Tushnet’s distinction between weak and strong forms of constitutional review,11 the Greek system can be seen as a mixed one. This is due to the fact that, although no court except the Supreme Special Court (Article 100 GC) is formally competent to annul a law, the jurisprudence of the highest courts, especially the case law of the Council of State, has the power to bring about factual general misapplication of a statute. In practice, however, when a High Court, especially the Council of State, finds in a specific case that a stipulation violates a constitutional norm, Parliament feels politically obliged to change the law and the administration not to implement it. In the opposite case, if the statutory provision is continually and often used, citizens usually resort to the courts in massive numbers in order to have the provision declared unconstitutional in their own individual cases. This is an element of irrationality since the lower courts would very rarely adopt a different position than that expressed in the case law of the higher courts. Neither the Parliament nor the Government is in a position to modify the consequences of a court decision, given the fact that the latter has inter partes effect, with the notable exception of judgments of the Supreme Special Court. In cases where a law is found to be unconstitutional by one of the other supreme courts, Parliament can change basic elements of the law that led to its unconstitutionality and re-enact it. It may also vote upon an amendment that has the same tenor as the previous law but avoids repeating the specific provisions which were found to be unconstitutional. If it did so – and nothing in the Constitution prohibits it – the courts would again set these provisions aside and would not implement them. However, jurisprudence is not a typical source of law, which means that courts may alter their case law and transform or repeal, at a later date, their prior judgments.

II.  Constitutional Culture A.  The Underlying Constitutional and Political Traditions The underlying constitutional tradition concerning the diffused and declaratory character of the judicial constitutionality review and especially the competence of every court to set aside and not apply any unconstitutional statutory provision goes back to the Greek courts’ jurisprudence of the nineteenth century, when legendary decisions,12 with content similar to the

11 Mark Tushnet, Weak Courts, Strong Rights. Judicial Review and Social Welfare Rights in Comparative Constitutional Law (Princeton NJ, Princeton University Press, 2008). 12 The ground-breaking judgments include Athens Court of First Instance 3504/1892, AP 169/1893 and 23/1897.

Greece  251 US Supreme Court case Marbury v Madison (1803), declared that no laws might alter the contents of the superior law of the state, which is the Constitution. Greek constitutionalism accepted early Emmanuel-Joseph Sieyès’ declaration that ‘une constitution est un corps de lois obligatoires, ou ce n’est rien …’.13 Nevertheless, the democratic, ie majoritarian, tradition of Greek ­constitutionalism preserves a form of parliamentary sovereignty, in the sense that no court – except the Supreme Special Court under rare circumstances – may formally abolish a parliamentary statute. To facilitate understanding how European integration and Greece’s EU membership have influenced the Constitution, it is worth mentioning that the latter is relatively rigid. In substantive terms, there is a hard core (‘eternity clause’) consisting of non-amendable fundamental ­principles relating to the parliamentary, not presidential, republican form of Government; human dignity; equality; the preservation of public offices only for Greeks; the prohibition of titles of nobility or distinction; the free development of personality and personal liberty; religious freedom; and the division of powers. In procedural terms, the Constitution may not be altered unless a period of five years has elapsed since the last revision. Moreover, revision requires a qualified majority in two consecutive parliamentary periods, ie a three-fifths majority before the elections and an absolute majority of the total number of parliamentarians post-election (or vice versa). This means that the revision procedure is cumbersome and lengthy. This explains why tacit and implicit changes usually occur in the (material) Constitution rather than formal amendments, as will be shown below.

B.  The Stance towards European Integration The Constitution plays an important role in the public debate. Political arguments are often dressed as constitutional ones. This has happened more intensely during the last years of the fiscal debt crisis, when the political stance against the Memoranda and the Greek bail-out programmes was based on their alleged unconstitutionality. Faced with such allegations, Courts retained a rather self-restrained stance. This stance has been especially critical during the crisis. At that time the Courts practically adopted a presumption of constitutionality of the laws introducing austerity measures based on the implicit supremacy of the legislator. While in previous years they tended to accept the citizens’ demands and award them with state benefits, thus contributing to the public deficit, they overturned this stance during the crisis. Consequently, the living Constitution has proven to be flexible enough to accommodate not only new interpretations and developments in the law, but also economic necessities. In some cases interpretation even resulted in the idleness of specific constitutional provisions without their formal deletion through constitutional revision. More specifically, European integration, as will be shown below, has caused tacit changes in the Constitution. These changes have been accepted smoothly by both state organs and Greek legal doctrine. It should also be noted that Greek constitutional theory and jurisprudence are rather cautious in accepting any far-reaching or, much less, unrestricted judicial review of laws, holding to the theory of self-restraint and preferring the ‘verfassungskonforme

13 ‘The Constitution is a body of obligatory legal provisions; otherwise it is nothing’, Emmanuel Joseph Sieyès, Opinion de Sieyès. Sur les attributions et l’organisation du Jury Constitutionaire proposé le 2 thermidor, prononcée à la Convention Nationale le 18 du même mois, l’an 3 de la République (Paris, Imprimerie Nationale, 3 [1795]).

252  Lina Papadopoulou Auslegung des Gesetzes’, which allows for an interpretation of the law in accordance with the Constitution.14 Moreover, the political forces in Greece and especially the ruling parties, since 1982 (at least until 2015, with the escalation of the dubious15 referendum in July 2015), have been euro-friendly, a fact that has led to very high majorities in favour of all steps of European integration.

III.  Constitutional Foundations of EU Membership A.  Relevance of General Provisions on International Law and Treaties 1.  Giving Effect to International Law According to Article 2 GC, Greece adheres to the generally acknowledged rules of international law and pursues the strengthening of peace. Under Article 28§1 GC, international law is divided into two broad categories: international conventions (or treaties16), on the one hand, and ‘generally accepted norms of international law’ (including general principles of international law, customary law, and general principles of state law applicable in the international legal order), on the other. This division seems to correspond to the widely accepted dichotomy of international law rules comprising ‘conventional international law’ and rules of ‘general international law’. It is possible, however, that the second category may comprise norms that are often included in conventional law. More specifically, generally recognised rules of international law, that is, the general principles of the law of nations and international customary law, are automatically integrated into domestic law.17 On the other hand, the President of the Republic concludes, as Article 36§1 GC18 stipulates, ‘treaties of peace, alliance, economic co-operation and participation in international organisations or unions and he shall announce them to the Parliament with the necessary clarifications, whenever the interest and the security of the State thus allow’. However, it should be

14 Spyropoulos and Fortsakis, Constitutional Law (n 1) 207; Philippos Spyropoulos, ‘Die Vermutung der Verfassungsmässigkeit der Gesetze und die verfassungskonforme Gesetzesauslegung’, in G Kassimatis and M Stolleis (eds), Geschichte und Staatsrechtslehre. Griechisch-deutsche Wechselwirkungen (Frankfurt/Main, Klostermann, 2001) 283. 15 Cf Lina Papadopoulou, ‘The 2015 Greek Referendum on Bailouts’, in J Smith (ed), Palgrave Handbook on European Referendums (Palgrave Macmillan, 2021) 625ff. 16 Cf Art 2 § 1(a) of the Vienna Convention on the Law of Treaties. 17 Spyropoulos and Fortsakis, Constitutional Law (n 1) 78. 18 Art 36 GC: 1. The President of the Republic, complying absolutely with the provisions of Article 35 paragraph 1, shall represent the State internationally, declare war, conclude treaties of peace, alliance, economic cooperation and participation in international organizations or unions and he shall announce them to the Parliament with the necessary clarifications, whenever the interest and the security of the State thus allow. 2. Conventions on trade, taxation, economic cooperation and participation in international organizations or unions and all others containing concessions for which, according to other provisions of this Constitution, no provision can be made without a statute, or which may burden the Greeks individually, shall not be operative without ratification by a statute voted by the Parliament. 3. Secret articles of an agreement may in no case reverse the open ones. 4. The ratification of international treaties may not be the object of delegation of legislative power as specified in Art 43 paras 2 and 4.

Greece  253 stressed that the participation of the President of the Republic in the making of a treaty is of symbolic nature only.19 Furthermore, Conventions on trade, taxation, economic co-operation and participation in international organisations or unions and all others containing concessions for which, according to other provisions of this Constitution, no provision can be made without a statute, or which may burden the Greeks individually, shall not be operative without ratification by a statute voted by the Parliament. (Article 36§2 GC.)

The ratification of treaties may not be the object of delegation of legislative competence by the Parliament to the Government (Article 36§4 GC). Consequently, treaties like those mentioned above take effect and become part of the national legal order as long as the statute ratifying them has been promulgated and published by the President in the Government Gazette. Exceptionally, treaties which do not fall under Article 36§2 GC, ie treaties not containing concessions, obligations or burdens, do not need to be ratified by the Parliament, but only announced to it, and take effect through the presidential decree promulgating them. Although the monist view may not be totally excluded, since Article 28§1 GC20 stipulates that after their ratification21 international law rules become ‘an integral part of domestic Greek law’,22 the same Article, as well as Article 36§2 GC, presupposes the ratification of treaties as a precondition for their validity as part of the Greek legal order, and thus implies the adoption of a dualist model.23 The dualist view is also strengthened by Article 28§2 GC, which allows for the transfer of constitutionally endorsed competences to international organisations.

2.  Placement of International Law in the Greek Legal Order Article 28 GC declares the position of public international law with regard to the internal sources of law. Adopted prior to accession, Article 28§1 GC explicitly stipulates that treaties, after their ratification, and generally accepted rules of international law are an integral part of the domestic law and prevail over all prior and subsequent contrary statutory provisions (‘shall prevail over any contrary provision of the law’).24 This supra-legislative status of international law means that all Greek laws may be judged by national courts on the criterion of their ‘conventionality’, ie their non-violation of an international treaty. In case of a conflict between a treaty provision on the one hand and a provision of a parliamentary statute (‘formal law’) or normative administrative act (‘substantive law’) on the other, the latter gives way and is not applied by any Greek court which

19 Spyropoulos and Fortsakis, Constitutional Law (n 1) 300. 20 Art 28§1 GC: ‘The generally recognised rules of international law, as well as international conventions as of the time they are ratified by statute and become operative according to their respective conditions, shall be an integral part of domestic Greek law.’ 21 Greek constitutional law and doctrine distinguish between ‘sanctioning’ (kyrosis) and ‘ratification’ (epikyrosis), although in the text of the Constitution itself the two Greek words are used somewhat interchangeably and inconsistently. 22 Cf Xenophon Contiades and Alkmene Fotiadou, ‘The Hellenic Republic’, in L Besselink et al (eds) Constitutional Law of the EU Member States (Alphen aan den Rijn, Kluwer Law International, 2014) 718 (720). 23 Cf Krateros Ioannou, ‘Greece’, in R Blackburn and J Polakiewicz (eds), Fundamental Rights in Europe. The European Convention on Human Rights and its Member States, 1950–2000 (Oxford, OUP, 2001) 358; İbrahim Özden Kaboğlu and Stylianos-Ioannis Koutnatzis, ‘The Reception Process in Greece and Turkey’, in H Keller and A Stone Sweet (eds), A Europe of Rights. Assessing the Impact of the ECHR on National Legal Systems (Oxford, OUP, 2008) 451, 462. 24 Arghyrios Fatouros, ‘International Law in the New Greek Constitution’, (1976) 70 The American Journal of International Law 492, 501 ff. See indicatively AP, 150/2018, of 22 January 2018 and 452/2018 of 2 March 2018 (both on the matter of eminent domain). Similarly, AP 1529/2017 of 7 September 2017 (concerning the ‘Convention on the Law Applicable to Contractual Obligations’ of 1980, known as the Rome Convention).

254  Lina Papadopoulou is competent to rule upon the specific case. However, even if found to be unconventional, the provision is not annulled. Instead; it merely remains inactive in that specific case, as also happens in the case of unconstitutionality. As in the case of constitutionality control, the scrutiny of conventionality is also diffused. This is especially the case with the control of non-violation of the European Convention on Human Rights and Freedoms (ECHR), as well as EU law, as is also demanded by the Court of Justice of the European Union (CJEU). Consequently, the view that conventionality control should be implemented ex officio (rather than on the sole basis of a litigant’s claim), again similarly to the constitutionality control, seems to be the most convincing one. Thus, it is evident that Article 28§1 GC has not only accommodated all treaties and played a vital role in the acceptance of EU law but it is also important with respect to the steps towards a European financial constitutionalism that are nowadays being taken. While the supremacy of international law over statutory law (stipulated by Article 28§1 GC) has been established, it is disputable whether the former (including EU law) also supersedes the provisions of the Constitution itself. As a matter of fact, there is no explicit reference to this relationship in the constitutional text itself. Thus, a wide margin of interpretation has been left open for theory and jurisprudence.

B.  Legal Basis for EU Membership 1.  The Accession to the EEC (EU) Greece was the first country to sign an Association Agreement with the EEC, as early as 1961. Already in 1958 it had applied for such an Agreement, which came into force in 1962. The ‘Athens Agreement’ aimed at Greece’s accession to the European Economic Communities (EEC) within 22 years. The Agreement was partly frozen for seven years (1967–74), at the initiative of the Commission of the EEC as a reaction to the military regime that assumed power in Greece on 21 April 1967, and was reactivated on the restoration of a parliamentary regime in 1974. The Constitution of 11 June 1975 entered into force in the light of the envisaged accession of Greece to the European Communities. Just one day after its entry into force on 6 December 1975 the New Democracy Government then in power under Constantine Karamanlis submitted to the Council of Ministers its application for the country’s membership of the Communities. The Accession Treaty was signed in Athens on 28 May 1979, ratified by Law 945/1979, and Greece became the tenth member of the then EEC on 1 January 1981.

2.  The Normative Basis for Greece’s Participation in European Integration (Article 28 GC) Greece’s accession to and participation in the EU is constitutionally based on Article 28 GC. That article is a gateway admitting international and supranational law into the national legal order, an outlet channelling national competences to international and supranational organisations, and a vehicle of supranational integration – the foundation of the model for Greece’s ‘open statehood’.25 Its stipulations continue to be the sedes materiae and are activated at each new stage of European integration upon ratification of each new treaty.

25 Theodora Antoniou, Europäische Integration und griechische Verfassung. Ein Beitrag zur Auslegung des Artikel 28 der griechischen Verfassung von 1975 (Frankfurt am Main, Peter Lang, 1985) 7 f and 157.

Greece  255 Article 28 GC has existed in the Constitution since the 1975 post-Junta Constitution was adopted. Although this predated Greece’s accession application, it had been specifically designed with a view to such membership. It was designed to allow for European integration in an open way, one that would minimise the need for subsequent amendments to the constitutional text. Besides paragraph 1, which applies to international law as a whole, the full meaning of the next two paragraphs is being revealed if one considers the way in which the Communities had functioned and developed up until that point. Article 28§2 and 3 GC remain the two main gateways which permit the insertion of EU law into the Greek domestic legal order.26 In the 2001 revision27 an ‘interpretative clause’ was added to Article 28, in order to reaffirm what had always been known, namely that this provision constitutes the foundation for the country’s participation in European integration. Thus, this constitutional provision is the ‘European clause’ of the Greek Constitution and accommodates its ‘integrational function’. Moreover, in the 2001 revision, a further ‘interpretative clause’ was added to Article 80 GC28 to accommodate constitutionally Greece’s accession to the EMU and its adoption of the euro.29 It is worth presenting the full text of Article 28 GC, which reads as follows: 1.

2.

3.

The generally recognised rules of international law, as well as international conventions as of the time they are sanctioned by statute and become operative according to their respective conditions, shall be an integral part of domestic Greek law and shall prevail over any contrary provision of the law. The rules of international law and of international conventions shall be applicable to aliens only under the condition of reciprocity. Competences provided by the Constitution may by virtue of treaty or agreement be vested in agencies of international organizations, when this serves an important national interest and promotes cooperation with other States. A majority of three-fifths of the total number of Members of Parliament shall be necessary to vote the law sanctioning the treaty or agreement. Greece shall freely proceed by law passed by an absolute majority of the total number of Members of Parliament to limit the exercise of national sovereignty, insofar as this is dictated by an important national interest, does not infringe upon the rights of man and the foundations of democratic government and is effected on the basis of the principles of equality and under the condition of reciprocity.

Interpretative clause: Article 28 constitutes the foundation for the participation of the Country in the European integration process.

Being the product of a long discussion and consultation process, Article 28 GC is characterised by a regulatory openness and flexibility. Thus, it provides a sure, safe, and implicitly evolving constitutional foundation for the EU integration process. However, the European Communities/ Union themselves were not explicitly mentioned until 2001.

26 Cf Julia Iliopoulous-Strangas and Eugenie Prevedourou, ‘Le droit de l’Union européenne et les Constitutions nationales. Rapport Hellénique’, in FIDE XX Congress, Vol I (London, BIICL, 2002) 259, 274. 27 Resolution of 6 April 2001 by the 7th Revisionary Parliament of the Greeks. The 2001 Revision was based on broad political consensus and very extensive: four new articles were inserted and 48 were revised. See Spyropoulos and Fortsakis, Constitutional Law (n 1) 57. 28 Art 80: (1) ‘No salary, pension, subsidy or remuneration shall be entered in the State budget or granted, unless it is provided for by statute concerning the organisation or other special statute.’ (2) ‘The minting or issuing of currency shall be regulated by law.’ Interpretative clause: ‘Paragraph 2 does not impede the participation of Greece in the process of the Economic and Monetary Union, in the wider framework of European integration, according to the provisions of Article 28’. 29 Konstantinos Gogos, ‘Die Diskussion über eine zukünftige Verfassung der Europäischen Union in Griechenland’, in H Timmermann (ed), Eine Verfassung für die Europäische Union: Beiträge zu einer grundsätzlichen und aktuellen Diskussion (Wiesbaden, Springer, 2001) 215, 217.

256  Lina Papadopoulou The addition of the interpretative clause moved the Greek Constitution out of the group of national constitutions without explicit reference to European integration into the group of those with such a normative clause, although nothing changed substantively. This allowed legal theorists to talk more persuasively about a constitutionalisation of Greece’s participation in the European Communities/Union, enhancing the view that the application of EU law and even its primacy constitute a constitutional obligation, which finds its limits in, and may be analysed only on the basis of Article 28§3 GC.

3.  Article 28 GC as a Basis for Accepting EU Law in the National Legal Order Primary Community/EU law bears all the hallmarks of a classic multilateral treaty, since the accession or entry of a state into an international organisation or a multilateral convention rests on an international agreement or treaty. According to this prevailing theory and practice, Article 28§1 GC has to be applied for the ratification of the EU Treaty, especially when the latter amends EU law without further recognition of constitutionally vested powers in EU bodies or restrictions on the exercise of national sovereignty. In practical terms, this would mean that a simple majority in the Parliament would suffice. In most cases, however, new EC/EU Treaties have resulted in further transfer of competences. In such cases, Article 28§2 GC remains the legal basis under the Constitution for further transfer of policy-making powers from the Member State to the EU. Article 28§3 GC explicitly empowers the legislator to restrict the exercise of national sovereignty under specific conditions in order to serve a ‘major national interest’. These limitations may occur either in the transfer of competences, when Article 28§2 GC is activated simultaneously, or in the adoption of human rights protection treaties or the establishment of a free trade zone.30 These two clauses – operating in a politically EU-friendly environment – create an equally friendly normative environment for absorbing primary law and enabling secondary law. It is, thus, of no surprise that Greek law has been greatly influenced by EU law (evropaikon dikaion), whose primacy over sub-constitutional law has been smoothly accepted from the very beginning. Nevertheless, and despite the fact that Article 28 GC does not clarify the position of EU law in the hierarchy of norms with regard to the Constitution, the interpretation of the latter and its material content have been directly affected by the fact that the national legal order co-exists with the EU institutional order.31

4.  Practice and Doctrinal Debate on the Relationship between the Last Two Paragraphs of Article 28 GC The relationship between the last two paragraphs of Article 28 GC and the question regarding under what circumstances either of them should be activated have preoccupied doctrinal debate. This concerns the desired normative result (transfer of competences for paragraph 2 and limitations on the exercise of national sovereignty in paragraph 3), as well as the different sets of substantive and procedural requirements the two paragraphs entail. The regulatory link between the two paragraphs is disturbed by the increased procedural requirements of the second

30 Fatouros, ‘International Law’ (n 22) 499. 31 Δημήτρης Τσάτσος, Συνταγματικό δίκαιο Ι. Θεωρητικό θεμέλιο [Dimitris Tsatsos, Constitutional Law I. Theoretical Foundations] (Athens, Sakkoulas, 1994) 29.

Greece  257 paragraph and the stricter substantive requirements of the third. Thus, the question arises of the casuistic manner of their implementation, especially regarding the ratification of the founding and amending EU treaties. a.  Substantive Requirements Concerning the substantive content of these paragraphs, in both cases, the service of an important national interest is presupposed. Moreover, in paragraph 2 such service is presupposed for the purpose of promoting co-operation with other states, while in paragraph 3 there is an extra limitation concerning human rights and democracy. Despite the fact that the national ­interest is a vague legal concept, which is subject only to parliamentary scrutiny and not to judicial review, any claim that such an interest is being served is not proven by the mere invocation of the country’s ‘national interests’, but requires detailed specification.32 The four conditions set out in Article 28 §3 GC are to be observed not only for the country’s accession to the Communities/ Union but also for the whole duration of its participation in those bodies. b.  Procedural Requirements The existence of an international treaty or agreement is a prerequisite in paragraph 2, but not in paragraph 3, where a one-sided decision of the Parliament suffices, as long as equality between states and a mutual reciprocity are guaranteed, which means that the other state(s) undertake(s) a similar obligation to restrict the exercise of its/their own sovereignty. Thus, the terms of equality and reciprocity exclude unilateral or unequal obligations to foreign powers or countries.33 The prerequisites of the non-infringement of human rights and democracy set the outer limits to any restriction of the exercise of national sovereignty. These requirements apply not only in respect of accession to the EC/EU but to every new step towards European integration, and are scrutinised by the Parliament. As yet, courts have not exercised any control based on these substantive prerequisites. In any case such control would run counter to the prerogatives of Parliament, and the courts’ commonly accepted and demanded self-restraint. The non-recourse to courts can be explained by the high political consensus that exists for all new treaties in the Parliament and the absence of a Constitutional Court. Interestingly enough, a different qualified majority is provided for in each case: three-fifths of the total number of members of Parliament (ie 180 out of 300) need to vote for the transfer of competences, while a simple majority of only 151 suffices for limitations on the exercise of national sovereignty. This differentiation between the two paragraphs was the outcome of lengthy discussions in the 5th Revisionary Parliament of 1975 (Ε‘ Αναθεωρητική Βουλή), which decided to adopt different majorities, despite the close relationship between their subject-matter.34 c.  Choice of Paragraph: The Prevailing Opinion Given the silence of the law ratifying the Accession Treaty and the verbal uncertainty of the two paragraphs of Article 28 GC, all possible interpretational versions have been supported by doctrine, which has held one of the following three positions: (a) the second paragraph is



32 Fatouros, 33 Ibid,

34 Ibid.

499.

‘International Law’ (n 22) 498.

258  Lina Papadopoulou applicable, since the accession of Greece to the European Communities constitutes a transfer of competences provided for in the Constitution to institutions of international organisations;35 (b) the third paragraph is applicable, since EC/EU membership results in limitations on the exercise of national sovereignty;36 (c) the view that both paragraphs form a ‘regulatory complex’ and have to be applied cumulatively and simultaneously in the ratification of every new EC/EU Treaty.37 The prevailing view supports a joint reading of the two paragraphs, which must be interpreted and implemented in conjunction with each other since they regulate the same thing. According to this prevailing view, it has also been argued that the ‘regulatory core’ of paragraph 3 is more comprehensive and includes, as a subset, the cases covered by paragraph 2, in the sense that any recognition of transfer of competence to organs of international organisations constitutes a restriction on the exercise of national sovereignty, which, however, may stem from other reasons. According to this view, in the transfer of competences to international organisations, the stricter substantive requirements of the third paragraph should be met alongside the strict requirements of the second paragraph. In practice, there has never been a need to distinguish between the two paragraphs, since all Treaties have been ratified with a majority higher than the qualified one provided for in Article 28§2 GC. It is significant that, despite the different majorities required, the first and most fundamental Article of Law 945/1979 ratifying the Act of Accession did not specify which of the paragraphs of Article 28 GC it was based on. This can be attributed to the fact that a great majority that voted in favour. So, the comfortable view that both provisions should be applied cumulatively has never posed any real problems. In doctrinal terms, however, the question remains ‘why the legislator has not formulated the text as one single paragraph, if paragraphs 2 and 3 form a normative complex that is always applied together’.38 d.  A Minority Opinion: Attributing Normative Autonomy to the Two Paragraphs Α minority opinion has attempted to provide a different answer and distinguish the normative spectrum of each of the two paragraphs, while still recognising that on many occasions both paragraphs still need to be applied cumulatively. The criterion has been the operation of the political entity to which competences have been transferred, or in favour of which limitations on the exercise of national sovereignty have been enforced. In particular, it has been proposed that the application of the one or the other provision is directly related to the extent to which

35 See indicatively Emmanuel Roucounas, ‘Le droit international dans la Constitution de la Grèce du 9 juin 1975’, 29 Revue hellénique de droit international (1976) 51. 36 The application of para 3 has been supported by Πρόδρομος Δαγτόγλου [Prodromos Dagtoglou], ‘Ο Έλλην νομικός και το ευρωπαϊκό κοινοτικό δίκαιο’ [‘The Greek Lawyer and European Community Law’], (1976) 24 Νομικό Βήμα [Νοmiko Vima] 929, 932. 37 See indicatively, Fatouros, ‘International Law’ (n 22); Dimitrios Evrigenis, ‘Aspects institutionnels et juridiques de l’élargissement. L’impact de l’adhésion sur les institutions et le droit des pays candidats: Grèce’, in W Wallace and I Herreman (eds), A Community of twelve. The impact of further enlargement on the European Communities (Bruges, De Tempel, 1978) 13; Dimitrios Evrigenis, ‘Legal and Constitutional Implications of Greek Accession to the European Communities’, (1980) 17 CML Rev 157; Julia Iliopoulos-Strangas, ‘Rechtsfragen der Mitgliedschaft Griechenlands in einer Europäischen politischen Union’, (1985) 20 Europarecht 199, 204; Gogos, ‘Diskussion’ (n 27) 216. 38 Some commentators believed that this is due to a poor wording of the constitutional text. See eg Αργύριος Φατούρος [Arghyrios Fatouros], ‘Διεθνείς πράξεις με σοβαρές επιπτώσεις στην άσκηση της εθνικής κυριαρχίας΄ [‘International Acts with Important Influence on the Exercise of National Sovereignty’], in K Ioannou et al, Δημόσιο Διεθνές Δίκαιο – Σχέσεις διεθνούς και εσωτερικού δικαίου [International Public Law. Relations between International and Domestic Law] (Athens, Sakkoulas, 1990) 85.

Greece  259 the international organisations concerned are of intergovernmental or supranational character and the manner in which decisions are made. According to this opinion, paragraph 2 is applied when constitutionally vested competences are transferred to intergovernmental organisations where decisions are taken by unanimity. In this case these competences are taken from the arsenal of national constitutional provisions, although no decision may be made without the state’s consent, and so the exercise of national sovereignty is not restricted. On the other hand, the exercise of national sovereignty is restricted when decisions in a specific field are made on the basis of a (qualified or simple) majority, which means that a state might be out-voted and still be obliged to implement a decision with which it disagrees.39 In this case supranational norms constitutive of the international organisation cause the national Constitution to be set aside. Then more stringent substantive requirements of Article 28§3 GC come to the forefront, ie human rights and democracy. This is because the reciprocal limitations on the exercise of national sovereignty result in the creation of a supranational legal system (eg the EU) to which the national constitutions of all participating states no longer apply. This gap creates a legal and political need to transfer the fundamental principle of liberal democracy to this new supranational ‘domain’ for the exercise of sovereignty. In this case we are dealing with a form of supranational democracy, where political decisions are subject to reason on the basis of the supranational interest, beyond the fragmented and partial national interests. Given the fact that these limitations are mutual and reciprocal, the result is a transformation of sovereignty, which, instead of being an undivided and exclusive property of the state, now becomes a shared and commonly exercised feature of political power.40 e.  Practical Application of the Provisions of Article 28§§2-3 GC The Ratification Law 945/1979 was voted by 191 of the 300 deputies. Then Prime Minister Constantine Karamanlis declared that this enlarged majority was not an interpretative precedent binding on any subsequent occasions. This reservation has not yet become critical, as all amendments to the Treaties establishing the Communities and the Union have been voted for by both major parties and, thus, have received a majority of votes far greater than the one required even by the second paragraph of Article 28 GC. The ratification of the Accession Treaty of Spain and Portugal (Law 1572/1985), the Single European Act (Law 1681/87) and the Treaty of Schengen followed the ordinary process of treatyratification (Articles 36§§1-2 and 28§1 GC). In the ratification of the Treaties of Maastricht (Law 2077/1992), Amsterdam (Law 2691/1999) and Lisbon (Law 3671/2008) the applied provisions were not specified. All three were ratified with higher than two-thirds majorities of the total number of deputies. On the other hand, in the laws ratifying the Treaty of Nice (Law 3001/2002) and the ‘Treaty Establishing a Constitution for Europe’ (Law 3341/2005–268 votes in favour, 17 against, 15 abstentions) there is a clause stipulating that these Treaties are ‘ratified and have the validity laid down in Article 28§1 GC’. The acknowledgement of either one paragraph or the other has political rather than legal connotations.

39 Λίνα Παπαδοπούλου [Lina Papadopoulou], Εθνικό Σύνταγμα και Κοινοτικό Δίκαιο: Το ζήτημα της ‘Υπεροχής’ [National Constitution and Community Law: the Question of ‘Supremacy’] (Athens, Sakkoulas, 2009) 402 ff. 40 Papadopoulou, National Constitution (n 37) 406.

260  Lina Papadopoulou

C.  EU Integration Outside the EU legal order: ESM, Fiscal Compact Integration instruments outside the EU legal order such as the European Stability Mechanism (ESM) or the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union (TSCG, the so-called Fiscal Compact) do not differ, in formal terms, according to the Greek Constitution, from ordinary EU law. All these are treaties under international law that need to be passed with simple majority in Parliament based on Article 28§1 GC. Their normative power is superior to common statutes, although, as mentioned above, there is no specific relation to the Constitution is mentioned. However, should such treaties be considered to allow the transfer of constitutionally vested competences – which would imply a tacit change of the Constitution without formal revision – they would need to be passed by a qualified majority of three-fifths of the Deputies. In other words, the Constitution does not differentiate between EU and international law. EU law is (a type of) international law for which simple majority suffices. The criterion for a differentiation of the majority required is the transfer of competences (three-fifths majority required) and/or the restriction on the exercise of national sovereignty (absolute majority of all Deputies required plus the substantive presuppositions mentioned above). Accordingly, all three Memoranda of Understanding between Greece and its creditors have been voted upon by the Parliament based on Article 28§1 GC, since they were not Treaties involving transfer of competences to institutions of international organisations but only political programmes with objectives of general policy and the means of achieving it in a given timetable. According to the Council of State,41 adjudicating upon the first Memorandum (Law 3845/2010), the latter was not an international agreement, since the contracting parties did not assume any mutual commitments and it did not foresee any sanctions against the Greek authorities in case of non-implementation.42 Similarly, the amendment of Article 136 of the Treaty on the Functioning of the European Union (TFEU), the ESM Treaty and the Fiscal Compact were ratified through Law 4063/2012. Through its ratification both treaties have acquired the legal validity foreseen in Article 28§1 GC, ie supra-statutory legal force. Both instruments were approved and ratified – together with the European Council’s Decision of 25 March 2011, which modified Article 136 TFEU (2011/199/ΕU) – through Law 4063/2012 (FEK 71/30/03/2012) based on Article 28§1 GC and with the ­supra-statutory validity foreseen therein.

IV.  Constitutional Limits to EU Integration The most important, ‘horizontal’ limit is that the ‘competence of competence’, that is, state ­sovereignty43 as such, cannot be transferred. Only specific competences are eligible to be transferred to international organisations. According to Article 28§3 GC, it is the ‘exercise’ of national sovereignty and not sovereignty itself that can be transferred and this restriction should be mutual.

41 CoS 668/2012 (Plenary), 1283–4/2012 (Plenary), 3783/2015 (Section A), 2432/2017 (Section A). 42 Maria Meng-Papantoni, ‘Legal Aspects of the Memoranda of Understanding in the Greek Debt Crisis’, 18 Zeitschrift für Europarechtliche Studien (2015) no 3, 15. 43 Αντώνης Μανιτάκης [Antonis Manitakis], Ελληνικό συνταγματικό δίκαιο [Greek Constitutional Law] (Athens, Sakkoulas, 2004) 385.

Greece  261 The ‘leeway’ for the exercise of public power generated through this mutual restriction of national sovereignty must, according to Article 28§3 GC, be established on the basis of human rights and democracy, that is, in constitutional terms.

A.  Limits to the (Further) Transfer of Powers to the EU through Treaty Amendments 1.  ‘Core Competences’ Which ‘Must’ Remain with the National State Part of Greek constitutional theory has been influenced by the doctrine of Italian origin concerning the controlimiti, or ‘counter-limits’, to European integration.44 This is a jurisprudential concept according to which there is a number of ‘core competences’, or ‘non-transferable constitutional qualities’, which may not be transferred to the EU but need to remain with the national state. Within this context, it has been purported that the substantive requirements of Article 28§3 GC, ie human rights and democracy, establish a constitutional reservation of sovereignty, a hard core of sovereignty that may not be removed from the national state. According to this view, a state cannot abandon its internal sovereignty, in the sense that it is inconceivable that a state would concede its highest public power to define, through its constituent and revisionary powers, the terms or rules of its establishment and operation as a polity. This has to do with ‘constitutional self-determination’ and the reservation of the ‘constitutional identity’ of each state.45 This theory of counter-limits rests upon the theory of democratic statism, according to which only the national state is capable of ensuring democracy and human rights – it is the necessary eggshell for democracy to be born and to flourish. This view was supported by the absence in the Treaties of any reference to democracy and human rights in the first few decades of the Communities’ existence. The reasoning also goes back to the ‘no-demos-thesis’, according to which there is no European demos competent to exercise the pouvoir constituant and bring about a pan-European Constitution establishing European democracy and a human rights regime. According to this view, Article 28§3 GC also functioned as a rule for competence sharing. This view has become outdated, especially after the Union’s attempts to deal with its democratic deficit. A more open reading of this clause suggested an ‘extroverted’ interpretation of the substantive terms democracy and human rights. Already in 1978 Evrigenis concluded that the concepts of ‘human rights’ and ‘foundations of democratic government’ determined the ‘political philosophy’ of international organisations in favour of which Greece could make restrictions on its national sovereignty.46 In other words, it has been proposed that some requirements for European integration may be deduced from Article 28§§2–3. If this view is combined with the one mentioned above, which holds that the limitations to the exercise of national sovereignty are tantamount

44 This theory was initially and mainly introduced by Αντώνης Μανιτάκης [Antonis Manitakis], ‘Τα όρια της κοινοτικής αρμοδιότητας και η συνταγματική θεώρησή τους’ [The limits of the Community’s competence and their constitutional validation], (1984) Το σύνταγμα, 472, 490 and was until recently followed by a significant proportion of Greek public laywers. 45 Manitakis, Constitutional Law (n 41) 386. 46 Evrigenis, ‘Aspects institutionnels’ (n 35) 135; Ευάγγελος Βενιζέλος [Evangelos Venizelos], Μαθήματα Συνταγματικού Δικαίου [Lectures of Constitutional Law] (Athens, Sakkoulas, 1991) 153.

262  Lina Papadopoulou to the substitution of the national Constitution by EU law, it may be inferred that a European constitutional law, comprising the provisions specifying European democracy and human rights, is necessitated by the Greek Constitution. Its requirements – together with those of other national constitutions – may serve as ‘pre-constitutional rules’ enabling the European constitution to be born (Constitutio europae nascitura).

2.  No Scrutiny of Probable Unconstitutionality of New European Treaties Judicial review of European Treaties is procedurally impossible before their ratification. The institutions that can scrutinise them with respect to possible conflicts between the new provisions of European law and the Greek Constitution are the Greek Government and particularly the Prime Minister, who signs each Treaty, and the Parliament, that is required to ratify it. The absence of scrutiny or specific and detailed reports by scientific advisory committees of the relevant constitutional institutions is indicative of the deficits that characterise the functioning of parliamentary democracy in Greece. In any case, in practice, so far no constitutional revision has ever taken place prior to the ratification of European treaties, nor has the issue of their unconstitutionality ever been posed by the Government or the Parliament. The Constitution welcomes EU law on the sole basis of Article 28 GC, without revision of the specific constitutional provisions that might be affected. Thus, any changes in the substantive constitution are not explicitly reflected in the body of the formal Constitution, but remain informal.47

B.  Courts in Action: Scrutiny of Secondary Legislation against the Yardstick of National Constitutional Law 1.  The Receptive Attitude of Greek Courts Already from the very beginning, the Council of State (the Supreme Administrative Court – Simvoulio Epikratias) has demonstrated a receptive attitude towards accepting that, like the whole corpus of public international law, European law is also part of domestic law and as such is fully enforceable.48 Both primary and secondary law issued prior to Greece’s accession to the EEC enjoy clear and undisputed supremacy over common statutes due to Article 28§1 GC. This provision clearly defines the position of the European Treaties, which are by their legal nature treaties of public international law. It does not, however, provide an answer concerning the position of secondary EU law since accession. The validity of secondary EU law issued a posteriori to the accession is derived from Article 28§2 GC, which stipulates that ‘competences provided by the Constitution may by treaty or agreement be vested in agencies of international organisations’. This necessarily means that the law to be produced by the competent organs of these international organisations has to

47 Cf Lina Papadopoulou, ‘Die implizite Änderung der griechischen Verfassung durch das EU-Recht’, (2014) 74 Zeitschrift für ausländisches öffentliches Recht und Völkerrecht 141, 161; Antonis Manitakis, ‘In Greece, Constitutional Changes Without Constitutional Revision’, Huffington Post, 14.03.2016. 48 George Gerapetritis, ‘The Hellenic Council of State vis-à-vis the European Court of Justice: From voluntary seclusion to inevitable constitutional dialogue’, in N Alivizatos et al (eds), Essays in honour of Georgios I. Kassimatis (Athens/ Brussels, Sakkoulas/Bruylant, 2004) 83.

Greece  263 be accepted as such. Otherwise the transfer of the legislative competence in specific fields will have no real effect and the provision of Article 28§2 GC will be practically void. Consequently, not only has the Council of State recognised from the outset the enforceability of EU law and its prevalence over any statutory provision, but it has also ruled that domestic courts must examine on their own motion the compatibility of domestic law with EU law.49 So, the Greek judge is situated to act both as a national and a European judge. While acting in this double role (‘dédoublement fonctionnel’), the judge is called upon to interpret the statutory provision in accordance with both the domestic Constitution and EU law and set it aside without applying it in the event that he finds it breaches either or both. In practice, Greek judges have not taken this role very seriously but have rather exhibited inertia in this field.50 It is commonly recognised that the courts need to scrutinise the conformity of national legislation with both the Constitution and EU law.51 In its judgment 161/2010 (Plenum), the Council of State declared that the control of constitutionality has to enjoy priority and only if the law is deemed to be in conformity with the Constitution does the court need to proceed with the control of conventionality (conformity with EU law).52 The Courts’ competence to control the (un)constitutionality of EU law is based on the following sequence of thoughts. EU law after its ratification by law becomes an integral part of domestic Greek law, as stipulated in Article 28§1 GC. Therefore, it is law, in the material sense of the term, subject to the same substantive restrictions as domestic laws. The latter must be in harmony with, or at least not be contrary to, the Constitution, which, all Greek institutions, including the judiciary, are required to secure respect of.53 This view is consistent with the alleged superiority of the Constitution with respect to EU law, the application of which is regulated by specific constitutional provisions as well as fundamental constitutional principles. Consequently, according to this stance, the constitutional review of EU law is possible even on a case-by-case basis, as each EU act, like any domestic law, is subject to the national Constitution. These considerations are reflected in an exemplary way in CoS judgment 3242/2004, which invoked the competence of all Greek courts (Articles 93§4 and 87§2 GC) to perform a constitutionality control of all laws without exception, ie substantive laws, which include EU regulations or directives. The above position in favour of the controllability of the constitutionality of EU law denied the unique nature and relative autonomy of the EU legal order, and underestimated the normative function of Article 28§§2-3 GC. In contrast, more convincing is the view that national courts are not competent to judge an EU act on the basis of the national Constitution, as the two systems, despite being interconnected, have their own systems for monitoring and evaluating their respective acts. According to this view, the constitutionality control that Greek courts may exercise on EU law is limited by the criteria laid down by Article 28§§2-3 GC which are more specific in relation to the general provision of Article 93§4 GC.

49 CoS 425/2015 (plenary session), 1116/2014 (plenary session), 3670/2006 (plenary session). Gerapetritis, ‘Council of State vis-à-vis the ECJ’ (n 46) 83. 50 Constantin Yannakopoulos, ‘L’influence du droit de l’Union européenne sur le système de contrôle juridictionnel de constitutionnalité des lois: les paradigmes français et grec’, (2012) 91 Revue française de droit constitutionnel, 537. 51 Exemplary rulings on the obligation of courts to scrutinise national law against the yardstick of EU law are the early judgments CoS 3502/1994 and 249/1997, which underlined that Greek judges must set aside and not implement any Greek law provision contrary to EU law (and the ECHR). Similarly, CoS 180/2008 (Β Section). 52 Similarly to the French organic law 2010–1523 concerning the application of Art 61-1 of the French Constitution. 53 Art 93§4 GC (‘The courts shall be bound not to apply a statute whose content is contrary to the Constitution’) in conjunction with Art 87§2 (‘In the discharge of their duties, judges shall be subject only to the Constitution and the laws; in no case whatsoever shall they be obliged to comply with provisions enacted in violation of the Constitution’).

264  Lina Papadopoulou However, if the controlimiti of Article 28§3 GC are considered to be extroverted terms applicable to the EU, then the control necessarily becomes one of a systemic and not a case-by-case type. It concerns the institutional system of the European Sympolity54 in general. Furthermore, it becomes marginal and exceptional, in the sense that national courts may hear a case only in the exceptional, hypothetical and unlikely event that the institutional system deviates from any concept of democratic government, or if it undermines the democratic governance of Member States or only of Greece, meaning a prior violation of the term of equality and reciprocity, or ignores, in a profound, lasting, and persistent way the fundamental rights defined in Article 110 GC, which delineates the hard, non-revisable core of the Constitution. Therefore, the only possible judicial review of the constitutionality of EU law is the marginal judicial review of compliance with the clauses of Article 28§3 GC guaranteeing the core of the constitutional and political culture, not only of the Greek constitutional order and the constitutional order of each Member State. Such a scrutiny has not yet been performed by the courts.

2.  A Consistent (Europe-friendly) Interpretation of the Constitution … Given the procedural possibilities of a last-instance control by the CJEU, and in order to ensure uniformity and legal certainty within the EU as a whole, it is proposed and widely accepted that the constitutional rule should be adapted to EU law, based on an interpretation of the former in accordance with the latter, similar to the constitutionality control of laws.55 More specifically, the view has been expressed that the EU legal order demands the enrichment of the interpretation of law in general, and the interpretation of the Constitution in particular, with a cognitive process that enables it to sufficiently correspond to the new historical fact of the European dimension that the function of law seems to be acquiring … The interpreter, in approaching his subject, must examine the impact of European law on the meaning of the interpreted rule, thus taking into consideration the broader system of the European legal culture into which Greek law has been incorporated.56

Moreover, the revisionary function, as a sui generis constitution-producing process, is particularly subject to the ‘Europe-friendly’ approach in respect of the constitutional material under revision.57 A more solid basis for the formulation of a group of indirect but extremely crucial restrictions for the revisionary legislator would be recourse to the concept of the ‘European legal system’, which encompasses the international (EU-specific) obligations of the Greek state with – according to the traditional theory on the hierarchy of the legal sources – a supra-constitutional validity. In fact, however, it has been proved that if the legislator amending the Constitution ignores these obligations, this may create fields of tension between the Constitution and international and EU law, with unpredictable consequences both for the institutional effectiveness and the symbolic function of the constitutional text. 54 For this concept see DimitrisTsatsos, The European Sympolity: Towards a New Democratic Discourse (Brussels, Bruylant, 2009). 55 See, indicatively, Evangelos Venizelos, ‘The relationship between National Constitution and European Community Law after the signature of the Treaty for the European Constitution and the revision of the 2001 Greek Constitution’, (2005) 1, Hellenic Revue of European Law 1, 20, 23. See also Appellate Court of Athens 14/2008 (5th Section), which found the foundation for such an obligation in Art 249§3 EC Treaty and in the ECJ jurisprudence. 56 Tsatsos, Constitutional Law I (n 29). 57 Cf Προκόπης Παυλόπουλος [Prokopios Pavlopoulos], ‘Το Σύνταγμα μεταξύ διεθνούς και Ευρωπαϊκού Δικαίου. Η ιεραρχία της έννομης τάξης’ [‘The Constitution between international and European law. The hierarchy of the legal order’], (2015) 10 (2) Εφημερίδα Διοικητικού Δικαίου, 154, who proposes that both legal sources – the national Constitution and European law – should be interpreted in accordance with each other.

Greece  265

3  …. Leading to an Implicit Change in the Meaning of Constitutional Provisions Often, an interpretation of the Constitution in accordance with EU law exceeds the enrichment of the conceptual content of the constitutional terms and results in an indirect and implicit change in the meaning of constitutional provisions (even contra legem). In this way, the interpretative process allows for a continuous adjustment of the material Constitution to modern sociological and normative requirements and the goals of European integration. This reinterpretation, however, sometimes substitutes the adaptation of the constitutional text through the formal revisionary process, despite the fact that according to the prevailing opinion in Greek legal doctrine such a change may only be brought about through Article 110 GC, which regulates the formal revision procedure. Nevertheless, many examples show that in the end this kind of change does take place through Article 28§§2-3 GC due to the prevalence of EU law. A classic example is that concerning the occupation of public offices by non-nationals. The Constitution continues to contain the non-revisable Article 4§4 according to which ‘[o]nly Greek citizens shall be eligible for public service, except as otherwise provided by special laws’. Since the Maastricht Treaty, however, this provision has been interpreted as allowing all EU citizens to occupy public offices in accordance with EU law and CJEU’s jurisprudence. Article 48 TEC (Article 39 TEC [Maastricht], now Article 45 TFEU) and Regulation 1612/1968 may be considered58 to be the ‘special law’ foreseen by Article 4§4 GC.59 The Council of State has held an interpretation of Article 4§4 GC in accordance with EU law, the CJEU case law and a narrow specialisation of the concept ‘public functions’, essentially adopting the quality criteria set by the CJEU for the concept of ‘public administration’ in Article 39 para 4 TEC (Article 45 TFEU).60 With the restrictive interpretation of the term ‘public functions’, however, it has opened up access to the public service not only to European citizens but also to all other foreigners. If, instead, it had accepted the amendment operation of Article 28 GC as a result of its integrationist function, the restrictive interpretation of Article 4§4 GC may indeed have been limited exclusively to European citizens. Another example of an interpretation of the Constitution in accordance with EU law that effectively results in an indirect and implicit modification of a constitutional provision was the right of Europeans residing in Greece without Greek citizenship to vote and stand as a candidate in local authority elections. According to prevailing opinion, Article 102§2 GC,61 though not featuring an explicit provision on this matter, presupposed that only Greeks had these political rights. Under the influence of EU law, however, the opposite interpretation prevailed, accepting that the law could allow for voting rights for non-Greeks they were enshrined in primary and secondary EU law.

58 See, for example, One-Member First Instance Court of Athens judgment 2228/1992, 59 ECJ 02.07.1994 C-290/94 (Commission of the European Communities v Hellenic Republic) ECLI:EU:C:1996:265; see also CoS opinion on Decrees 380/1996, 501/2001. 60 See the CoS opinions on Decrees 36/2004, 61/2004, 196/2004, 40/2003, 29/2003, 423/2002, 274/2002. It should be noted that the elaboration of all decrees of regulatory nature falls under the jurisdiction of the Council of State, which has the competence to give an opinion concerning their legality. Law 2431/1996 then explicitly provided for the possibility of European citizens being appointed to positions of public service, in public entities, local authorities or organisations and entities controlled by the state. 61 ‘Local government agencies shall enjoy administrative independence. Their authorities shall be elected by universal and secret ballot.’

266  Lina Papadopoulou As already shown in the example of the occupation of public offices by non-Greek EU nationals, and also in the ‘main shareholder’ and ‘private universities’ cases analysed above, the meaning of the national Constitution is being changed in order not to oppose EU law, although no formal revision of its wording has taken place. Formal revisions are often carried out in other legal systems, but not in the Greek one, not least due to the cumbersome procedure involved. As a matter of fact, it may be observed that the more difficult a revision procedure is, the more likely it is that substantive constitutional stipulations will change through interpretation and not through a formal revisionary procedure. Going back to constitutional culture, one may also note that a certain normative hypocrisy exists, to the extent that specific norms typically survive, despite being seriously circumvented in practice. The example of the so-called ‘private universities’ is typical for both explanations. First, it would have been very difficult to find the large majorities required by Article 110§2 GC, since the prohibition of private tertiary education is a symbol of both leftist and state-centric politics. Secondly, upholding this prohibition serves a kind of hypocritical loyalty to public education, even if at the same time public education suffers in many different ways.

C.  The Question of Primacy vis-à-vis the Constitution The ‘European legal system’ comprises two distinct, independent subsystems – the CoE and the EU – each of them serving different institutional functions, political and legal objectives. From both of these subsystems obligations arise that the constitutional legislator has to take into consideration both when exercising a constituent function and also during the revisionary process. As a consequence, in the case of Greece, the Constitution has long ceased to be the supreme source of law, even though this is not expressly acknowledged by national scholars and courts and is even denied by the high authorities of the State,62 as it will be shown below.

1.  A Longstanding Doctrinal and Jurisprudential Debate As already noted, the Constitution does not provide for the hierarchical supremacy of EU law (or other public international law) over national constitutional law. As a consequence, this matter has been left to theorists and the judiciary and has preoccupied doctrinal debate. It is worth mentioning that in Greek legal doctrine and partially also in case law, the word yperohi (υπεροχή = supremacy) is used, rather than the more appropriate word protereotita (προτεραιότητα = precedence or primacy, Vorrang). This wording can be traced back to the official translation of the landmark decisions of the CJEU, where the word yperohi rather than protereotita is used.63 All three possible opinions have been supported: (a) the supremacy of the national Constitution, (b) the supremacy (or, in a differentiated version, the primacy) of EU law, and (c) the placement of both on an equal footing, together with other treaties, such as the ECHR. Most constitutional lawyers have remained loyal to the national Constitution, purporting that no legal source may be superior to it. The foundation of this opinion has been either democratic statism,

62 Cf Pavlopoulos, ‘Constitution’ (n 55). 63 See, for example, the official translation of ECJ 09.03.1978 106/77 (Amministrazione delle Finanze dello Stato v Simmenthal SpA) ECLI:EU:C:1978:49, para 17; cf CoS 3242/2004, para 19; CoS 3670/2006, para 21.

Greece  267 based on the premise that democracy is not possible outside the nation state and the no-demosthesis, since no democratic constituent power has ever been formed,64 or the positivist account that it is logically impossible for the Constitution to recognise another legal source as superior to itself. A counterargument is that a strict, literal approach to the Constitution could not allow for a self-restraint clause open to the Constitution’s self-abolition or for an international organ to violate the provisions of the Constitution because in such a case the international forum would be more powerful, although enjoying far less democratic legitimacy and direct accountability. On the other hand, EU law disciples65 insist on the approach suggested by the CJEU ie the yperohi of EU law based on the special, autonomous nature of European law itself and the treaties. From this methodological point of view, Article 28 GC and Act 945/1979 (which ratified Greece’s Accession) mark the ‘zero hour’, ie the timely and logical starting-point of the validity of EU law in Greece. They are not, however, the foundation of this validity, which is to be acknowledged de iure communitatis, and includes supremacy or primacy over all national law, including the Constitution. The intermediate opinion that the national Constitution and the EU treaties are to be placed on an equal footing, together with the ECHR, has also been proposed.66 In the event of conflict, the lex specialis rule applies. On the other hand, the lex posteriori rule is undermined since it would violate Article 110 GC regulating formal amendments to the Constitution, which excludes implicit changes in the latter. Furthermore, the author accepts67 the classical foundation of national monism, according to which the Constitution is, within the context of the national legal order, the sole highest legal source. A minority opinion68 sees the primacy of EU law enshrined in Article 28§§2 and 3, which allow for Greece to transfer competences which were initially provided for in the Constitution itself, and restrict the exercise of its national sovereignty. According to this opinion, Greece is deprived of these competences, in a definitive manner for as long as the country remains an EU Member State, given that these same competences have been realised at EU level (principle of pre-emption). Thus, EU law obliges the national organs, including judges, to set aside and not implement (though not abolish) Greek law, even constitutional norms, in favour of EU law. Opponents of this view purport that no legal source may recognise higher validity than its own in another legal source.69 This could only be the case if Article 28 GC were itself superior to the other constitutional provisions, a concept that is not valid.70

64 A point recently raised by the former President of the Hellenic Republic and Professor of Public Law, Pavlopoulos, ‘Constitution’ (n 55). 65 See, eg, Πρόδρομος Δαγτόγλου [Prodromos Dagtoglou], ‘Η συμμετοχή μας στις Ευρωπαϊκές Κοινότητες και το Σύνταγμα’ [‘Our Participation in the EC and the Constitution’], in Stangos and Moutsios (eds), Προσχώρηση και συμμετοχή της Ελλάδος στις Ευρωπαϊκές Κοινότητες. Η συνταγματική θεώρηση [Greece’s Accession to, and Participation in the EC: the Constitutional Viewpoint] (Thessaloniki-Athens, Sakkoulas, 1987) 19. 66 Βασίλης Τζέμος [Vassilis Tzemos], Ο βασικός μέτοχος [The Basic Share-holder] (Athens-Komotini, Sakkoulas Ant, 2006) 190 ff. Tzemos rests his conclusion mainly on a positivist basis, given that no constitutional provision either stipulates or excludes the superiority of the one or the other. However, he recognises that the ECHR develops a de facto validity superior to the Constitution (200). 67 Ibid, 190. 68 Papadopoulou, National Constitution (n 37) 413 ff. 69 Pavlopoulos, ‘Constitution’ (n 55) invokes the fundamental principle of Roman law, nemo plus juris ad allium transfere potest quam ipse habet. 70 Antoniou, Europäische Integration (n 23) 39.

268  Lina Papadopoulou In a few Greek court judgments an attempt has been made to support the primacy of EU law over constitutional provisions (mainly in Article 28 GC).71 According to some commentators, this would not be viable, since national courts only have the national Constitution as their ‘rule of recognition’, and recognising a different one would result in a kind of judicial revolution. Thus, it is of no surprise that in most of their judgments the highest courts have declared the inviolability of the domestic Constitution. For a long time Greek courts avoided looking for links with EU law and rarely sent preliminary references to the CJEU,72 mainly in order to avoid the possibility of a constitutional provision being found at odds with EU law.73 They often used the CILFIT case law and invoked the theory of acte clair. They also regularly and with few exceptions74 refrained from taking a position on the question of the relationship between EU and Greek law, in particular the Constitution, and on the primacy issue. Even later, when the issue began to be raised, the case law of both the Council of State and other Greek courts varied and on many occasions, different sections, or even the Plenum of the same Court, due to a different composition, made conflicting decisions, not guided by a landmark judgment.75 A shift in this trend is reflected in two judgments of the Council of State (3670/200676 and 778/200777) on two hot issues of tension between the Greek Constitution and EU law. These are the so-called ‘principal shareholder’ case and the recognition of professional qualifications based on academic titles acquired in non-public Universities, which will be examined in detail below.

2.  Areas of Tension between the Greek Constitution and the European Legal System a.  The ‘Principal Shareholder’ Case Undisputedly, the most prominent example of tension between the Constitution and EU law is Article 14§9 sections e–f GC,78 which were added in the 2001 constitutional revision. This

71 For example, CoS 3312/1989, Athens Court of First Instance 9099/1994; Art 28 GC is directly invoked by the minority opinion in judgments CoE 3242/2004 and 3243/2004, while CoS 3670/2006 (Plenum) does not refer to constitutional provisions in order to establish control of the compatibility of the constitutional rule with Community law; rather, it is based on the demarcation between the scope of the two relevant provisions. 72 According to a survey conducted in 2005, no more than 15 cases per year out of a total of 4,500–5,000 judgments found a link to EC/EU law. Up until 2005 the Council of State had referred to the ECJ only 14 times for preliminary references, and more than half of these times after 2000; See Φιλοκτήμων Αρναούτογλου [Feloktemon Arnaoutoglou], ‘Συμβούλιο της Επικρατείας και Ευρωπαϊκά Δικαστήρια’ [‘Council of State and European Courts’], (2005) Νομικό Βήμα [Νοmiko Vima] 197. In one case the Council of State did not respect the decision of the ECJ, concluding, when the case returned to it, that the applicant did not have a legal interest in raising the specific legal argument. 73 See for example CoS 989/1987, 1093/1987, 3190/1988, 3785–3789/1990, 1545/1995, 2176/1996, 308/1998, 4674–5/1998 and 3457/1998. 74 For example, CoS 3502/1994 and CoS 249/1997. 75 Gerapetritis, ‘Council of State vis-à-vis the ECJ’ (n 46) 83 f. 76 CoS (Plenary) 3670/2006, 08 December 2006. 77 Cos (Section D) No 778/2007 (Section D), 13 March 2007. 78 Art 14 (9) GC: ‘(a) The ownership status, the financial situation and the means of financing of information media must be made known as specified by law; (b) The measures and restrictions necessary for fully ensuring transparency and plurality in information shall be specified by law; (c) The concentration of the control of more than one information media of the same type or of different types is prohibited; (d) More specifically, concentration of more than one electronic information media of the same type is prohibited, as specified by law; (e) The capacity of owner, partner, major shareholder or managing director of an information media enterprise, is incompatible with the capacity of owner, partner, major shareholder or managing director of an enterprise that undertakes towards the Public Administration or towards a legal entity of the wider public sector to perform works or to supply goods or services; (f) The prohibition of the previous

Greece  269 provision prohibits the award of public works contracts to undertakings whose shareholders, executives or their spouses are also shareholders or executives of media undertakings. The question raised here concerning the relationship between national, ie Constitutional, and EU law is whether excluding a bidder from a public works contract on the grounds that s/he holds media interests is contrary to EU public procurement law. The problems of this controversial constitutional provision are summarised in the consecutive adoption of three executing laws, each one replacing the former,79 a series of judgments by the Council of State and the Court of Audit, as well as the notification of a reasoned opinion from the European Commission to the Greek Government (27.04.2005). This opinion determined a deadline of three weeks in order to ensure compliance with the provisions of EU law and avoid initiating the procedure of Article 226 (2) EC Treaty (now Article 258 TFEU) that would bring the Greek Government before the European Court of Justice. Not only did the legislative specification of this constitutional provision prove either to be incompatible with secondary EU law or to raise questions of constitutionality. But the constitutional provision itself also raised concerns over its conformity with EU law. On the basis of the above-mentioned provisions, the Michaniki company, an unsuccessful tenderer for a works contract, brought an action before the Council of State for the annulment of the certificate which the National Radio and Television Council had issued to the main shareholder of the Pantechniki company, the successful tenderer for the contract at issue. The challenged part of the certificate stated that there was no incompatibility (according to Article 3 of Law 3021/2002), despite the fact that the son of that shareholder was an executive of two Greek media companies. In judgments 3242 and 3243/2004 of the Council of State, the majority held that the incompatibility provisions of Article 14§9 GC did not infringe Article 24 of Directive 93/37, since the meaning of the Directive was clear and it did not relate to the exemption of media professionals etc. from public works contracts referred to in Article 14§9 GC. Therefore, the regulatory domains of the two provisions did not overlap and, thus, there was no conflict between the national Constitution and EU law. The Plenum of the Council of State, however, in its 3670/2006 judgment, referred questions to the ECJ concerning the interpretation of Directive 93/37/EC on public works contracts, with a view to assessing the compatibility of the constitutional provision at issue with EU law. At the same time, it accepted that Greek judges were obliged to scrutinise the conformity of a Constitutional norm with EU law (including secondary law). In doing so, it is submitted80 that the CoS implied the federal principle. A first, crucial question in this case was how to define the concepts of ‘principal shareholder’ and ‘intermediary persons’ in such a way that the prohibition deriving from Article 14§9 GC would not impermissibly broaden the reasons which, according to the relevant EC directives, lead to the exclusion of candidates from bidding for public contracts. The ECJ maintained that EU law did not preclude the option for Member States to maintain or adopt additional rules in order to ensure the principle of equal treatment and transparency, as long as those provisions were section extends also over all types of intermediary persons, such as spouses, relatives, financially dependent persons or companies; (g) The specific regulations, the sanctions, which may extend to the point of revocation of the license of a radio or television station and to the point of prohibition of the conclusion or to the annulment of the pertinent contract, as well as the means of control and the guarantees for the prevention of infringements of the previous sections, shall be determined by law.’ 79 Laws 3021/2002, 3310/2005 and 3414/2005. 80 Ioannis Tassopoulos, ‘Between Constitutionalism and Legal Pluralism: Perspectives from Greece’, in G Tusseau (ed), Debating Legal Pluralism and Constitutionalism: New Trajectories for Legal Theory in the Global Age, (Cham, Springer, 2020) 229, 234.

270  Lina Papadopoulou proportionate to achieving their objective. In the light of the foregoing, the ECJ concluded that a national provision such as the one at issue, which introduced general incompatibility between the public works sector and the media sector without allowing those involved any opportunity to show that, in their case, there had been no favouritism, was disproportionate and therefore unacceptable under EU law. In his opinion, which was tolerant towards national specificities, Advocate General (AG) Poiares Maduro based his reasoning (paragraphs 31–33) on the EU’s fundamental obligation to respect the national constitutional identity of Member States, since European integration should not undermine the maintenance of their political entity. The AG added that ‘[i]t is, nevertheless, necessary to point out that that respect owed to the constitutional identity of the Member States cannot be understood as an absolute obligation to defer to all national constitutional rules’.81 Partly following the AG, the ECJ accepted that Community law does not preclude the adoption of national measures designed to avoid, in procedures for the award of public works contracts, the risk of occurrence of practices capable of jeopardising transparency and distorting competition, a risk which could arise from the presence, amongst the tenderers, of a contractor active in the media sector or connected with a person involved in that sector, and thus to prevent or punish fraud and corruption.

However, it also added that such measures should be compatible with the principle of proportionality, something which does not apply to the provision at issue, which established a system of general incompatibility between the public works and media sectors and excluded ‘from the award of public contracts public works contractors who are also involved in the media sector on account of a connection as owner, main shareholder, partner or management executive, without affording them any possibility of showing, with regard to any evidence advanced, for instance, by a competitor, that, in their case, there is no real risk’ for transparency and equality.82 Following the ECJ judgment, the Council of State issued its final decision (3471/2011),83 which affirmed the unconstitutionality of the relevant statutory provision on the basis of an interpretation of the Constitution in accordance with EU law, pursuant to the ECJ judgment. Despite the fact that an attempt was made to support an interpretation of Article 14§9 GC that would be compatible with EU law,84 it is obvious that tension has arisen between the constitutional provision and the EU Directives on the award of public contracts. This tension could lead to further interpretative problems or even the degradation of the symbolic and normative function of the constitutional text. In practice, the constitutional provision has been inactivated. b.  The Recognition of Professional Qualifications Based on Academic Titles Acquired in Non-public Universities A second case of tension between the Constitution and EU law is that concerning the recognition of professional rights to holders of academic titles issued by institutions of tertiary education established in Greece and linked to non-Greek European universities. The issue of (non-)recognition was closely related to the constitutional prohibition of the foundation of

81 ECJ 08.10.2008 C-213/07 (Michaniki AE v Ethniko Symvoulio Radiotileorasis and Ypourgos Epikrateias) Opinion AG Poiares Maduro, ECLI:EU:C:2008:544 paras 30–33. 82 ECJ 16.12.2008 C-213/07 (Michaniki AE v Ethniko Symvoulio Radiotileorasis and Ypourgos Epikrateias) ECLI:EU:C:2008:731 paras 60, 62. 83 CoS (Plenary) 3471/2011 (plenary session), of 04 November 2011. 84 Venizelos, ‘Relationship’ (n 53).

Greece  271 private universities in accordance with Article 16§§5 and 8 GC. Universities in Greece are fully self-governing public law legal entities that operate under the supervision of the state and are entitled to financial assistance from it.85 Consequently, no academic titles could be recognised by the competent Greek authority if issued on the basis of studies that had been completed in tertiary educational institutions established in Greece which were not legal entities of public law, even if the titles had been awarded by cooperating European universities established in other EU Member States. In its established case law,86 the Council of State kept rejecting applications for the recognition of academic titles, despite the fact that sufficient doubts existed to require a reference to the ECJ. It eventually referred the issue to the latter, basing its case on a distinction between the academic and the professional rights emanating from academic titles. CoS 2809/199787 is worth noting, since it was exemplary in this issue. It explicitly proclaimed the supremacy of the Constitution over EU law and ruled that the application of the Directive 89/48/EEC on the recognition of higher education diplomas88 would entail the ‘supremacy’ of EU law over the Constitution, which the latter did not allow for. It added that the Parliament, when it ratified the Accession Treaty, had neither constituent nor revisionary power, so it could not have altered this limitation. It is no surprise that the Commission had repeatedly cautioned Greece with action before the CJEU for not recognising degrees awarded by private educational institutions operating and established in Greece that collaborated with European universities as equivalent to those of Greek universities and hence as university degrees.89 Educational policy was, according to Article 149 EC Treaty (and still is based on Article 165 TFEU, in combination with Article 6 [e] TFEU) an issue of intergovernmental cooperation, demarcating the competence of the EU on the basis of the principle of subsidiarity. Nevertheless, EU law had become relevant on the basis of Articles 39 ff, 43 ff (especially Article 47) and 49 ff of the TEC (now Articles 45 ff, 49 ff and 56 ff TFEU). These stipulations regulate the mutual recognition of diplomas, certificates and other academic titles. Two developments also formed a new framework of debate: on the one hand, the judgment of the ECJ in the Valentina Neri v Italy case,90 and, on the other hand, Directive 2005/36/EC on the recognition of professional qualifications, which clearly substantiated the ambiguous positions expressed by the Court in the Neri case. There is no doubt that the controversial Directive standardised the scope of application, the procedure, and the results of the recognition of professional qualifications, leaving it up to every Member State to verify the following: a) that the education offered is officially certified by the educational institution of the Member State of origin; b) that the awarded titles are identical to those that would have been awarded if the education had been provided in the country of origin of the title, and c) the extent to which each title provides the same professional rights as those within the territory of the Member State that awards it. Through the Community’s intervention in mutual recognition of academic titles for professional purposes, titles awarded by private institutions of post-secondary (especially higher)

85 Cf Andreas Pottakis, ‘Higher Education Reform in Greece: Another Missed Opportunity?’, 14 European Public Law (2008) 507. 86 See for example CoS 1680/1999, 2274/1990, 2809/1997 and 3457/1998, 1440/2000, 2667/2003, 2392/2004, 2709/2004, 2716/2005, 907/2005 1841/2006. 87 The same reasoning is also to be found in CoS 2807–2809/1997. 88 Directive 89/48/EEC of 21 December 1988 on a general system for the recognition of higher-education diplomas awarded on completion of professional education and training of at least three years’ duration. [1989] OJ L 19/16, adopted through Presidential Decree 52/1993 (GG Α 20). 89 Pottakis, ‘Higher Education Reform’ (n 82) 514. 90 ECJ 13.11.2003 C-153/02 (Valentina Neri v European School of Economics) ECLI:EU:C:2003:614.

272  Lina Papadopoulou education could under certain conditions, in principle through the system of university ­franchising, be recognised as ‘professionally equal’ to those awarded by state institutions of higher education. Moreover, although Directive 2005/36/EC and the Neri case are not prima facie opposed to the constitutional provision, taking into consideration also the limited EU competence in educational matters, they indirectly force the constitutional legislator to adopt a differentiated reading, if not a revision of the Constitution in order to address the indirect effects of EU law. Judgment CoS 778/2007 (Section D) marks the turning-point in the jurisprudence of the CoS in the matter under consideration. It admitted that the recognition of a diploma awarded by a foreign university (in this case Italian) to Greek citizens was compulsory, despite the fact that the studies had been carried out partly in a branch established in Greece, which was not recognised as an educational institution. This obligation derived from both primary – referring to the free movement of workers – and secondary EU law, especially Directive 92/51,91 whose provisions were listed in detail in the decision. This U-turn can be attributed to the effect of the ECJ’s judgment in the Neri case and the possibility of action by the European Commission against Greece on the issue of non-recognition. The ECJ adjudicated on the issue and concluded that Directive 89/48 did not ‘contain any limitation as regards the Member State in which an applicant must have acquired his professional qualifications’ and that ‘it is sufficient that the education and training were received “mainly in the Community”’. Consequently, a host Member State is obliged … to recognise a diploma awarded by an authority of another Member State even if that diploma is awarded on completion of education and training received, in whole or in part, in the host Member State and even if, according to the legislation of that State, that education and training is not recognised as higher education.92

This judgment conforms to the strategy of avoiding direct conflicts between EU law and the national Constitution. The organisation of education without connections with educational institutions in other Member States is not affected by EU law. In other words, the state monopoly on higher education in Greece is not formally challenged. The issue is resolved in a way that does not allow the conclusion that EU law, primary or secondary, requires or favours the revision of Article 16§§5, 7 and 8 GC. It is worth noting that the issue of preserving the monopoly on state universities has also attracted considerable attention from the wider public. This fact may explain the self-restraint of the ECJ and its differentiated and more tolerant stance towards Article 16 GC, compared with its counterpart in the ‘principal shareholder’ case.93 Nevertheless, in practice the monopoly has lost its practical importance. This is because it remains a tipping point only for academic titles, in that it does not allow the holder of the title to proceed with further studies in a public Greek university, but has no meaning for professional purposes. It undermines Article 16§§5 and 8 GC, without obliging the Greek legislator to amend it, though it almost renders it devoid of meaningful content. EU law has, thus, caused an implicit change in the relevant constitutional provisions. It did so by changing its field of application and excluding from it non-public educational institutions established in Greece.

91 It is worth noting that the CoS interpreted Directive 92/51/EEC in almost the same way as ECJ 19.04.2007 C-274/05 (Commission v Greece) Opinion AG Bot, ECLI:EU:C:2007:225, concerning Directive 89/48/EC. 92 ECJ 23.08.2008 C-274/05 (Commission of the European Communities v Hellenic Republic) ECLI:EU:C:2008:585, paras 28, 35, concerning failure of a MS to fulfil obligations resulting from Directive 89/48/EEC regulating the free ­movement of workers and more specifically the recognition of diplomas. 93 Pottakis, ‘Higher Education Reform’ (n 82) 513.

Greece  273 This way EU law influences Greek courts’ margin of appreciation in interpreting Article 16§§5 and 8 GC. Admittedly, EU law is pushing only gently towards the recognition of the operation of private higher education institutions. Nevertheless, in the last decade there has been no political consensus for such a change, and it is unlikely that such will form in the foreseeable future. For the revisionary legislator the dilemma is not whether to recognise the possibility of founding private higher education institutions, but rather whether to proceed to the consolidation of detailed constitutional conditions and limitations concerning their operation, or to transfer the related competence to the common legislator. Taking into account the importance of achieving, to the extent possible, the homogeneity of academic guarantees that regulate university institutions, either private or public, it is evident that the constitutional legislator must explicitly consolidate academic freedom and state supervision over private universities. However, it is doubtful whether the constitutional legislator is in a position to make the non-profit-making character of private universities a condition for their operation, especially in view of the content of the Directive. c.  Employment Contracts of Limited Duration in the Public Sector There seems to be a conflict between Article 103 GC, on the one hand, and Council Directive 1999/70/EC,94 especially clause 5 of its Annexed ETUC-UNICE-CEEP framework agreement, aiming to ‘prevent abuse arising from the use of successive fixed-term employment contracts or relationships’ on the other. Specifically, since the completion of the constitutional revision of 2001,95 Article 103§8 GC prohibits limited duration contracts of private law employment in the public administration and in the wider public sector to be converted to contracts for an indefinite period on the basis that they cover fixed and permanent needs. This constitutional provision (Article 103§8 GC) was added to the Constitution via its 2001 revision, while the transposition of Council Directive 1999/70/EC was still pending. Such a conversion had been possible and used to take place, until the said constitutional revision, based on Law 2112/1920 functioning as an ‘equivalent legislative measure’ in the sense of the aforementioned Clause 5. Due to the new constitutional provisions, however, no more scope had been left either for the legislator or for the courts to assess such contracts and classify the legal relationship properly, as contracts of indefinite duration, as such conversion of staff in the public sector to permanent civil servants would be in breach of the prohibition enshrined in Article 103 §8 GC, as revised.

94 Directive 1999/70/EC of 28 June 1999 concerning the framework agreement on fixed-term work concluded by ETUC, UNICE and CEEP, [1999] OJ L 175/43. 95 In 2001 paras 7 and 8 were added to Art 103 GC, worded as follows: ‘7. The engagement of employees in the public administration and in the wider public sector … shall take place either by competitive entry examination or by selection on the basis of predefined and objective criteria, and shall be subject to review by an independent authority, as specified by legislation. … 8. The legislation shall specify the conditions and duration of private law employment relationships in the public administration and in the wider public sector, as this defined in each case, in order to fill either permanent posts other than those provided for in the first section of para 3, or posts to meet temporary, unforeseen or urgent needs, within the meaning of the second section of para 2. The legislation shall also specify the duties that may be undertaken by the staff mentioned in the preceding sentence. Conversion by legislation of staff covered by the first section to permanent civil servants or conversion of their contracts into contracts of indefinite duration is prohibited. The prohibitions of the present paragraph shall also apply to those employed on the basis of a contract for performance of a specific task.’

274  Lina Papadopoulou The Directive regarding the public sector was transposed into national law by Presidential Decree 164/2004. The adaptation of the domestic law to the EU law was made with respect to Article 103§8 GC. However, since the provisions of the abovementioned Presidential Decrees entered into force on 19 July 2004, transitional provisions were laid down for the earlier period. According to these provisions, fixed-term contracts or work contracts constitute contracts of indefinite duration if they (1) were concluded prior to the entry into force of the Presidential Decree and were active until its entry into force, (2) lasted at least two years after renewal, and (3) concern activities related to the fixed and permanent needs. The subsequent conflict between the constitutional provision and the provisions of Directive 1999/70/EC was resolved in favour of the Constitution by the case law. Relevant judgments of the Court of Cassation (AP, the Supreme Civil and Penal Court) tend to accept that the Constitution prevails over (older) secondary EU law and, thus, the fixed-term contracts or works contracts concluded after 17 April 2001 (date of completion of the revision), may no longer be considered as contracts of indefinite duration.96 This legal reasoning of AP97 reinstates the supremacy of the Constitution. There is, however, a minority opinion in a relevant judgment, which insists on the primacy of EU law over the relevant statutory and (tacitly also) constitutional stipulations.98 However, the CJEU took a different stance when requested by a first instance (one judge) Court.99 In a 2021 judgment100 the CJEU concluded (para 53) that Clause 5(1) of the framework agreement must be interpreted as meaning that, where an abuse of successive fixed-term employment contracts, within the meaning of that provision, has occurred, the obligation incumbent on the referring court to undertake, so far as possible, an interpretation and an application of all the relevant provisions of domestic law capable of duly penalising that abuse and of eliminating the consequences of the infringement of EU law, extends to the application of a provision of national law that permits the conversion of the succession of fixed-term contracts to one employment contract of indefinite duration

– even contrary to a constitutional provision that absolutely prohibits, in the public sector, such a conversion.101 In this way, the CJEU tried to guarantee the full efficacy of EU law and, at the same time, its primacy over national law, including the national Constitution.

96 AP 1872/2017 (Section B), 126. 244/2015. However, according to the case law of the CoS, the Constitution itself tolerates (based also on Art 108 §8 GC, which foresees a transitional period) the settlement of the ongoing employment relationships in the public sector concerning employees that even after the Revision of the Constitution were serving based on successive fixed-term contracts ignoring the rights they could derive from Directive 1999/70 / EC, in view of the delay in adapting Greek legislation to this Directive. See the CoS judgments 1253/2006 (Section C), 441/2007, 2228–9/2007, 31/2008, 2718/2010, 2575/2011, 639/2014. 97 AP 1401/2017, 3360, 3423/2015, 71,410/2014 1154/2012, 123/2012, 16/2012. 98 In its decision 63/2018 (Section B2) of 10 January 2018 the AP ruled in favour of the conversion of the employment relations with limited duration to such with indefinite duration since the first contracts had already been signed before the promulgation of the amended Constitution of 2001. Interestingly enough there is a concurrent one-judge minority opinion coming to the same conclusion based on the ‘supremacy’ of EU law resulting in the national Courts’ obligation to safeguard full effectiveness of EU law and interpret national law in accordance with Community law. 99 E Πρεβεδούρου [E Prevedourou] ‘Οι συμβάσεις ορισμένου χρόνου στον δημόσιο τομέα μεταξύ υπεροχής του ενωσιακού δικαίου, συνταγματικής απαγόρευσης και σύμφωνης ερμηνείας: Με αφορμή την απόφαση Δ.Ε.Ε. 11.2.2021, C-760/18, M.Β. κ.λπ. κατά O.T.A. «Δήμος Αγίου Νικολάου»’ [‘Fixed term employment contracts in public sector between supremacy of EU law, constitutional prohibition and interpretation in conformity: in the occasion of CJEU 11.2.2021, C-760/18, M.V. and Others v Organismos Topikis Aftodioikisis (O.T.A.) «Dimos Agiou Nikolaou»’], (2021) 80 (4) Επιθεώρησις Εργατικού Δικαίου, 389. 100 CJEU 11.02.2021 C-760/18 (M.V. and Others v O.T.A. «Dimos Agiou Nikolaou») ECLI:EU:C:2021:113. 101 In accordance with CJEU 04.07.2006 C-212/04 (Adeneler and Others) EU:C:2006:443, when the Court did not find (para 102) ‘a general obligation on the Member States to provide for the conversion of fixed-term employment

Greece  275

3.  Limits to European Integration Outside the EU Legal Order Whereas limits to European integration and transfer of constitutional competences fall under Article 28§3 GC (see above), EU integration outside the EU legal order, as long as it is designed through international law treaties, does not enjoy primacy over the Greek Constitution – at least according to the prevalent view. Thus, ideal-typically at least, and no matter how improbable that might be, these treaties may be scrutinised by Greek courts based on all constitutional provisions. In the unlikely case of non-application of a treaty, of course, the state would be internationally liable. Even more so, courts have proven to be very self-restrained when it comes to economic measures – and rightly so! Courts have not used the Constitution to undermine the application of European measures, formally structured either as EU or as international law, especially under the circumstances of financial distress, like that which Greece has been experiencing the last 10 years. The point which has – more than others – been extensively discussed in theory is Article 3 § 2 of the Treaty on Stability, Coordination and Governance (TSCG). It obliged Member States to introduce the balanced budget amendment (often called a ‘golden rule’) to their national legal order with provisions of binding force and permanent character, preferably constitutional, or otherwise guaranteed to be fully respected and adhered to throughout the national budgetary processes. What was problematised here was the fact that an international law obligation concerned the (positive) amendment of the Constitution itself. However, as already mentioned above, Article 28 GC foresees the supra-statutory status of international law treaties, and budgets are statutory laws. Hence, besides the fact that this obligation has been dropped at the end, a constitutional amendment was not necessary anyway.102

V.  Constitutional Rules and/or Practice on Implementing EU Law A.  Binding (Parliamentary or Governmental) Resolutions for the Ministers in the Council of the EU? There are no binding parliamentary or governmental resolutions for the Ministers in Council of the EU. Ministers often informally communicate with the Prime Minister when they are expected to make decisions of great importance. This is because it is the Prime Minister who bears the principal political responsibility and he is the one who has chosen the Ministers and may dismiss them if they act contrary to the Government’s policies.

contracts into contracts of indefinite duration’, but (para 105) ‘in circumstances such as those of the main proceedings, the Framework Agreement is to be interpreted as meaning that, in so far as domestic law of the Member State concerned does not include, in the sector under consideration, any other effective measure to prevent and, where relevant, punish the misuse of successive fixed-term contracts, the Framework Agreement precludes the application of national legislation which, in the public sector alone, prohibits absolutely the conversion into an employment contract of indefinite duration of a succession of fixed-term contracts that, in fact, have been intended to cover “fixed and permanent needs” of the employer and must therefore be regarded as constituting an abuse.’ 102 On this issue see more extensively, Lina Papadopoulou, ‘Can constitutional rules, even if “golden”, tame Greek public debt?’, in M Adams, F Fabbrini and P Larouche (eds), The Constitutionalization of European Budgetary Constraints (Oxford, Hart, 2014) 223.

276  Lina Papadopoulou

B.  Participation of the Parliament in the Making of Secondary EU law (Principle of Subsidiarity) According to Article 70§8 GC, ‘[t]he Standing Orders of Parliament shall specify the manner in which the Parliament is informed by the Government on issues being the object of regulation in the framework of the European Union, and debates on these’. Article 41B of the Standing Orders (‘Opinions on regulatory acts of the European Union’) specifies the constitutional provision, stipulating that the Government should address any draft EU legislation to the Parliament’s Chairperson, as soon as it is communicated to the Council of Ministers, as well as consultation documents. The Chairperson should then call a meeting of the European Affairs Committee (EAC), probably together with the other competent Committees. The EAC was set up in 1990 and is the most competent body to rule on subsidiarity issues. Moreover, it has competence over a) EU institutional issues, b) issues concerning the relations between the Hellenic Parliament and other EU Member State parliaments, the European Parliament and the COSAC, and c) European policies and EU legislative acts. According to Article 32A of the Standing Orders, the EAC is composed of 30 deputies, chaired by one of the Vice Presidents of the Parliament, and its composition should be proportionally representative of all the political groups in the Parliament. Greek Parliamentarians may also participate in its meetings, by right of speech only. The Committee normally meets jointly with members of the Standing Parliamentary Committee. The EAC is only called to meetings irregularly by its Chairperson, or by request of one third of its members, or of the Government. In addition, its Chairperson holds the initiative of choosing the agenda items. The EAC’s decisions are final and do not require referral to the Plenary. It may also invite Ministers for hearings. It can deliver their opinion and address it to the Parliament, to the competent Ministers, and the EU institutions. The Committee’s opinions may be submitted for debate and vote in the Plenary upon request of the Speaker or the competent Committees. The Parliament has rarely taken the opportunity to enter into political dialogue with the Commission103 or to raise objections to draft EU legislation based on the principle of subsidiarity and send reasoned opinions104 to the European Commission.105 In the period 2006–07 the Parliament expressed and addressed to the Commission no opinions at all, while in 2008 it expressed three,106 in 2009 10,107 in 2010108 and in 2011 four,109 in 2012 six110 (though none of

103 Within the framework of the political dialogue, national Parliaments can send opinions to the Commission, which endeavours to reply within three months. Opinions can concern Commission documents or policy areas for which the Commission is competent. 104 To qualify as a reasoned opinion, according to the definition in Protocol No 2, an opinion must clearly state a breach of subsidiarity and be sent to the Commission within eight weeks of the transmission of the proposal to national Parliaments. 105 See the Annual Reports of the Commission on Relations between the European Commission and National Parliaments at http://ec.europa.eu/dgs/secretariat_general/relations/relations_other/npo/political_dialogue_en.htm. 106 Annual Report 2008 on Relations between the European Commission and National Parliaments, 07.07.2009, COM(2009) 343 final. 107 Annual Report 2009, 02.06.2010, COM(2010) 291 final. 108 Annual Report 2010, 10.06.2011, COM(2011) 345 final. 109 Annual Report 2011, 10.07.2012, COM(2012) 375 final. 110 Annual Report 2012, 30.07.2013, COM(2013) 565 final.

Greece  277 the opinions in the period 2010–12 was a reasoned opinion), in 2013 three reasoned opinions (out of four opinions altogether),111 and in 2014 no opinions at all.112

C.  Implementation of Secondary Law The implementation of EU Directives and their reception into the national legal order may take place through a statute (formal law) voted upon by the Parliament, or a normative act issued by an institution of the executive (eg Presidential Decree or Ministerial Decision, in most cases a joint one, the Minister for National Economy signing in most cases together with the competent Minister). In order for such normative acts to be issued, prior legislative authorisation is needed, according to Article 43 GC. A key tool for incorporating Directives continues to be Law 1338/1983 (on implementation of Community law), as amended, which enables the use of existing or ad hoc authorisations for the issue of ministerial decisions and presidential decrees. The choice of instrument is vested with the Minister who is responsible for each EU legal act, depending on the material regulated thereby. The Ministry is also obliged to draft a table matching the provisions of the Directive with those of the national legal act. Measures that are necessary for the implementation of Regulations, as long as this is provided for by the Regulations themselves, are implemented either through presidential decrees or through ministerial decisions. In the case of implementation through formal law by the Parliament, the Ministry has to draft a statute which needs to be accompanied by (a) a report on the results of the public deliberation which takes place online, (b) a report by the State General Accounting Office (Geniko Logistirio Kratous) concerning the fiscal consequences of the statute, according to Article 75§3 GC and (c) the opinion of the Court of Audit in accordance with Article 73§2 GC (in case of a bill granting pensions). In cases where a Directive is implemented through a Presidential Decree, it must be noted that the latter must be scrutinised by the 5th Section of the Council of State for its legality, otherwise the Decree is void. The administration has to take into consideration and incorporate the remarks noted in the Council’s opinion (Praktiko Epexergasias). Otherwise the President may not issue the Decree. After they have been voted upon and issued, all acts (laws, presidential decrees, ministerial decisions) need to be published by the President in the official Government Gazette.

111 Two reasoned opinions may be noted, both submitted in 2013: the first one concerned COM (2013) 151 on ‘Conditions of entry and residence of third-country nationals for purposes of research, studies, pupil exchange, training, voluntary service and au pairing’ as against Art 79§5 TFEU. The second one concerned COM (2012)788 on ‘Approximation of the laws, regulations and administrative provisions of the Member States concerning the manufacture, presentation and sale of tobacco and related products’. In both cases the Commission replied to the Hellenic Parliament’s objections, which were based on alleged non-compliance with the principle of subsidiarity, purporting that these objections were not valid, Annual Report 2013, 05.08.2014 COM(2014) 507 final. 112 However, a number of objections were raised by many Parliaments, amongst them the Greek one, about the Tobacco Products Directive, concerning the number and content of the provisions empowering the Commission to adopt delegated acts in accordance with Art 290 TFEU, the size of the health warnings on packages and the proposal to ban slim cigarettes. See Annual Report 2014, 02.07.2015 COM(2015) 316 final.

278  Lina Papadopoulou

D.  Enforcement Through the Courts, Including the Safeguarding of the EU Charter of Fundamental Rights In increasing, though still limited, numbers, litigants have been using arguments based on EU law in order to challenge legislation in EU law-related fields. Courts evaluate these arguments and the judges – especially those of the higher courts – have gradually been acquiring better knowledge of EU law. The entry into force of the Treaty of Lisbon and the Charter of Fundamental Rights of the European Union (the Charter) coincided with the outbreak of the public debt crisis in Greece. Austerity measures are often challenged before the courts and litigants also invoke the Charter, given the relevance of EU law to the Memoranda of Understanding, which the measures are designed to implement. Within this context, some examples are indicative: In CoS (Section B) 4788/2014 the Court dismissed the applicants’ allegations, claiming that the increase of the cost for a lawsuit (from 10 to 100 euros) and for an action in a penal court (from 10 to 50 euros) did not violate Article 47 of the Charter (nor Article 6 ECHR or Article 20 GC), since it respected the principle of proportionality. CoS judgment 1254/2015 dealt with the constitutionality and conventionality – including Article 16 of the Charter and Directive 98/59/EC – of a statutory provision regulating collective redundancies. In this case the Council of State decided to stay proceedings and make a preliminary reference to the CJEU. The reference concerned mainly the compatibility of the prior authorisation of the redundancies by the administrative authorities with the Directive 98/59 in particular and, more generally, Articles 49 and 63 TFEU. The CJEU stated that national legislation which provides that a national authority endowed with powers either to approve or project redundancies on grounds relating to the protection of workers and of employment, cannot be considered contrary to freedom of establishment as guaranteed by Article 49 TFEU or the freedom to conduct a business enshrined in Article 16 of the Charter.113 However it pointed out that the three criteria, which the competent public authority is called upon to take into account for the purpose of deciding whether it opposes collective redundancies, are formulated in very general and imprecise terms. This, according to the CJEU, results in severe interference with the freedom to conduct a business. Moreover, in several criminal cases the accused invoked114 – sometimes successfully – the principle of ne bis in idem guaranteed in Article 50 of the Charter, in order to avoid a second conviction by a Greek court. Articles 50 and 52(1) of the Charter have the same normative content as Article 4 of the 7th Additional Protocol to the ECHR.115 More specifically, the ‘­criminal’ character of administrative sanctions is assessed on the basis of criteria similar to the Engel criteria adopted by the ECtHR.116 The CJΕU rules that the accumulation of penalties constitutes a violation of the principle of ne bis in idem, which may be justified under certain conditions.117 It is necessary that parallel procedures and penalties do not go beyond what is appropriate and ­necessary to achieve the objectives pursued by the legislature of combating tax evasion and

113 CJEU 21.12.2016 C-201/15 (AGET Iraklis) ECLI:EU:C:2016:972. 114 See for example Appeal Court of Athens (Penal) 674/2011 and 4703/2011, AP (Penal) judgments 1568/2010 111/2014, 1125/2014. The judges often analyse the terms and conditions under which the principle ne bis in idem applies within the Union after the entry into force of the Treaty of Lisbon and the Charter. 115 CoS 951/2018, 1102–1104/2018. 116 ECtHR (Plenary), Engel and Others v The Netherlands, (Application no 5100/71; 5101/71; 5102/71; 5354/72; 5370/72), 8 June 1976. 117 CJEU 27.05.2015 C-129/14 PPU (Spasic) ECLI:EU:C:2014:586.

Greece  279 collection of taxes and/or duties, since there is either an irrevocable acquittal of a criminal court, finding that the tax offence at issue did not occur,118 or an irrevocable conviction by a criminal court, imposing a penalty capable of suppressing the infringement committed in an effective, proportionate, and dissuasive manner.119 It is worth mentioning that the AP in its judgment in Plenary 1/2011 accepted that res judicata had acquired a transnational force in the EU. In the same judgment the AP accepted that if the sentence for a crime has in fact been served, this is no longer a prerequisite for denying the execution of a European Arrest Warrant.120

VI.  Resulting Relationship between EU Law and National law Greece’s accession to the EC and the reception of subsequent EU Treaty amendments was made possible under Article 28 GC. It allows for both the transfer of constitutionally vested competences to international organisations and the limitation of the exercise of national sovereignty. The latter presupposes that the political entities, eg the EU, in favour of which the limitations are allowed, preserve democracy and fundamental rights as their fundamental principles. Whereas EU law, like international law as a whole, enjoys primacy or precedence over statutes, its relationship with the Constitution has always been a topic of discussion in both legal doctrine and jurisprudence. In practice, Greek courts nowadays accept the primacy of EU law and attempt to interpret the Constitution in accordance with EU law. Two major cases, those of the ‘principal shareholder’ and the ‘private universities’, have confirmed this stance. Since the formal Constitution is relatively rigid, with a lengthy and cumbersome revision procedure – a rigidity that is enhanced by the constitutional culture121 – it was the material one that has been implicitly and tacitly changed in order to adapt to the demands placed by EU law. Consequently, besides an initial long-lasting wariness and despite some delay in accepting the primacy of EU law in marginal cases considered to fall under the ‘national constitutional identity’ (such as the recognition of university diplomas), the relationship between EU law and national law has been a rather smooth one from its very beginning in 1981 until today.

References T Antoniou, Europäische Integration und griechische Verfassung. Ein Beitrag zur Auslegung des Artikel 28 der griechischen Verfassung von 1975 (Frankfurt am Main, Peter Lang, 1985). Φ Αρναούτογλου [F Arnaoutoglou], ‘Συμβούλιο της Επικρατείας και Ευρωπαϊκά Δικαστήρια’ [‘Council of State and European Courts’], (2005) Νομικό Βήμα [Νομικό Vima] 197. F Arnaoutoglou, ‘Grèce: Le Conseil d’Etat’ in J Iliopoulos-Strangas (ed), Cours suprêmes nationales et cours européennes: concurrence ou collaboration? (Athens, Sakkoulas, 2007) 179.

118 CoS 1887/2018, 951/2018 (Section B), CJEU 20.03.2018 C-537/16 (Garlsson Real Estate SA and Others v Commissione Nazionale per le Società e la Borsa [Consob]) ECLI:EU:C:2018:193. 119 CJEU 20.03.2018 C-524/15 (Criminal proceedings against Luca Menci) ECLI:EU:C:2018:197. 120 Xenophon Contiades, Charalambos Papacharalambous and Christos Papastylianos, ‘The Constitution of Greece: EU Membership Perspectives’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (Berlin, Springer Verlag, 2019) 641, 664. 121 Xenophon Contiades and Alkmene Fotiadou, ‘Models of Constitutional Change’, in X Contiades (ed), Engineering Constitutional Change. A Comparative Perspective on Europe, Canada and the USA (Oxford/New York, Routledge, 2012) 417, 450.

280  Lina Papadopoulou X Contiades and A Fotiadou, ‘Models of Constitutional Change’, in X Contiades (ed), Engineering Constitutional Change. A Comparative Perspective on Europe, Canada and the USA (Oxford/New York, Routledge, 2012) 417. X Contiades and A Fotiadou, ‘The Hellenic Republic’, in L Besselink et al (eds) Constitutional Law of the EU Member States (Alphen aan den Rijn, Kluwer Law International, 2014) 718. X Contiades, C Papacharalambous and C Papastylianos, ‘The Constitution of Greece: EU Membership Perspectives’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (Hague, TMC Asser Press, 2019) 641. Π Δαγτόγλου [P Dagtoglou], ‘Ο Έλλην νομικός και το ευρωπαϊκό κοινοτικό δίκαιο’ [‘The Greek Lawyer and European Community Law’], (1976) 24 Νομικό Βήμα [Νομικό Vima] 929. Π Δαγτόγλου [P Dagtoglou], ‘Η συμμετοχή μας στις Ευρωπαϊκές Κοινότητες και το Σύνταγμα’ [‘Our Participation in the EC and the Constitution’], in Stangos and Moutsios (eds), Προσχώρηση και συμμετοχή της Ελλάδος στις Ευρωπαϊκές Κοινότητες. Η συνταγματική θεώρηση [Greece’s Accession to, and Participation in the EC: the Constitutional Viewpoint] (Thessaloniki-Athens, Sakkoulas, 1987) 19. D Evrigenis, ‘Aspects institutionnels et juridiques de l’élargissement. L’impact de l’adhésion sur les institutions et le droit des pays candidats: Grèce’, in W Wallace and I Herreman (eds), A Community of twelve. The impact of further enlargement on the European Communities (Bruges, De Tempel, 1978) 13. D Evrigenis, ‘Legal and Constitutional Implications of Greek Accession to the European Communities’, (1980) 17 CML Rev 157. A Fatouros, ‘International Law in the New Greek Constitution’, (1976) 70 The American Journal of International Law 492. Α Φατούρος [A Fatouros], ‘Διεθνείς πράξεις με σοβαρές επιπτώσεις στην άσκηση της εθνικής κυριαρχίας΄ [‘International Acts with Important Influence on the Exercise of National Sovereignty’], in K Ioannou et al, Δημόσιο Διεθνές Δίκαιο – Σχέσεις διεθνούς και εσωτερικού δικαίου [International Public Law. Relations between International and Domestic Law] (Athens, Sakkoulas, 1990) 85. G Gerapetritis, ‘The Hellenic Council of State vis-à-vis the European Court of Justice: From voluntary seclusion to inevitable constitutional dialogue’, in N Alivizatos et al (eds), Essays in honour of Georgios I. Kassimatis (Athens/Brussels, Sakkoulas/Bruylant, 2004) 83. K Gogos, ‘Die Diskussion über eine zukünftige Verfassung der Europäischen Union in Griechenland’, in H Timmermann (ed), Eine Verfassung für die Europäische Union: Beiträge zu einer grundsätzlichen und aktuellen Diskussion (Wiesbaden, Springer, 2001) 215. J Iliopoulos-Strangas, ‘Rechtsfragen der Mitgliedschaft Griechenlands in einer Europäischen politischen Union’, (1985) 20 Europarecht 199. J Iliopoulous-Strangas and E Prevedourou, ‘Le droit de l’Union européenne et les Constitutions nationales. Rapport Hellénique’, in FIDE XX Congress, Vol I (London, BIICL, 2002) 259. K Ioannou, ‘Greece’, in R Blackburn and J Polakiewicz (eds), Fundamental Rights in Europe. The European Convention on Human Rights and its Member States, 1950–2000 (Oxford, OUP, 2001) 358. G Katrougalos, The Constitutional History of Greece, in the Balkan Context, www.cecl.gr/RigasNetwork/ databank/reports/r1/GR_1_Katrougalos.htm. Α Μανιτάκης [A Manitakis], Ελληνικό συνταγματικό δίκαιο [Greek Constitutional Law] (Athens, Sakkoulas, 2004) 385. A Manitakis, ‘In Greece, Constitutional Changes Without Constitutional Revision’, Huffington Post, 14.03.2016, www.huffingtonpost.com. Α Μανιτάκης [A Manitakis], ‘Τα όρια της κοινοτικής αρμοδιότητας και η συνταγματική θεώρησή τους’ [‘The limits of the Community’s competence and their constitutional validation’], Το σύνταγμα 1984, 472. M Meng-Papantoni, ‘Legal Aspects of the Memoranda of Understanding in the Greek Debt Crisis’, (2015) 18 (3) Zeitschrift für Europarechtliche Studien, 15. İ Özden Kaboğlu and S-I Koutnatzis, ‘The Reception Process in Greece and Turkey’, in H Keller and A Stone Sweet (eds), A Europe of Rights. Assessing the Impact of the ECHR on National Legal Systems (Oxford, OUP, 2008) 451. Λ Παπαδοπούλου [L Papadopoulou], Εθνικό Σύνταγμα και Κοινοτικό Δίκαιο: Το ζήτημα της ‘Υπεροχής’ [National Constitution and Community Law: the Question of ‘Supremacy’] (Athens, Sakkoulas, 2009).

Greece  281 L Papadopoulou, ‘Die implizite Änderung der griechischen Verfassung durch das EU-Recht’, (2014) 74 Zeitschrift für ausländisches öffentliches Recht und Völkerrecht 141. L Papadopoulou, ‘Can constitutional rules, even if “golden”, tame Greek public debt?, in M Adams, F Fabbrini and P Larouche (eds), The Constitutionalization of European Budgetary Constraints (Oxford, Hart, 2014) 223. L Papadopoulou, ‘The 2015 Greek Referendum on Bailouts’, in J Smith (ed), Palgrave Handbook on European Referendums (Cham, Palgrave Macmillan, 2021) 625. Π Παυλόπουλος [P Pavlopoulos], ‘Το Σύνταγμα μεταξύ διεθνούς και Ευρωπαϊκού Δικαίου. Η ιεραρχία της έννομης τάξης’ [‘The Constitution between international and European law. The hierarchy of the legal order’], (2015) 10 (2) Εφημερίδα Διοικητικού Δικαίου, 154. A Pottakis, ‘Higher Education Reform in Greece: Another Missed Opportunity?’, (2008) 14 European Public Law 507. E Πρεβεδούρου [E Prevedourou] ‘Οι συμβάσεις ορισμένου χρόνου στον δημόσιο τομέα μεταξύ υπεροχής του ενωσιακού δικαίου, συνταγματικής απαγόρευσης και σύμφωνης ερμηνείας: Με αφορμή την απόφαση Δ.Ε.Ε. 11.2.2021, C-760/18, M.Β. κ.λπ. κατά O.T.A. «Δήμος Αγίου Νικολάου»’ [‘Fixed term employment contracts in public sector between supremace of EU law, constitutional prohibition and interpretation in conformity: in the occasion of CJEU 11.2.2021, C-760/18, M.V. and Others v Organismos Topikis Aftodioikisis (O.T.A.) «Dimos Agiou Nikolaou»’], (2021) 80 (4) Επιθεώρησις Εργατικού Δικαίου, 389. E Roucounas, ‘Le droit international dans la Constitution de la Grèce du 9 juin 1975’, (1976) 29 Revue hellénique de droit international 51. V Skouris, ‘Constitutional Disputes and Judicial Review in Greece’, in C Landfried (ed), Constitutional Review and Legislation (Baden-Baden, Nomos, 1988) 177. P Spyropoulos, ‘Die Vermutung der Verfassungsmässigkeit der Gesetze und die verfassungskonforme Gesetzesauslegung’, in G Kassimatis and M Stolleis (eds), Geschichte und Staatsrechtslehre. Griechisch-deutsche Wechselwirkungen (Frankfurt/Main, Klostermann, 2001) 283. P Spyropoulos and T Fortsakis, Constitutional Law in Greece (Alphen aan den Rijn, Kluwer Law International, 2009). I Tassopoulos, ‘Between Constitutionalism and Legal Pluralism: Perspectives from Greece’, in G Tusseau (ed), Debating Legal Pluralism and Constitutionalism: New Trajectories for Legal Theory in the Global Age (Cham, Springer, 2020) 229. Δ Τσάτσος [D Tsatsos], Συνταγματικό Δίκαιο Ι. Θεωρητικό θεμέλιο, [Constitutional Law I. Theoretical Foundations] (Athens, Sakkoulas, 1994). D Tsatsos, The European Sympolity: Towards a New Democratic Discourse (Brussels, Bruylant, 2009). Β Τζέμος [V Tzemos], Ο βασικός μέτοχος [The Basic Share-holder] (Athens-Komotini, Sakkoulas Ant, 2006). M Tushnet, Weak Courts, Strong Rights. Judicial Review and Social Welfare Rights in Comparative Constitutional Law (Princeton NJ, Princeton University Press, 2008). Ε Βενιζέλος [E Venizelos], Μαθήματα Συνταγματικού Δικαίου [Lectures of Constitutional Law] (Athens, Sakkoulas, 1991). E Venizelos, ‘The relationship between National Constitution and European Community Law after the signature of the Treaty for the European Constitution and the revision of the 2001 Greek Constitution’, (2005) 1 Hellenic Revue of European Law 1, 20, 23. C Yannakopoulos, ‘L’influence du droit de l’Union européenne sur le système de contrôle juridictionnel de constitutionnalité des lois: les paradigmes français et grec’, (2012) 91 Revue française de droit constitutionnel, 537.

282

9 Spain AIDA TORRES PÉREZ AND MARIBEL GONZÁLEZ PASCUAL*

I.  Main Characteristics of the National Constitutional System The 1978 Spanish Constitution (henceforth SC, or the Constitution) was the landmark of the transition process from dictatorship to democracy in Spain.1 The death of General Franco in November 1975 brought to an end the prolonged dictatorship that had been established after the Civil War in 1939. The Constitution was drafted by the Parliament that emerged after the June 1977 general elections and subsequently ratified through a referendum on 6 December 1978. The Constitution was the outcome of a broad political consensus with the aim to create a stable institutional framework to ground a democratic system and ensure the protection of fundamental rights. The main traits of the national constitutional system are: a parliamentary system, a model of political decentralisation in Autonomous Communities, a Constitutional Court, a bill of rights, and a rigid system of constitutional amendment. Spain is a parliamentary monarchy, in which the Head of the State is the King (or the Queen). The King performs a representative function and symbolises the unity of Spain, but lacks any actual political power. Although King Juan Carlos I was appointed by General Franco as his successor, he played a prominent role in Spain’s transition to democracy, which served as a source of legitimacy for him. In June 2014, King Juan Carlos I abdicated, amidst low popularity, and was succeeded by his son, King Felipe VI. The Spanish Parliament (Cortes Generales) is the institution that represents the people and is composed of two houses: Congress and Senate. As a result of the constitutional design, the Congress has a more prominent role than the Senate in the legislative process and the political life of Spain. The Congress consists of 350 deputies elected by the people. Under the Constitution, the electoral districts correspond to the Spanish provinces (50 in number), and the system is based on the principle of proportional representation. The electoral formula is the so-called ‘D’Hondt’ method. Although this is a proportional method, it has a majoritarian effect in small electoral districts. In Spain, 28 provinces elect fewer than five deputies, and 17 between six and nine. As a consequence, in practice the D’Hondt method favours the two largest political parties: the Socialist Party (PSOE) and the Popular Party (PP). To some extent, the two-party dominance was challenged in the 2015 and 2016 general elections, when two emerging political parties (Podemos and Ciudadanos) managed to get nearly 35 per cent of the seats. Corruption cases, the * The manuscript was closed in October, 2018. 1 For this section, see Víctor Ferreres Comella, The Constitution of Spain. A Contextual Analysis (Oxford, Hart Publishing, 2013).

284  Aida Torres Pérez and Maribel González Pascual economic crisis, and disenchantment with the traditional political parties fostered the support for the emerging ones, which promised a democratic regeneration from the left (Podemos) and the right (Ciudadanos). The Senate is defined as the chamber of territorial representation, but it does not effectively represent the Autonomous Communities. The electoral system for senators is a mixed one: there are senators elected by the people in each province (four senators per province); and senators appointed by the Parliament of each Autonomous Community (one senator per Autonomous Community, plus another one for every million residents). Moreover, the Senate has very few competences directly related to the territorial model. For instance, the Senate’s approval is needed in order to adopt measures to force an Autonomous Community to fulfil its obligations under the SC. However, this is an exceptional procedure that has only been used once, against Catalonia after the 2017 declaration of independence. Spain has a parliamentary system in which the Prime Minister is appointed by the Congress. To be appointed, the candidate needs to secure the support of an absolute majority of the deputies in the first round, or a simple majority in a second one. Adolfo Suarez, who won the 1979 general elections as leader of the Union of the Democratic Centre (Unión de Centro Democrático), was the first Prime Minister under the SC. Subsequent Prime Ministers have been leaders of the PP or the PSOE. Once appointed, the Prime Minister freely selects the ministers that form the Government. The Constitution provides a set of mechanisms for parliamentary control over the Government. Should the Prime Minister lose the confidence of the Congress, the Prime Minister would be removed. Oversight mechanisms, however, tend to be controlled by the parliamentary majority that supports the Prime Minister. The 2018 censorship motion was the first successful one. It followed the judgment in a major corruption case involving formed members of the PP (the so-called Gürtel case). As a result, Prime Minister Mariano Rajoy (PP) was forced to resign, and Pedro Sánchez (PSOE) replaced him. The model of territorial decentralisation of power was probably one of the most controversial issues in the drafting of the Constitution and it has been a continuous source of tensions since. The Constitution designed a quasi-federal system, the so-called ‘State of Autonomies’. Spain has 17 Autonomous Communities that enjoy political autonomy and have their respective Parliaments and Governments. The Constitution lists the competences that are reserved to the central state, while the Autonomous Communities may assume any other powers through the respective Statute of Autonomy. The Statute of Autonomy is the ‘constitution’ of each Autonomous Community and provides the foundations for its political organisation. The territorial model of political decentralisation is currently under pressure. This pressure resulted mainly from the secessionist movement in Catalonia. In order to expand and reinforce its autonomy, the Statute of Catalonia was amended in 2006. The amendment was challenged before the Constitutional Court, and in 2010 several articles were struck down. Since then, support for Catalonian independence has grown exponentially. In the 2015 autonomous elections, parties supporting independence obtained 47 per cent of the popular vote and an absolute majority in the Catalan Parliament. In October 2017, the Catalan Government organised a referendum that had not been authorised by the central Government. Subsequently, the Catalan Parliament declared Catalonia’s independence. This situation led the central Government to adopt exceptional measures under Article 155 SC forcing the Autonomous Community to abide by the Constitution. Furthermore, the Catalan Prime Minister and several members of the Government were accused of rebellion. The judiciary consists of a body of hierarchically ordered courts, with the Supreme Court at the top. Courts are structured in four subject matter jurisdictions: civil, criminal, labour,

Spain  285 and administrative courts. The Constitution provides for a specific body to secure the judiciary’s external independence, ie, the General Council of the Judiciary. The General Council is in charge of selecting and training judges, making judicial appointments, promotion, inspection and disciplinary sanctions. The Council is composed of 20 members, of whom 12 shall be judges, appointed by the Congress and the Senate. The Constitution set up a Constitutional Court to guarantee the supremacy of the Constitution following the Kelsenian model of centralised judicial review of legislation. The Constitutional Court comprises 12 members. Four are appointed by the Congress, four by the Senate, two by the Government, and two by the General Council of the Judiciary. The main functions of the Constitutional Court are the following. First, the Court performs an abstract review of legislation through direct challenges brought by political actors (recurso de inconstitucionalidad) or questions referred by judges (cuestión de inconstitucionalidad). Second, the Court has a very prominent function in securing the protection of fundamental rights through individual complaints (recurso de amparo). After exhausting remedies before ordinary courts, individuals may bring an action before the Constitutional Court for the violation of their fundamental rights. Finally, the Court safeguards the system of allocation of powers between the Central State and the Autonomous Communities (conflictos de competencia). The Constitution embodies a broad catalogue of fundamental rights in Title I, which includes civil, political, and social rights. Social rights, however, enjoy a different level of protection and effectiveness. They are included in Chapter III (Title I), entitled ‘guiding principles on social and economic policy’, and they are not directly enforceable. Instead, they can only be invoked before the courts in accordance with the laws that develop them. Article  10(2) SC opens the Constitution to international human rights law and sets forth that constitutional rights need to be interpreted according to the Universal Declaration of Human Rights and other human rights treaties ratified by Spain.2 In practice, the Constitutional Court has frequently referred to the European Convention on Human Rights (ECHR) and the corresponding case law as sources of interpretation. Finally, the Constitution is rigid when it comes to amendments. It lays down two different procedures requiring a broad consensus. The general constitutional amendment procedure requires a three-fifths majority in the Congress and the Senate (Article 167 SC). While referendum is not compulsory, if one-tenth of the deputies or the senators ask for it, then the amendment shall be ratified by referendum. An even more demanding procedure needs to be followed (Article 168 SC) in order to amend the clauses contained in the Preliminary Title (which includes the main structural principles of the constitutional system), the Crown, the most basic rights, or to perform a general constitutional amendment. This procedure requires a two-thirds majority in the Congress and the Senate, general elections, and a referendum. The Constitution has been amended only twice so far. Both amendments were prompted by the process of EU integration. The first amendment, in 1992, paved the way for ratification of the Maastricht Treaty. In particular, Article 13(2) SC was amended to include the right of foreigners to stand as candidates in local elections. Next, in 2011, in the midst of the economic crisis, Article 135 SC was amended to incorporate the balanced budget rule. It has been almost 40 years since the Constitution was enacted and there are increasing calls for an overall constitutional amendment. Such amendment is called in order to confront the current challenges faced by a political system that was born in the transitional period from dictatorship to democracy, as well as to renew the commitment of the people and the legitimacy of the Constitution. 2 Alejandro Saiz Arnaiz, La Apertura Constitucional al Derecho Internacional y Europeo de los Derechos Humanos. El Artículo 10.2 de la Constitución Española (Madrid, Consejo General del Poder Judicial, 1999).

286  Aida Torres Pérez and Maribel González Pascual

II.  Constitutional Culture Spain’s constitutional history is a turbulent one. This is due to complex and deep unrest as well as fractures within Spanish society,3 where the liberal democratic tradition is the minority. In this regard, the current Constitution was a turning point, not only because it brought about the longest and most serene democratic period in Spanish history, but also because the Constitution became a main player in the collective imagination and also a relevant tool for the profound change of the political system. The successful political transition was grounded in a reform of the legal system from within. The Constitution framers were aware of the need to settle historically divisive issues, and addressed them with a conciliatory spirit.4 This consensus has been a major source of legitimacy for the Constitution, and for the political system deriving from it throughout the years. Furthermore, it strengthened the Constitutional Court’s wide powers by providing it with extensive political support for decades. Besides, the rule of law had been traditionally disregarded by Spanish authorities making a new and powerful court an essential element. As a matter of fact, the Constitutional Court, relying on the normative understanding of the Constitution, allowed for a deep change of the legal system without replacing the members of the judiciary. In a way, the constitutional culture derived from the Constitution and not vice versa,5 the Constitutional Court being the main player. The Constitutional Court’s status rose during the first decades of democracy. However, when the new Catalan Statute of Autonomy was challenged before the Spanish Constitutional Court, political parties openly blocked the appointment procedure and amended the Constitutional Court Act, attempting to influence the judgment. Besides, the Catalan Statute of Autonomy was approved after a referendum which made the decision very controversial in Catalonia. This chain of events eventually undermined the Court’s popular support.6 Thus, the weak constitutional culture of the political elite eventually impinged on the Constitutional Court’s legitimacy. The lack of political arrangements, the prominence of a Constitutional Court, whose legitimacy is currently being challenged, and the myth of the clear meaning of the Constitution have blocked the constitutional system.7 Furthermore, the constitutional consensus made constitutional amendments a taboo for decades, which prevented necessary reforms from being approved. In this regard, Spanish regionalism has become particularly intricate. The dilemma seems only to be resolved either by breaching the rules or by amending the Constitution. In this regard, the 2011 amendment provoked further criticism to constitutional stagnation.8 Therefore, the Constitution has been crucial for the creation and development of a democratic political system. Nowadays, there is increasing popular pressure seeking constitutional amendment, which is indicative of the constitutional maturity of the political community.9

3 Francisco Tomás y Valiente, ‘La Constitución de 1978 y la historia del Constitucionalismo español’, (1980) Anuario de Historia del Derecho Español 721, 746. 4 See Ferreres Comella, Constitution of Spain (n 1) 19–23. 5 Pedro Cruz Villalón, ‘Constitución y cultura constitucional’, (1988) 211 Revista de Occidente 11, 19. 6 Maribel González Pascual and Daniel Toda Castán, ‘Katalonien und Spanien; Bruch oder Verfassungsreform?’, (2016) 69 Die Öffentliche Verwaltung 269, 270 f. 7 Maribel González Pascual, ‘Methods of Interpreting Competence Norms: Judicial Allocation of Powers in a Comparative Perspective’, (2013) 14 German Law Journal 1501, 1519. 8 It should be recalled that the amendment of Art 135 SC was debated and approved by both Chambers in 12 days thanks to the support of the two main political parties. 9 Víctor Ferreres Comella, ‘Amending the National Constitutions to save the Euro: is this the right strategy?’, (2013) 48 Texas International Law Journal 233.

Spain  287 Moreover, seeking constitutional reform as a tool to settle current controversies shows that the Constitution is still a key element of the collective imagination.

III.  Constitutional Foundations of EU Membership and Closely Related Instruments Despite the lack of any explicit reference to the EU, the constitutional foundation of EU membership is grounded on Article  93 SC. This Article was specifically drafted in light of the future accession of Spain to the EU. Furthermore, the Constitutional Court declared that Article 93 SC is the ‘ultimate foundation’ of Spain’s incorporation to the process of European integration and Spain’s bond with EU law.10 Article 93 SC sets forth that the ratification of international treaties that transfer constitutional powers to an international organisation or institution shall be authorised by the Organic Law. As such, the Constitution enables the transfer of sovereign powers to international organisations under parliamentary approval through the Organic Law. This type of act requires an absolute majority in Congress (half of the deputies plus one) and, thus, a broader consensus than for ordinary legislation, for which simple majority is the rule. Still, this majority is lower than the majority required for a constitutional amendment. Approval by Organic Law was the procedure followed to authorise the accession of Spain to the European Communities (EC)11 and the subsequent treaty amendments. A referendum is not compulsory for the ratification of international treaties. Article 92 SC enables the Prime Minister to call for a referendum on politically salient issues. Indeed, in 2005, a referendum was held on the Treaty Establishing a Constitution for Europe (TECE). The outcome in Spain was positive, but the process of ratification failed after the negative referenda in France and the Netherlands. Regarding judicial oversight, the Constitution provides a mechanism for the ex ante control of international treaties. Prior to ratification, international treaties may be reviewed by the Constitutional Court to check their compatibility with the Constitution. Judicial review can be triggered by the Government, the Congress or the Senate. The decision of the Constitutional Court is compulsory. If the Constitutional Court declares that the treaty breaches the Constitution, the treaty may only be ratified after the Constitution is amended accordingly. This mechanism has been set in motion twice regarding EU Treaties. In Declaration 1/1992,12 the Constitutional Court held that the Treaty of Maastricht clashed with the Constitution. The Maastricht Treaty granted EU citizens the right to vote and to stand as candidates in local elections in the country of residence, while according to the Constitution, aliens may only be given the right to vote in local elections, excluding the right to stand as candidates. The decision of the Constitutional Court prompted the first amendment of the Constitution. The amendment was limited to including the right of foreigners to stand as candidates in Article 13(2) SC. The ordinary constitutional amendment procedure was used (Article 167 SC), thus, without holding a referendum. Later on, the Government brought the TECE before the Constitutional Court. In particular, the Court considered the conformity with the Constitution of the clause stating the primacy of

10 Tribunal

Constitucional de España (hereinafter TCE) 13.12.2004 Declaración ECLI:ES:TC:2004:1D para 2. Law 10/1985, 2 August, authorising the accession of Spain to the European Communities. 01.07.1992 Declaración 1/1992 ECLI:ES:TC:1992:1D.

11 Organic 12 TCE

288  Aida Torres Pérez and Maribel González Pascual EU law (Article I-6 TECE); and the horizontal clauses on the scope of application of the Charter of Fundamental Rights (the Charter) and its interpretation (Articles II-111 and II-112 TECE). In Declaration 1/2004,13 the Constitutional Court confirmed TECE’s constitutionality. Following previous case law, the Constitutional Court held that the primacy of EU law was grounded on Article 93 SC, which enables the transfer of powers to an international organisation.14 Moreover, the Court acrobatically distinguished between the concepts of primacy and supremacy and argued that the primacy of EU law did not impinge upon the supremacy of the Constitution. While supremacy indicated a relationship of hierarchy and validity, primacy related to the applicability of norms. As the argument went, the primacy of EU law over domestic legislation had been accepted on the basis of Article  93 SC and thus the Constitution was still supreme.15 In the case of an insurmountable conflict between the Constitution and EU law, the Constitutional Court claimed the last word for the sake of popular sovereignty and constitutional supremacy.16 With regard to the Charter provisions, the question was broadly about the overlap between the constitutional and European declarations of rights. The Constitutional Court acknowledged that the Charter would apply as part of EU law through Article 93 SC, taking into account its limited scope of application. At the same time, following Article 10(2) SC, the Court held that the Charter’s interpretive force would extend beyond the scope of application of EU law.17 The Court understood that, on the basis of Article III-113 SC (now Article 53 of the Charter), the Charter would perform as minimum level of protection and concluded that there was no clash between the Constitution and the TECE.18 Generally, Article 96 SC governs the interaction between the domestic legal system and international law. International treaties are conceived to prevail over domestic law in case of conflict. As mentioned above, the Constitutional Court has acknowledged the specific characteristics of EU law, and in particular the primacy of EU law over domestic legislation on the grounds of Article 93 SC. In areas where powers have been transferred, EU law will prevail over domestic legislation, since the Constitution has authorised the transfer of powers to the EU. At the same time, however, according to the Constitutional Court, Article  93 SC has not granted EU law constitutional status. In sum, Article 93 SC is the foundation of both the transfer of powers to the EU and the acknowledgment of the primacy of EU law. The sole constitutional article that explicitly mentions the EU is Article  135 SC. On 27 September 2011, before the Fiscal Compact was signed, Article 135 SC was completely redrafted. The new provision was mainly a response to the high cost of Spanish bonds’ yields, the international markets being the constitutional amendment’s primary addressee. The Autonomous Communities were, however, the secondary addressee. Article 135 SC19 establishes the principle 13 TCE 1/2004 (n 10). 14 Ibid, para 4. 15 Antonio López Castillo, Alejandro Saiz Arnaiz and Víctor Ferreres Comella, Constitución Española y Constitución Europea análisis de la Declaración del Tribunal Constitucional (DTC 1/2004, de 13 de diciembre) (Madrid, Centro de Estudios Políticos y Constitucionales, 2005); Ricardo Alonso García, ‘Constitución española y Constitución europea: Guión para una colisión virtual y otros matices sobre el principio de primacía’ (2005) 73 Revista Española de Derecho Constitucional 339. 16 TCE 1/2004 (n 10) para 4, in fine. 17 Juan Ugartemendía Eceizabarrena, ‘La posición del Tribunal Constitucional ante la incorporación de los Derechos Fundamentales de la Unión Europea’ in S Ripol Carulla and J Ugartemendía Eceizabarrena (eds), España ante los tribunales internacionales europeos. Cuestiones de política judicial (Vitoria, Instituto Vasco de Administración Pública, 2008) 251, 269 f. 18 TCE 1/2004 (n 10) para 6. The interpretation of Art 53 of the Charter as a minimum floor of protection was not embraced by the CJEU in Melloni (CJEU 26.02.2013 C-399/11 (Melloni) ECLI:EU:C:2013:107). 19 Article 135 SC: ‘1. All public administration services shall adapt their actions to the principle of budgetary stability.

Spain  289 of budgetary stability, local authorities being required to have a balanced budget, whereas the central state and the regional administrations must not exceed the structural deficit margins established by the EU. Furthermore, public debt must observe the constraints determined by the Treaty on the Functioning of the EU (TFEU), and its payment must be given absolute priority. Nevertheless, the constraints of both public debt and structural deficit may be modified by the Congress in the event of a natural disaster, economic recession or extraordinary emergency which could cause a severe prejudice to economic or social sustainability.20 The deficit target and the debt limits were intended to enter into force in 2020. However, according to the first transitory provision of the Act, whenever an Excessive Deficit Procedure (henceforth EDP) is triggered regarding Spain, the deficit target will have to comply with the duties arising from the EDP.21 An individual complaint was launched before the Constitutional Court regarding the procedure followed to approve the constitutional amendment, but it was considered ill-founded.22 Later, the Budgetary and Financial Stability Act 2/2012, approved under Article 135 SC, was challenged by the Canarias Government also before the Constitutional Court. That constitutional reform was built on an already existing set of laws approved in 2001 aiming at coordinating the Autonomous Communities’ deficits. Nevertheless, the internal stability pact was toothless, given that the Autonomous Communities not fulfilling the deficit targets could not be sanctioned. The Budgetary and Financial Stability Act, however, confers extensive inspection and sanction powers to the central Government if the Autonomous Communities fail to achieve the structural and public debt targets. Article  11(6) of the Budgetary and Financial Stability Act was one of the provisions challenged. This provision establishes the structural deficit allowed to administrations by relying on the methodology followed by the EU Commission. According to the Canarias Government, however, this article empowers the central Government to decide unilaterally the amount of structural deficit that each administration may incur, given that there is no mandatory EU rule 2. Neither the Central State nor the Regional Governments may incur a structural deficit that exceeds the margins established, as the case may be, by the European Union for its Member States. A Constitutional Law shall set the maximum structural deficit permitted for the Central State and for the Regional Governments in relation to the gross domestic product thereof. Local authorities shall be required to present a balanced budget. 3. The Central State and the Regional Governments shall require legislative authorisation to issue public debt or secure credit. Loans to meet payment on the interest and capital of the public debt held by the public administration services shall always be deemed to be included in budget expenditure and their payment shall receive absolute priority. Such loans may not be subject to amendment or modification while they are in line with the conditions established by legislation on the issue thereof. The volume of public debt held by the public administration services as a whole in relation to the gross domestic product of the Central State may not exceed the reference value established in the Treaty on the Functioning of the European Union. 4. The limits on the structural deficit and the volume of public debt may only be exceeded in the event of natural disaster, economic recession or situations of extraordinary emergency beyond the control of the Central State and which seriously prejudice the financial situation or the economic or social sustainability of the Central State in the opinion of an absolute majority of the Members of the Lower House of Parliament. 5. A Constitutional Law shall develop the principles provided for in this article, as well as the involvement in the respective procedures by the institutional bodies for coordination between the public administration services in matters of fiscal and financial policy. At any event, it shall govern: a) the distribution of the deficit and debt limits between the various public administration services, the exceptional cases when the same may be exceeded and the method and term to correct deviations that may arise in relation to one or another; b) the methodology and procedure for calculating the structural deficit; c) the responsibility of each public administration service in the event of failure to comply with budgetary stability targets. 6. Pursuant to their respective governing statutes and within the limits referred to in this article, the Regional Governments shall adopt the appropriate provisions to effectively apply the principle of stability in their regulations and budgetary decisions.’

20 On

the shortcomings of Art 135 SC, see Ferreres Comella, ‘Amending’ (n 9). transitional provision Budgetary and Financial Stability Act 2/2012. 13.01.2012, Auto 9/2012 ECLI:ES:TC:2012:9A.

21 First

22 TCE

290  Aida Torres Pérez and Maribel González Pascual on the matter.23 The Constitutional Court declared this provision constitutional on the basis of EU membership. The Constitutional Court recalled the regions’ obligation to comply with EU law and also the need to coordinate economic policy within the EU on the basis of the principle of loyal cooperation between the EU and the Member States. According to the Constitutional Court, the EU is entitled to assess Member States’ deficits and also to decide which methodology to follow to determine it, the basis being EU law direct effect (Article 288 TFEU). Therefore, ‘it is not only constitutionally necessary to observe the maximum structural deficit determined by the EU (Article 126 TFEU and Article 135(2)SCCE) but the EU provisions related to the method to be followed to assess the deficit’.24 Therefore, paradoxically, the Constitutional Court clearly endorsed the strength of EU law and its impact on the Constitution in a case concerning a non-mandatory EU rule. Such reasoning can only be understood in the frame of the economic crisis, during which the Constitutional Court has been extremely deferential to any economic measure adopted by the central Government and to any recommendation stemming from the EU. In this regard, the Constitutional Court’s interpretation is extremely deferential to the EU legal system as there is a constitutional duty anchored in Articles 93 and 135 SC to comply with EU soft law.25 Such an interpretation is alien to the traditional approach of the Constitutional Court to both EU law and soft law, and bears witness to the ultimate strength of the Economic and Monetary Union (henceforth EMU) soft law. Therefore, the amendment to Article 135 SC had a particular bearing on the Spanish constitutional legal framework, considering the case law of the Spanish Constitutional Court regarding EU law following that amendment.

IV.  Constitutional Limits to EU Integration The Constitution does not set any explicit substantive limits to European integration. As explained above, Article 93 SC merely conditions the transfer of competences to parliamentary approval through Organic Law. Until Declaration 1/2004, the Constitutional Court held that Article 93 SC merely had a procedural nature.26 In Declaration 1/2004, in reviewing the compatibility of the TECE with the Constitution, the Constitutional Court acknowledged that Article 93 SC had a substantive dimension as well, as it opened up the constitutional order to other legal systems.27 Indeed, the Court argued that there are implicit substantive limits to the transfer of powers to an international organisation derived from the Constitution. These limits include state sovereignty, constitutional basic structures, and the system of fundamental values and principles enshrined in the Constitution, fundamental rights in particular.28 Eventually, the Court concluded that the TECE did not encroach upon those limits. 23 In fact the Government relies on the following document published by the European Commission, Directorate-General for Economic and Financial Affairs: Gilles Mourre, George-Marian Isbasoiu, Dario Paternoster and Matteo Salto, The cyclically-adjusted budget balance used in the EU fiscal framework: an update (Economic Papers 478, Brussels, March 2013), which explicitly states that ‘the views expressed are the author’s alone and do not necessarily correspond to those of the European Commission’. 24 TCE 18.12.2014 Sentencia 215/2014 ECLI:ES:TC:2014:215. 25 In a similar vein the Constitutional Court has also declared unconstitutional several regional laws regarding right to housing on the basis of both the fifth review of the Financial Assistance Programme for Spain and the letter sent by the EU Commission to the Spanish Government on June 2013. 26 TCE 14.02.1991 Sentencia 28/1991 ECLI:ES:TC:1991:28, para 4; TCE 1/2004 (n 10) para 4. 27 Xabier Arzoz Santisteban, La tutela de los derechos fundamentales de la Unión Europea por el Tribunal Constitucional (Madrid, Instituto Nacional de Administración Pública, 2015) 77. 28 TCE 1/2004 (n 10) para 2; José Martín y Pérez de Nanclares, ‘Hacia una nueva concepción del artículo 93 de la Constitución española: comentario a la Declaración 1/2004 del Tribunal Constitucional’, (2005) 52 Revista Española de Derecho Constitucional 799.

Spain  291 The Constitutional Court has acknowledged that the Constitution is not a parameter of validity of EU law, and has not claimed the power to review the constitutionality of secondary EU legislation. Neither has it claimed to have jurisdiction to review EU legislation against the yardstick of EU law. Only in cases of insurmountable conflict between EU law and the Constitution, when the excesses of EU legislation under the Treaties have not been resolved using the mechanisms provided by the EU legal order, has the Constitutional Court claimed the power to intervene to safeguard popular sovereignty and the supremacy of the Constitution. Thus, the Constitutional Court reserved for itself the last word in exceptional circumstances.29 So far, the Constitutional Court has not activated the counter-limits doctrine. In Melloni,30 the Constitutional Court faced a clash between the constitutional interpretation of the right to a fair trial and the Framework Decision on the European Arrest Warrant (EAW).31 The Constitutional Court lodged its first-ever request for preliminary ruling before the Court of Justice of the European Union (CJEU). The CJEU confirmed the validity of the Framework Decision under the Charter, and refused to enable the Constitutional Court to enforce a higher level of constitutional protection to review the decision to execute an arrest warrant against a person who had been condemned in absentia in Italy. In the end, while the Constitutional Court rhetorically referred to the counter-limits doctrine developed in Declaration 1/2004 and to its ultimate power to safeguard the Constitution, the Court reversed the settled interpretation of the right to a fair trial to fulfil its obligations under EU law.32

V.  Constitutional Rules and Practice on Implementing EU Law The Spanish Parliament plays a minor role in EU affairs, the Government being a particularly strong executive branch. In fact, the Parliament does not seem very interested in controlling government activities concerning the EU.33 Autonomous Communities, however, have asked for greater participation in EU decision-making since 1986. As a consequence, regional scrutiny over the Government with respect to EU affairs is much more effective than the parliamentary scrutiny. Spanish regionalism has developed from 1983 onwards, when the Castile-Leon Statute of Autonomy was finally approved. This longstanding process of devolution, along with the wellknown conflicts between the central state and the Autonomous Communities in Spain, might explain the flaws that still surround the Autonomous Communities’ involvement in EU law.34 These two axes, Parliament versus Government and central state versus Autonomous Communities, affect both the participation of the Government in EU institutions and the implementation of EU law. Therefore, in order to fully grasp the impact of the EU upon the constitutional system, both axes will be analysed regarding the scrutiny of the Government’s participation in the Council, the subsidiarity principle test, and the transposition and implementation of EU law. 29 TCE 1/2004 (n 10) para 4. 30 TCE 13.02.2014 Sentencia 26/2014 (Melloni) ECLI:ES:TC:2014:26. 31 2002/584/JHA, Council Framework Decision of 13 June 2002 on the European Arrest Warrant and the surrender procedures between Member States, [2002] OJ L 190/1. 32 For a critical approach to the use of the counter-limits doctrine in this decision, see Arzoz Santisteban, La tutela (n 28) 96 ff. 33 Empirical research has highlighted the low performance of the Spanish Parliament in EU affairs. Such a conclusion has been drawn from the analysis of parameters such as the commissions’ meetings, reports drafted or hearings within the Spanish Parliament regarding EU affairs. See also: José Bueno y Vicente, La contribución de las Cortes Generales a la integración europea. 1986-2004 (Madrid, Congreso de los Diputados, 2006); Ariadna Salazar Quiñonez, La Participación de los Parlamentos Nacionales en la UE (Madrid, Congreso de los Diputados, 2015). 34 See sections V.A and V.B.

292  Aida Torres Pérez and Maribel González Pascual

A.  Parliamentary and Regional Scrutiny of Spanish Ministers in the Council From the beginning of the Spanish Parliament’s engagement with the EU, it has been reactive, similar to other southern European parliaments.35 The main player in the Parliament is the Joint Committee for the EU, which is composed of members of the lower and upper houses and governed by Act 8/1994. Its main task has been to be informed and give opinions on EU legislative acts and to hold hearings with governmental officials regarding EU affairs. Act 8/1994was adapted to the Lisbon Treaty with Act 24/2009 of 22 December. The new Act conferred on the Joint Committee the competences to adopt reasoned opinions on compliance with the subsidiarity principle, to ask the Government to bring action for annulment before the CJEU on grounds of infringement of the subsidiarity principle, to make a proposal for the lower and upper chambers to oppose a ‘passerelle clause’, to be informed of new membership applications, and to monitor Eurojust’s and Europol’s activities (Article 3, Act 8/1994). This amendment strengthened also the control of the Joint Committee over the Government, but the new competences overburdened it.36 In fact, this committee has become one of the most active in the Parliament, but ‘the scrutiny might be turning more “legal” and mechanical in nature’.37 Furthermore, in our view, there are still flaws in the regulation, which diminish the Joint Committee’s competence to control the Government. First, Government appears before the Joint Committee only after Council meetings take place. Either the foreign affairs minister or the secretary for the EU must appear before the Joint Committee but only at the end of each Presidency. The Joint Committee can require the presence of any member of the Government. However, in practice it is usually low-rank officers and experts outside the Government, instead of ministers, that attend the Joint Committee meetings. Thus, the Joint Committee cannot actually control the Government participation in the Council. In line with this flawed legislation, the Joint Committee does not deliver any binding resolution on the Government. Secondly, even though the access of Parliament to information regarding the EU has improved thanks to Protocol 2 of the TFEU,38 there are still deficiencies which impair its capacity to process it. In this regard, according to Article 3 of Act 8/1994, the Government must send reports to the Joint Committee regarding both the EU legislative proposals of interest for Spain and the Spanish Government’s position on EU affairs. Nevertheless, the Act 8/1994 does not establish a deadline for these reports, which are normally too brief.39 Last but not least, it is the parliamentary majority – not the opposition – that decides which mechanisms to deploy to control the Government. In fact, the presence of members of the Government, the need for a parliamentary debate, or the request for further information is in the hands of the parliamentary majority. However, both the Congress and the Senate have normally backed the Government. 35 José Magone, ‘South European national Parliaments and the European Union’, in J O’Brennan and T Raunio (eds), National Parliaments within the Enlarged European Union. From ‘victims’ of integration to competitive actors? (Abingdon, Routledge, 2007) 116, 122. 36 Elviro Aranda, ‘La alerta temprana en el procedimiento legislativo de la UE’ (2013) 44 Revista de Derecho Comunitario Europeo 132. 37 Sonia Piedrafita, ‘The Spanish Parliament and EU Affairs in the Post Lisbon Treaty Era: All Change?’, (2014) 20 The Journal of Legislative Studies 451. 38 TFEU, Protocol (No 2) on the application of the principles of subsidiarity and proportionality, [2008] OJ C 115/206-9. 39 Control of European Council ex post meetings is particularly guaranteed, since the Government must appear in the plenary of the Congress to report on the decisions made (Art 4, Statute 8/1994 (consolidated version)).

Spain  293 It must be pointed out that the Spanish two-party system has imploded during the Eurocrisis, the Government being supported by 85 out of 350 MPs at the moment. This deep change of the political system has instigated a vivid Parliament also in EU matters. In this regard, the tools provided by the rulings of both Chambers are being particularly exploited by the MPs in an unprecedented way. Regional governments have been the ones who actually control, to a certain extent, the Spanish Government participation in the Council. In 1994 the central state and the Autonomous Communities signed an agreement according to which Autonomous Communities could determine the Spanish negotiation in EU affairs when exclusive regional competences are involved. Furthermore, when the competence is shared by the central state and the Autonomous Communities, the Spanish position should be agreed between them. Nonetheless, Autonomous Communities could only determine the initial Spanish position. The central Government is entitled to modify it if necessary during the negotiations in Brussels. In any case, the real hindrance for this regional participation was the lack of agreement among the regional governments. Any agreement required regional consensus, and such consensus was achieved only once in the period 1992–2004. Regional participation in EU affairs was deepened after an agreement between the central Government and the regional governments in 2004. Currently, regional governments may be part of the Spanish delegations in four out of 10 Council configurations.40 Even though the Autonomous Communities can only intervene in the Council if the content of the regional intervention has been previously agreed with the central state, the regional presence in the negotiations allows them to watch over the central Government. In fact, consensus has been frequently reached among Autonomous Communities in the framework of the 2004 agreement, regardless of the deep internal political disputes.41 Regional cooperation and regional interaction with the central state is indeed much more functional in EU affairs than in any other matter. Thus, Autonomous Communities seem to follow a different cooperative strategy in EU affairs than in other matters. It could be argued that Autonomous Communities have learnt that the EU implies a different set of players and rules.42 A horizontal structured cooperation among all the Autonomous Communities is becoming the norm in EU matters, instead of the competitive bilateral negotiations between Autonomous Communities and the central state which characterise Spanish regionalism. Nonetheless, a deeper and wider regional participation in EU affairs is still required. Autonomous Communities’ participation in EU matters is lower than other European regions with similar powers.43 Moreover, given the deficient involvement of the Spanish Parliament, the only national democratic players watching over the Spanish Government in the Council are the regional governments. Thus, an improvement of regional participation would clearly increase the control of the Government’s activities in the EU.

B.  Subsidiarity Test The Joint Committee is responsible for the adoption of the reasoned opinions on the compliance of EU draft legislative acts with the principle of subsidiarity. The parliamentary legal service 40 Autonomous Communities participate in the Configurations dealing with agriculture and fisheries; education, youth, culture and sport; employment, social policy, health and consumer affairs and environment. 41 Autonomous Communities reached consensus on 108 regional positions on EU matters during the period 2005–12. 42 Maribel González Pascual, Las Comunidades Autónomas en la UE. Condicionantes, evolución y perspectivas de futuro (Barcelona, Institut d’Estudis Autonòmics, 2013) 96 f. 43 Michael Thatam, ‘Devolution and EU policy-shaping: bridging the gap between multi-level governance and liberal intergovernmentalism’, (2011) 3 European Political Science Review 53.

294  Aida Torres Pérez and Maribel González Pascual sends to the Committee the EU legislative proposals that may be subject to the subsidiarity test, with a short memorandum about the Government’s position and a note specifying whether the Parliament of any other Member State is considering a negative opinion. Furthermore, the Joint Committee can ask the Government for a report regarding the compliance of the EU legislative proposal with the subsidiarity principle. The members of the Bureau and the Board of Spokespersons of the Joint Committee decide which proposals to scrutinise assigning a rapporteur to each of them. According to the Resolution of the Bureaus of both Chambers, approved on 27 May 2010, the rapporteur presents the report and, afterwards, a debate among the members of the Joint Committee should be triggered, the final decision being taken by vote. However, in practice Joint Committee resolutions are adopted with consent of the attendants with no debate or voting.44 Moreover, the decisions are made by the majority which might explain that the opinions are in line with the Government position on the matter.45 All in all, it must be pointed out that the Joint Committee is increasingly relying on the Early Warning System to engage in a deeper dialogue with the Commission. In this regard, the Joint Committee highlighted on several occasions that not only the subsidiarity principle but also the proportionality principle had been violated. Moreover, since 2012 the Joint Committee delivers opinions with observations, in which the Committee considers that the subsidiarity has been observed but stresses other kinds of flaws regarding the EU proposal.46 Hence, the practice is evolving and could eventually put in place a system that deepens the participation of the Parliament in EU affairs, especially in the frame of the implosion of the two political-parties system. It is also worth mentioning that regional parliaments are in principle also involved in the Early Warning System. The legal service of the Spanish Parliament forwards EU legislative proposals to all of them. From this moment on, the regional parliaments have four weeks to draft an opinion. However, the deadline is quite short and the regional opinion falls short of causing any effect. The Joint Committee simply attaches it to its own reasoned opinion, if there is one. Unsurprisingly, there is a decreasing participation of regional parliaments in the subsidiarity test.47 Hence, the regional governments’ involvement in EU affairs has been widened, but that of the regional parliaments has not.

C.  EU Law Transposition and Implementation: Parliament and Autonomous Communities Spanish performance in the application and enforcement of EU law has generally been quite poor, the number of judgments of the CJEU against Spain under Articles 258 and 260 TFEU being a 44 Piedrafita, ‘Spanish Parliament’ (n 37). 45 In the cases in which the rapporteurs have differed from the Government position, the Joint Committee has either appointed another rapporteur (Report 6/2012, on the compliance with the principle of subsidiarity) or supported the Government position instead of the rapporteur’s (debate on the proposal a directive of the European Parliament and of the Council on improving the gender balance among non-executive directors of companies listed on stock exchanges and related measures, Diario de Sesiones de las Cortes Generales, Comisiones mixtas 46, 18 December 2012). The Joint Committee’s documents regarding the subsidiarity test can be found under www.congreso.es/portal/page/portal/ Congreso/Congreso/CongresoUE. 46 The Joint Committee has delivered nine opinions with observations in which it has either considered that the EU rule is not necessary or made specific recommendations to improve the EU rule quality. 47 Maribel González Pascual, ‘Los parlamentos regionales y el mecanismo de alerta temprana’, (2012) 93 Revista Vasca de Administración Pública/Herri-Arduralaritzako Euskal Aldizkaria 60.

Spain  295 telling example. Nevertheless, in 2011 the EU pilot files and the infringement procedures against Spain started a clear downward trend.48 It could be wondered whether the new Article 135 SC could have played a role in this shift, given that up until then there had been neither an effective legislation nor actual practice regarding either the transposition or the application of EU law. The main concern was to keep the balance between the Government and the Parliament on the one hand, and between the central state and the Autonomous Communities on the other hand. However, there has been ‘a fierce swipe off the equilibrium of powers within the parliamentary system’.49 Generally, the Spanish Government plays a core role within the legislative procedure, but its position is particularly strengthened when it is necessary to transpose or implement EU law.50 In this regard, EU law is mostly transposed or implemented via Regulations approved by the Government. Furthermore, even when EU law is enforced by the Parliament, the legislative practice is not satisfactory. In this regard, usually several directives or framework decisions are transposed by a single statute, which is approved in a short period of time. Hence, a particularly long, not always coherent, legislative proposal is drafted and tabled by the Government before the Parliament under severe time constraints. This is done even though such statutes require a profound debate for members of the Parliament to grasp its actual scope. Moreover, it is also common practice to include in a statute, which transposes a directive, controversial provisions unrelated to the EU law content.51 These questionable practices take advantage of the minimal resistance and scarce debate in the Parliament when a statute is related to the EU. In fact, the Government aims at circumventing an open debate on controversial measures by being presented as EU law matters.52 The constitutionality of these measures has not been challenged, but the transposition of directives via a Royal Decree Law (Real Decreto-ley, RDL) has.53 In this regard, the Constitutional Court has considered that the Government may approve an RDL to transpose a directive only if Spain is about to be sanctioned.54 The Constitutional Court, however, has delivered numerous decisions concerning the transposition and execution of EU law by the Autonomous Communities. The two key points have been the actual impact of EU law upon the constitutional allocation of powers and the Autonomous Communities’ accountability for non-compliance with EU law. The Constitutional Court has repeatedly pointed out that the constitutional allocation of powers has not been modified by EU membership. Hence, the internal allocation of powers determines which institution is 48 Report from the Commission. Monitoring the application of Union law 2014 Annual Report, COM(2015) 329 final, Brussels, 9.7.2015 49 Rafael Bustos Gisbert, ‘National Constitutional Identity in European constitutionalism: revisiting the tale of the Emperor’s new clothes in Spain’, in A Saiz Arnaiz and C Alcoberro Llivina (eds), National Constitutional Identity and European Integration (Mortsel, Intersentia, 2013) 75, 82. 50 Anna Palau, ‘La europeización de la agenda legislativa estatal y autonómica (1998-2007)’, (2013) 160 Revista de Estudios Políticos 96. 51 Fabio Pascua, ‘La ejecución normativa del Derecho Comunitario en España: problemas y perspectivas’ in F Rubio Llorente (ed), Informe del Consejo de Estado sobre la inserción del Derecho Europeo en el ordenamiento español. Texto del informe, estudios y ponencias (Madrid, Consejo de Estado, 2008) 542. 52 In this regard, for instance, the transposition of the Directive (EU) 2016/1919 of the European Parliament and of the Council of 26 October 2016 on legal aid for suspects and accused persons in criminal proceedings and for requested persons in European arrest warrant proceedings, [2016] OJ L 297/1, was used to increase the police’s powers to interfere on the privacy of the suspects in criminal proceedings. 53 An RDL is adopted by the Government in exceptional circumstances, has the same force as statutes approved by Parliament, and enters into force once published. The lower chamber ratifies the text as a whole, after a short parliamentary debate, by a simple majority. 54 The constitutional court has only accepted the transposition via RDL in a case in which the Commission had already launched an infringement procedure against Spain, TCE 13.01.2012 Sentencia 1/2012, ECLI:ES:TC:2012:1.

296  Aida Torres Pérez and Maribel González Pascual entrusted with the EU law transposition and/or application.55 However, the Constitutional Court has also considered that EU law can be a useful interpretative tool to determine the allocation of powers, whenever the controversy concerns EU law implementation.56 This interpretation is not explicitly spelled out. Still, it should not be underestimated. In this regard, for instance, the Constitutional Court shaped the Autonomous Communities’ competence over river basins, relying on the so-called Water Directive. The Court considered that the European definitions57 were the ones in conformity with the constitutional provision entrusting the central state with the competence over river basins.58 Such an interpretation involved the unconstitutionality of provisions contained in the Statutes of Autonomy of both Castile-Leon and Andalusia. All in all, the main controversy regarding EU law transposition concerns the role of the central state. In this regard, directives often concern areas in which the central state and the Autonomous Communities share competence, the central state being entrusted with the competence over the general framework. However, in many cases the directive establishes just general framework. Thus, in principle, the Autonomous Communities might transpose the directive without the central state’s intervention. In other words, it could be argued that the central state competence is undermined by the EU in many cases. Nevertheless, this reasoning has not been endorsed by the Constitutional Court. The central state also has competence to establish the basis of economic policy as well as the main conditions for an equal enjoyment of constitutional rights in a wide sense throughout the territory. Competence over economic policy can be easily related to numerous EU rules, whereas competence on constitutional rights is linked to equal access to all kind of entitlements. As a consequence, the central state tends to transpose most directives on the basis of the equal access to constitutional rights, economic policy or general frameworks for several matters.59 The predominant role of the central state was justified to a large extent by the Autonomous Communities’ lack of accountability in cases of non-compliance with EU law. According to Article 93 SC, both the central Government and the Parliament must guarantee compliance with Spain’s international commitments. However, the Constitutional Court has held that Article 93 SC does not envisage any competence for the central powers to control the regional ones. Nevertheless, it has followed a systemic interpretation of the Constitution when confronted with the impact of EU law upon the allocation of powers.60 Therefore, on the one hand the Central State was not entitled to request compensation, based on domestic law, from the Autonomous Communities and/or other institutions that were responsible for the non-fulfilment of EU obligations. On the other hand, the Constitutional Court took a liberal reading of the central state competences in cases related to EU law. Still, several rules were approved over the years to make Autonomous Communities accountable at the domestic level for non-fulfilment of EU obligations on a case-by-case basis. In fact, these rules responded to actual infringement procedures in which Spain had been sanctioned.61 55 TCE 28.05.1992 Sentencia 79/1992 ECLI:ES:TC:1992:79. 56 TCE 17.02.2005, Sentencia 33/2005 ECLI:ES:TC:2005:33. 57 The provision followed was Art 2(13) of the Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy [2000] OJ L 327/1. 58 TCE 16.03.2011, Sentencia 30/2011 ECLI:ES:TC:2011:30; and TCE 17.03.2011, Sentencia 32/2011 ECLI:ES:TC:2011:32. 59 Rafael Bustos Gisbert, ‘La ejecución del derecho comunitario por el Gobierno central’, (2003) 67 Revista Vasca de Administración Pública/Herri-Arduralaritzako Euskal Aldizkaria 163, 179 f. 60 TCE 17.03.1992 Auto 80/1992 ECLI:ES:TC:1992:80A. 61 In fact, Art  121 bis of the Water Act was the first provision establishing that if Spain was sanctioned by the EU institutions, the administration responsible for this should compensate the central state. This provision was approved immediately after Spain was sanctioned by the ECJ judgment in C-278/01 (Commission v Spain) ECLI:EU:C:2003:635 (25.11.2003).

Spain  297 The first general rule on domestic liability for EU infringements was the final provision of the Sustainable Economy Act 18/2001, which was replaced by Article  8 of the already mentioned Budgetary and Financial Stability Act. Therefore, a scheme has been put in place which holds accountable each and every administration for non-compliance with EU law at the domestic level. This new scheme is grounded in the main role of the central Government disregarding a cooperative scheme,62 even though the cooperation between the central state and the Autonomous Communities in EU affairs has proved to be quite successful. However, if the Constitutional Court is to be consistent, it should change its own case law regarding the transposition and implementation of EU law. The central state’s creeping competence on EU affairs could be understood to a certain extent when Autonomous Communities were not held accountable if they did not comply with EU law. Under the new provisions, Autonomous Communities must be entrusted with EU law transposition and implementation in accordance with the constitutional allocation of powers. Such a development could alleviate the centralisation trend in Spain that is being fostered by the measures adopted in the frame of the European economic governance.

D.  Enforcement through the Courts, Including Protection of the EU Charter of Fundamental Rights Spanish courts have gradually internalised their role as ‘European courts’.63 Shortly after the accession of Spain to the EC, the Constitutional Court acknowledged the primacy of EU law over domestic legislation.64 At the same time, the Court held that EU law did not have constitutional status and that the enforcement of EU law was the function of ordinary courts. Consequently, it was for ordinary courts to make preliminary reference in case of doubt about the interpretation or validity of EU law, as well as to set aside national legislation clashing with EU law. In this vein, the Constitutional Court did not grant any constitutional relevance to the judicial enforcement of EU law.65 Over time, however, the Constitutional Court acknowledged the potential breach of the Constitution as a consequence of the failure to make a preliminary reference or to set aside national legislation conflicting with EU law. And yet, the evolution of the constitutional case law in this matter has been rather convoluted.66 In 2004,67 the Constitutional Court declared that the failure to make a preliminary reference before disapplying EU law violated the right to a fair trial (Article 24 SC). On the one hand, this decision was celebrated, since for the first time the Constitutional Court gave constitutional relevance to the obligation to make a preliminary reference. On the other hand, the reasoning was quite disturbing, since apparently the preliminary reference was conceived as a condition to be able to set aside national legislation clashing 62 Manuel Medina, ‘El Estado Autonómico en tiempos de disciplina fiscal’, (2013) 98 Revista Española de Derecho Constitucional 109, 145 f. 63 Daniel Sarmiento, La aplicación del Derecho de la Unión por el Tribunal Supremo en tiempos de crisis, Instituto de Derecho Europeo e Integración Regional, Working Papers 13 (2012). 64 TCE 28/1991 (n 26) para 6. 65 For a critical approach, see Miguel Azpitarte Sánchez, El Tribunal Constitucional ante el control del derecho comunitario (Madrid, Civitas, 2002) 122–26. 66 Aida Torres Pérez, ‘The Judicial Impact of European Law in Spain: ECHR and EU Law Compared’, (2011) 30 Yearbook of European Law 159, 169–72. 67 TCE 19.042004 Sentencia 58/2004 ECLI:ES:TC:2004:58.

298  Aida Torres Pérez and Maribel González Pascual with EU law.68 This interpretation was confirmed in 2006.69 Nonetheless, according to EU law, the preliminary reference is only compulsory for courts of last instance in case of doubt about the interpretation or validity of EU law, while there is no obligation to make the reference as a prior condition to set aside national legislation.70 Finally, in Metropole,71 the Constitutional Court expressly revised its previous doctrine. It held that there is no duty to make a reference to the CJEU before disapplying a legislative provision that violates EU law when the question at hand would be materially identical to a previous one resolved by the CJEU (Da Costa),72 or when the acte clair doctrine applies (CILFIT).73 The number of preliminary rulings lodged by Spanish courts has steadily increased over time.74 Since accession, the total amount has been 460 ­references,75 of which 78 came from the Supreme Court and one from the Constitutional Court. Furthermore, the Constitutional Court has strengthened the obligation to set aside domestic legislation clashing with EU law as a matter of constitutional law. In Iberdrola,76 the Court held that the failure to disapply a domestic provision, which had already been held in breach of EU law in an infringement action against Spain, violated the constitutional right to a fair trial (Article  24 SC).77 This judgment was significant since, in the past, the Constitutional Court had claimed that the compatibility between EU law and domestic law did not concern the Constitutional Court. Also, the Court emphasised the ex tunc effects of CJEU judgments. In November 2015, the Constitutional Court decided a case reiterating that while EU law does not enjoy constitutional nature, the failure to correctly apply EU law might have constitutional relevance from the perspective of the right to a fair trial (Article 24 SC).78 In previous judgments, the CJEU had already interpreted the principle of equality enshrined in Article 4(1) of Directive 1999/70/CE, and had even declared that the domestic provision at stake was at odds with the Directive. The Constitutional Court argued that when the CJEU has already interpreted EU law, the failure to enforce EU law according to the CJEU interpretation, might amount to an ‘unreasonable and arbitrary selection of the applicable law’ and result in a violation of the right to a fair trial.79 In the case at hand, the Constitutional Court pointed out that the High Court of Madrid had not in any way assessed previous CJEU case law, which had been brought to its attention by the interested party. The ordinary court had decided to ignore EU law, in violation of the primacy principle and the constitutional right to a fair trial.80 68 Ricardo Alonso García, ‘Spanish Constitutional Court. Judgment 58/2004, of 19 April 2004’ (2005) 42 CML Rev 535; Juan Ugartemendía Eceizabarrena, ‘El recurso a la prejudicial como cuestión de amparo (A propósito de la STC 58/2004, de 19 de abril)’, (2004) 11 Revista Española de Derecho Europeo 441. 69 TCE 19.06.2006 Sentencia 194/2006 ECLI:ES:TC:2006:194. 70 As was made clear in CJEU 19.01.2010 C-555/07 (Kücükdeveci) ECLI:EU:C:2010:21. 71 TCE 20.10.2010 Sentencia 78/2010 ECLI:ES:TC:2010:78. 72 ECJ 27.03.1963 28/62 (Da Costa en Schaake NV and Others v Administratie der Belastingen) ECLI:EU:C:1963:6. 73 ECJ 06.10.1982 283/81 (CILFIT v Ministero della Sanità) ECLI:EU:C:1982:335; Cristina Izquierdo Sans, ‘Cuestión prejudicial y artículo 24 de la Constitución española’, (2011) 23 Revista General de Derecho Europeo 1; Ricardo Alonso García, Cuestión prejudicial y tutela judicial efectiva (a propósito de las SSTC 58/2004, 194/2006 y 78/2010), Instituto de Derecho Europeo e Integración Regional, Working Papers 4 (2011) 8; Luis Arroyo Jiménez, ‘La aplicación judicial del derecho de la Unión Europea y el derecho a la tutela judicial efectiva. Una propuesta de sistematización’, (2014) 102 Revista Española de Derecho Constitucional 293. 74 CJEU, Annual Report 2017, 121–22. 75 Spain is number eight in terms of the total amount of references per Member State before the CJEU, ibid, at 122. 76 TCE 02.07.2012 Sentencia 145/2012 (Iberdrola) ECLI:ES:TC:2012:145. 77 Daniel Sarmiento, ‘Reinforcing the (domestic) constitutional protection of primacy of EU law Tribunal Constitucional (Spanish constitutional court), Judgment 145/2012of 2 July 2012, Iberdrola v Comisión Nacional de la Energía’, (2013) 50 CML Rev 875. 78 TCE 05.11.2015 Sentencia 232/2015 ECLI:ES:TC:2015:232. 79 Ibid, para 5. 80 Ibid, para 6.

Spain  299 The judicial enforcement of EU fundamental rights deserves particular attention from a constitutional perspective. Already in the 1990s, the Constitutional Court held that EU fundamental rights ought to be taken into account for the interpretation of constitutional rights in light of Article  10(2) SC.81 In practice, for example, references to several sources of EU law, including EU Treaties, Directives, and case law, have contributed to the interpretation of the right to non-discrimination on grounds of sex,82 and in particular the concept of indirect discrimination.83 Even before the Charter became legally binding, the Constitutional Court referred to it as a source for interpreting constitutional rights. Usually, the Charter was quoted to reinforce a conclusion already reached by the Constitutional Court regarding the interpretation of a fundamental right. For instance, the Constitutional Court held that although Article 14 SC does not explicitly mention sexual orientation among the grounds of prohibited discrimination, sexual orientation had to be regarded as one of them. To support this conclusion, the Constitutional Court quoted Article 21(1) Charter, which explicitly bans discrimination on grounds of sexual orientation.84 After the Charter entered into force (December 2009), the Constitutional Court kept making use of the Charter as a hermeneutic tool on the basis of Article 10(2) SC. At the same time, the Constitutional Court has emphasised that EU law, the Charter, and CJEU case law are not constitutional parameters for reviewing the validity of domestic legislation. The Court made this clear in a case in which the constitutionality of the Law on Social Security was challenged in terms of non-discrimination of part-time workers on grounds of sex.85 The Charter tends to be quoted as an argument ad abundantiam, to reinforce a reached conclusion of the Court. For instance, regarding the right to one’s own name, the Constitutional Court quoted Article 7 of the Charter and the CJEU judgment in Sayn-Wittgenstein.86 Also, in a case about discrimination on grounds of age, the Constitutional Court argued that even though Article 14 SC does not explicitly mention age as a ground of forbidden discrimination, according to previous case law, age shall be regarded as one of the circumstances that triggers heightened scrutiny. After reaching this conclusion, the Constitutional Court held that this interpretation was strengthened by Article 21(1) of the Charter, which explicitly bans discrimination on grounds of age, and the CJEU case law on the matter.87 On occasions, the content of constitutional and Charter rights might diverge. In a case about the expulsion of the mother of a Spanish minor, the Constitutional Court distinguished the content of Article  18(1) SC from Article  8(1) ECHR and Article  7 of the Charter. The Constitutional Court held that while the constitutional right protects the privacy of the family, the ECHR and the Charter protect family life as such. At the same time, the Court argued that the protection of family life is provided by a different constitutional clause (Article 39 SC), which does not receive protection through the individual complaint procedure before the Constitutional Court.88 81 TCE 28/1991 (n 26) para 5; TCE 22.03.1991 Sentencia 64/1991 ECLI:ES:TC:1991:64, para 4; Saiz Arnaiz, Apertura Constitucional (n 2) 185–203; Xabier Arzoz Santisteban, ‘La relevancia del derecho de la Unión Europea para la interpretación de los derechos fundamentales constitucionales’, (2005) 74 Revista Española de Derecho Constitucional 63. 82 TCE 01.07.1991 Sentencia 145/1991 ECLI:ES:TC:1991:145; TCE 28.02.1994 Sentencia 58/1994 ECLI:ES:TC:1994:58; TCE 23.07.1996 Sentencia 136/1996 ECLI:ES:TC:1996:136; TCE 20.11.2006 Sentencia 324/2006 ECLI:ES:TC:2006:324 para 6; TCE 12.02.2007 Sentencia 17/2007 ECLI:ES:TC:2007:17 para. 6. 83 TCE 15.01.2007 Sentencia 3/2007 ECLI:ES:TC:2007:3 para 3. 84 TCE 13.02.2006 Sentencia 41/2006 ECLI:ES:TC:2006:41. 85 TCE 14.03.2013 Sentencia 61/2013 ECLI:ES:TC:2013:61 para 5; TCE 14.02.2013 Sentencia 41/2013 ECLI:ES:TS:2013:41 86 TCE 07.10.2013 Sentencia 167/2013 ECLI:ES:TC:2013:167 referring to CJEU 22.12.2010 C-208/09 (Sayn-Wittgenstein) ECLI:EU:C:2010:806. 87 TCE 16.03.2015 Sentencia 22/2015 ECLI:ES:TC:2015:22 para 3. 88 TCE 04.11.2013 Sentencia 186/2013 ECLI:ES:TC:2013:186 para 6.

300  Aida Torres Pérez and Maribel González Pascual In Melloni, the Constitutional Court decided to submit its first preliminary reference before the CJEU. The case dealt with the potential conflict between the constitutional understanding of the right to a fair trial and the obligations under the Framework Decision on the EAW when the accused has been convicted in absentia. The submission of a preliminary reference was already a path-breaking decision, since up until then the Constitutional Court had claimed that the preliminary reference was only for ordinary courts to make.89 The Constitutional Court asked the CJEU for an interpretation and validity of the Framework Decision in light of Articles 47 and 48 of the Charter. In addition, the Constitutional Court also posed a question about the elusive meaning of Article  53 of the Charter. Unsurprisingly, the CJEU upheld the validity of the Framework Decision and followed the ECtHR case law in the interpretation of the right to a fair trial. The meaning to be given to Article 53 of the Charter was much trickier and the CJEU ruled that domestic courts were only allowed to enforce more protective constitutional rights in reviewing state acts implementing EU law as long as the primacy, unity, and effectiveness of EU law were not undermined.90 In the end, the Constitutional Court was compelled to reverse the settled interpretation of the right to a fair trial. Nonetheless, the Constitutional Court followed the rationale of Article 10(2) SC and the CJEU judgment was regarded as a hermeneutic tool, in the same way as the case law of the ECtHR. In this vein, even though the final outcome was in conformity with EU law, the Constitutional Court downplayed the binding nature of the preliminary reference.91 Finally, although ordinary courts tend to approach fundamental rights controversies under the Constitution, they are also increasingly making reference to the Charter. Usually, the Charter is quoted ad abundantiam.92 For instance, the Charter was quoted in cases about the protection of personal data (Article 8 of the Charter),93 the right to good administration (Article 41 of the Charter),94 or the freedom of information and the press (Article 11(2) of the Charter).95 On occasions, the Charter might become the main ground to declare a violation of a fundamental right. This happened in a case where the Supreme Court declared the violation of the right to good administration, and in particular ‘the right of every person to be heard, before any individual measure which would affect him or her adversely is taken’ (Article 41(1) of the Charter).96 In addition, Spanish courts have referred to the CJEU questions relating to the interpretation of the Charter. This has happened in cases such as Google Spain,97 regarding the ‘right to be forgotten’, in light of Articles 7 and 8 of the Charter or cases with a deep impact upon the Spanish legal system, such as Aziz98 and Sánchez Morcillo,99 regarding the right to consumer protection and to a fair trial (Article 47 of the Charter) in the context of the housing crisis. 89 Miguel Revenga Sánchez, ‘Rectificar preguntando. El Tribunal Constitucional acude al Tribunal de Justicia (ATC 86/2011, de 9 de junio)’, (2012) 41 Revista Española de Derecho Constitucional 139; Maribel González Pascual, ‘Mutual Recognition and Fundamental Constitutional Rights: The First Preliminary Reference of the Spanish Constitutional Court’, in M Claes et al (eds), Constitutional Conversations in Europe: Actors, Topics and Procedures (Mortsel, Intersentia, 2012) 161. 90 CJEU C-399/11 (Melloni) (n 18). 91 Aida Torres Pérez, ‘Melloni in Three Acts: From Dialogue to Monologue’, (2014) 10 European Constitutional Law Review 308, 321 ff. 92 See the cases included in the Case-Law Database of the Fundamental Rights Agency http://fra.europa.eu/en/ case-law-database/country/spain-431. 93 TCE 04.11.2014 Auto 267/2014 ECLI:ES:TS:2014:267A para 7; Tribunal Supremo (hereinafter TSE) 08.10.2010 Sentencia STS 5439/2010 ECLI:ES:TS:2010:5439 para 3. 94 TSE 14.10.2013 Sentencia STS 4968/2013 ECLI:ES:TS:2013:4968. 95 TSE 08.07.2010 Sentencia STS 5075/2010 ECLI:ES:TS:2010:5075. 96 TSE STS 4968/2013 (n 95) para 3. 97 CJEU 13.05.2014 C-131/12 (Google Spain and Google) ECLI:EU:C:2014:317. 98 CJEU 14.03.2013 C-415/11 (Aziz) ECLI:EU:C:2013:164. 99 CJEU 17.07.2014 C-169/14 (Sánchez Morcillo and Abril García) ECLI:EU:C:2014:2099.

Spain  301

VI.  Resulting Relationship between EU Law and National Law The Constitution was the outcome of a broad political consensus and provided the grounds for building a democratic system in Spain. Even though the EU is mentioned only once in the Constitution, European integration has had a significant impact upon the constitutional order. The two constitutional amendments were triggered by the EU. While the amendment to recognise foreigners’ right to stand as candidates in local elections was rather smooth, the amendment of Article 135 SC to incorporate the balanced budget rule was highly contentious. The constitutional foundations of EU membership are still found in Article  93 SC, which merely sets forth the procedure to transfer sovereign powers to international organisations. The Constitutional Court transformed this Article into a ‘European clause’. Article 93 has become the basis for acknowledging the primacy of EU law. Furthermore, the Constitutional Court interpreted that, even though this clause did not specify any substantive limits, there were implicit limits to integration in terms of sovereignty, basic constitutional structures, and fundamental rights. Nonetheless, so far, the counter-limits doctrine has not been activated. EU membership has impacted upon Spain’s constitutional design, weakening the Parliament even more. In fact, the key role played by the Government in EU matters has not been counterbalanced by an adequate parliamentary scrutiny, the Joint Committee on EU affairs’ powers being poorly shaped. The early warning system has fostered the Joint Committee activities but it might have worsened its general performance by overburdening it. Furthermore, the development and implementation of EU law lies mostly in the hands of the Government. In this regard, the Autonomous Communities have taken a more active stance towards the EU. Since 2004 regional governments participate in the Spanish delegations in four Council configurations. Interestingly enough, this participation is based on a horizontal cooperation which is at odds with the lack of regional cooperation in domestic affairs. Still, their participation is far from the one achieved by other regions in Europe with similar powers. Furthermore, the central state is entitled to transpose most of the EU Directives and Framework Decisions according to the Constitutional Court. Last but not least, the Autonomous Communities’ financial powers have been undermined by Article 135 SC, which remains the only constitutional provision that mentions Spain’s EU membership. The judicial enforcement of EU law has been bolstered by the constitutional case law over time. While at first the Constitutional Court did not grant any constitutional relevance to the enforcement of EU law by ordinary courts, subsequently the Court held that the failure to make a preliminary reference (when compulsory), or to set aside national legislation incompatible with EU law might encroach upon the right to a fair trial (Article 24 SC). Finally, in the field of fundamental rights, the Constitutional Court and ordinary courts tend to quote the Charter as a hermeneutic tool on the basis of Article  10(2) SC. The references to the Charter usually contribute to reinforcing the meaning given to constitutional rights. Exceptionally, in Melloni, the Constitutional Court was compelled to reverse the settled interpretation of the right to fair trial in order to ensure the effectiveness of EU law.

References R Alonso García, ‘Constitución española y Constitución europea: Guión para una colisión virtual y otros matices sobre el principio de primacía’, (2005) 73 Revista Española de Derecho Constitucional 339. R Alonso García, Cuestión prejudicial y tutela judicial efectiva (a propósito de las SSTC 58/2004, 194/2006 y 78/2010), Instituto de Derecho Europeo e Integración Regional, Working Papers 4 (2011).

302  Aida Torres Pérez and Maribel González Pascual R Alonso García, ‘Spanish Constitutional Court. Judgment 58/2004, of 19 April 2004’ (2005) 42 CML Rev 535. E Aranda, ‘La alerta temprana en el procedimiento legislativo de la UE’, (2013) 44 Revista de Derecho Comunitario Europeo 132. L Arroyo Jiménez, ‘La aplicación judicial del derecho de la Unión Europea y el derecho a la tutela judicial efectiva. Una propuesta de sistematización’, (2014) 102 Revista Española de Derecho Constitucional 293. X Arzoz Santisteban, La tutela de los derechos fundamentales de la Unión Europea por el Tribunal Constitucional (Madrid, Instituto Nacional de Administración Pública, 2015). X Arzoz Santisteban, ‘La relevancia del derecho de la Unión Europea para la interpretación de los derechos fundamentales constitucionales’, (2005) 74 Revista Española de Derecho Constitucional 63. M Azpitarte Sánchez, El Tribunal Constitucional ante el control del derecho comunitario (Madrid, Civitas, 2002). José Bueno y Vicente, La contribución de las Cortes Generales a la integración europea. 1986-2004 (Madrid, Congreso de los Diputados, 2006). R Bustos Gisbert, ‘National Constitutional Identity in European constitutionalism: revisiting the tale of the Emperor’s new clothes in Spain’, in A Saiz Arnaiz and C Alcoberro Llivina (eds), National Constitutional Identity and European Integration (Mortsel, Intersentia, 2013) 75. R Bustos Gisbert, ‘La ejecución del derecho comunitario por el Gobierno central’, (2003) 67 Revista Vasca de Administración Pública/Herri-Arduralaritzako Euskal Aldizkaria 163. P Cruz Villalón, ‘Constitución y cultura constitucional’, (1988) 211 Revista de Occidente 11. V Ferreres Comella, The Constitution of Spain. A Contextual Analysis (Oxford, Hart Publishing, 2013). V Ferreres Comella, ‘Amending the National Constitutions to save the Euro: is this the right strategy?’, (2013) 48 Texas International Law Journal 233. M González Pascual, ‘Mutual Recognition and Fundamental Constitutional Rights: The First Preliminary Reference of the Spanish Constitutional Court’, in M Claes et al (eds), Constitutional Conversations in Europe: Actors, Topics and Procedures (Mortsel, Intersentia, 2012) 161. M González Pascual, ‘Los parlamentos regionales y el mecanismo de alerta temprana’, (2012) 93 Revista Vasca de Administración Pública/Herri-Arduralaritzako Euskal Aldizkaria 60. M González Pascual, Las Comunidades Autónomas en la UE. Condicionantes, evolución y perspectivas de futuro (Barcelona, Institut d’Estudis Autonòmics, 2013). M González Pascual, ‘Methods of Interpreting Competence Norms: Judicial Allocation of Powers in a Comparative Perspective’, (2013) 14 German Law Journal 1501. M González Pascual and D Toda Castán, ‘Katalonien und Spanien; Bruch oder Verfassungsreform?’, (2016) 69 Die Öffentliche Verwaltung 269. C Izquierdo Sans, ‘Cuestión prejudicial y artículo 24 de la Constitución española’, (2011) 23 Revista General de Derecho Europeo 1. A López Castillo, A Saiz Arnaiz and V Ferreres Comella, Constitución Española y Constitución Europea análisis de la Declaración del Tribunal Constitucional (DTC 1/2004, de 13 de diciembre) (Madrid, Centro de Estudios Políticos y Constitucionales, 2005). J Magone, ‘South European national Parliaments and the European Union’, in J O’Brennan and T Raunio (eds), National Parliaments within the Enlarged European Union. From ‘victims’ of integration to competitive actors? (Abingdon, Routledge, 2007) 116. J Martín y Pérez de Nanclares, ‘Hacia una nueva concepción del artículo 93 de la Constitución española: comentario a la Declaración 1/2004 del Tribunal Constitucional’, (2005) 52 Revista Española de Derecho Constitucional 799. M Medina, ‘El Estado Autonómico en tiempos de disciplina fiscal’, (2013) 98 Revista Española de Derecho Constitucional 109. G Mourre, G-M Isbasoiu, D Paternoster and M Salto, The cyclically-adjusted budget balance used in the EU fiscal framework: an update (Economic Papers 478, Brussels, March 2013) A Palau, ‘La europeización de la agenda legislativa estatal y autonómica (1998-2007)’, (2013) 160 Revista de Estudios Políticos 96.

Spain  303 F Pascua, ‘La ejecución normativa del Derecho Comunitario en España: problemas y perspectivas’ in F Rubio Llorente (ed), Informe del Consejo de Estado sobre la inserción del Derecho Europeo en el ordenamiento español. Texto del informe, estudios y ponencias (Madrid, Consejo de Estado, 2008) 542. S Piedrafita, ‘The Spanish Parliament and EU Affairs in the Post Lisbon Treaty Era: All Change?’, (2014) 20 The Journal of Legislative Studies 451. M Revenga Sánchez, ‘Rectificar preguntando. El Tribunal Constitucional acude al Tribunal de Justicia (ATC 86/2011, de 9 de junio)’, (2012) 41 Revista Española de Derecho Constitucional 139. A Saiz Arnaiz, La Apertura Constitucional al Derecho Internacional y Europeo de los Derechos Humanos. El Artículo 10.2 de la Constitución Española (Madrid, Consejo General del Poder Judicial, 1999). A Salazar Quiñonez, La Participación de los Parlamentos Nacionales en la UE (Madrid, Congreso de los Diputados, 2015). D Sarmiento, ‘Reinforcing the (domestic) constitutional protection of primacy of EU law Tribunal Constitucional (Spanish constitutional court), Judgment 145/2012of 2 July 2012, Iberdrola v Comisión Nacional de la Energía’, (2013) 50 CML Rev 875. D Sarmiento, La aplicación del Derecho de la Unión por el Tribunal Supremo en tiempos de crisis, Instituto de Derecho Europeo e Integración Regional, Working Papers 13 (2012). M Thatam, ‘Devolution and EU policy-shaping: bridging the gap between multi-level governance and liberal intergovernmentalism’, (2011) 3 European Political Science Review 53. F Tomás y Valiente, ‘La Constitución de 1978 y la historia del Constitucionalismo español’, (1980) Anuario de Historia del Derecho Español 721. A Torres Pérez, ‘Melloni in Three Acts: From Dialogue to Monologue’, (2014) 10 European Constitutional Law Review 308. A Torres Pérez, ‘The Judicial Impact of European Law in Spain: ECHR and EU Law Compared’, (2011) 30 Yearbook of European Law 159. J Ugartemendía Eceizabarrena, ‘La posición del Tribunal Constitucional ante la incorporación de los Derechos Fundamentales de la Unión Europea’ in S Ripol Carulla and J Ugartemendía Eceizabarrena (eds), España ante los tribunales internacionales europeos. Cuestiones de política judicial (Vitoria, Instituto Vasco de Administración Pública, 2008) 251. J Ugartemendía Eceizabarrena, ‘El recurso a la prejudicial como cuestión de amparo (A propósito de la STC 58/2004, de 19 de abril)’, (2004) 11 Revista Española de Derecho Europeo 441.

304

10 France RAINER PALMSTORFER, LAURENT SERMET AND ERICH VRANES*

I.  Main Characteristics of the National Constitutional System and Constitutional Culture A. Introduction The Constitution of the Fifth Republic (FC) was adopted on 4 October 1958. Unlike earlier periods of French history, which were marked by ‘relative constitutional instability’,1 the ­ post-1958 French constitutional culture has produced a peculiar form of living constitutionalism. This living constitutionalism is based on three pillars. The ‘main’ one is of a political nature and is linked to the reform of the text of the constitution (ie, constitutional amendments) decided by the political institutions (ie, the French Parliament). The second one is of a judicial nature and is based on the case law of the French Constitutional Council (Conseil Constitutionnel).2 The last one refers to customary practice. According to Rogoff, ‘[t]his mix of mechanisms (political, judicial, and customary practice) for legitimating changes has resulted in a Constitution which now plays an increasingly important role in advancing the rule of law and symbolizing and promoting social and political integration’.3

B.  A Political Constitution: An Adaptive Text President de Gaulle’s initial constitutional footprint was about the distribution of powers. In this system, there was a mixture of parliamentary and presidential elements. There are two uncontested powers that are never politically responsible: the President and the Senate. By contrast, the National Assembly may be dissolved by the President and the Government may be withdrawn by the President or by the National Assembly. The President may be dismissed for criminal reasons only in relation to his/her office (Article 68 FC). The Constitution has been amended to allow for * This contribution is based on and contains material that has already been published in Erich Vranes, ‘Constitutional Foundations of, and Limitations to, EU Integration in France’, (2013) 19 European Public Law 535. 1 Martin Rogoff, ‘Fifty Years of Constitutional Evolution in France: The 2008 Amendments and Beyond’, in M Rogoff et al (eds), The Financial Crisis of 2008. French and American Responses. Proceedings of the 2010 Franco-American Legal Seminar (Portland ME, University of Maine School of Law, 2011) 1. 2 Decisions of the Constitutional Council are available at www.conseil-constitutionnel.fr. 3 Rogoff, ‘Fifty Years’ (n 1) 1.

306  Rainer Palmstorfer, Laurent Sermet and Erich Vranes the President to be judged for international crimes, if any, as France has ratified the Rome Treaty (International Criminal Court) (Article 53-2 FC). Since its adoption, the FC has been amended 24 times. The original version of the FC was clearly drafted by and for de Gaulle. It was also de Gaulle who, for the sake of acquiring increased legitimacy, initiated the second amendment of the Constitution in 1962. This amendment introduced direct presidential election. The frequent amendments have also changed the ‘original spirit’ of the Constitution.4 This being said, the number and the recent acceleration in the frequency of constitutional revisions has been an issue of concern, for one could possibly argue that these amendments are detrimental to the original text. Conversely, this adaptability seems necessary to ensure flexibility and the evolution of the regime. Previously, the Republican Regime was changed, and a new Constitution was adopted. Under the French Constitution of 1958, it appears to be possible to amend the Constitution without changing the overall regime. Since 1958, the Republic has been stable in this respect. The last amendments took place in July 2008 under Sarkozy’s presidency.5 As a rule, amendments of the Constitution follow the procedure governed by Article 89 FC. This procedure requires a three-fifths majority of the votes cast in the Congress of the French Parliament (Congrès du Parlement français), the latter being composed of the National Assembly (Assemblée nationale) and the Senate (Sénat). Alternatively, this high parliamentary threshold can be replaced by an approval in a referendum. Against the background of constitutional amendments, it has to be mentioned that there is no obvious eternity clause in the Constitution. There is also no explicit hierarchy between the different clauses of the Constitution.

C.  A Living Constitution: Customary Practices Constitutional practice is primarily a matter of the executive and legislative powers. An interesting case of customary practices came with the ‘cohabitation time’, when the President had to install a Prime Minister and a Government not belonging to the President’s party. This was because the French people had elected a President yet given its support to a different political wing in the parliamentary elections. This was due to the fact that the electoral mandate was seven years for the President, but restricted to five years for the members of the lower Chamber. To avoid the cohabitation issue, in 2000, the presidential mandate was reduced to five years (named as quinquennat) in order to allow the French people to vote both for the President and their representatives in a two-month time delay. During cohabitation, a customary practice developed due to political necessity. Contrary to the letter of Article 8 FC, it was understood that the President lost the power to choose and to revoke the Prime Minister and his/her Government, as they had a different form of political legitimacy.

D.  A Judicial Constitution: The ‘Block of Constitutionality’ As will be shown below, the introduction of the body of provisions in the FC dealing with the French Republic’s membership in the European Union (EU) (ie Title XV FC) was predominantly 4 However, several constitutional revision projects were unsuccessful. These projects included the establishment of a five-year term for presidency (1973), the incompatibility between the mandate and membership of parliamentary government (1974), the enlargement of the referendum for civil liberties (1984), the creation of a referral to the Constitutional Council by way of exception (1990 and 1993), the evolution of New Caledonia and French Polynesia (2000). 5 Cf Constitutional Act no 2008-724 of 23 July 2008.

France  307 caused by rulings of the Constitutional Council. Under the French constitutional system (cf Title VII: Articles 56–63 FC), this Council acts as the guardian of the FC by assessing the constitutionality of legislation by carrying out ex ante and ex post reviews. Using its competences, the Constitutional Council considerably developed the FC. As a result, the present constitutional order is composed not only of the 1958 Constitution. In 1971,6 the Constitutional Council started considering the legal texts to which the Preamble7 of the FC refers as belonging to the bloc de constitutionnalité. This is considered to comprise ‘fundamental principles recognised under the laws of the Republic’ and includes the 1789 Déclaration des droits de l’homme et du citoyen8 and the Preamble of the Constitution of 1946.9 That way the Constitutional Council made a creative use of the preamble references to reinvent and effectively extend the FC. Professor Favoreu, of Aix-en-Provence Law Faculty, qualified this phenomenon as a ‘constitutionality block’, which means that the notion of a normative Constitution is wider than the written Constitution itself (whether in its initial or amended form). In particular, it is the lack of provisions on human rights in the written Constitution that explains the emergence of this concept. A fundamental principle of the type the Council mentioned with respect to the block of constitutionality finds its origin in the republican tradition. Under the case law of the Constitutional Council, a fundamental principle recognised under the laws of the Republic has to meet three characteristics to be established. First, the principle has to have been formed before the Fourth Republic – during the First, Second or Third Republic, with the Third Republic being the most inspirational period for this constitutional category. Secondly, the principle must be based on a statute adopted during the republican time. It has to be emphasised that, through this concept, norms of statutory nature obtain constitutional status. Strictly speaking, it is not the statutes as such which belong to the fundamental principles. Rather, it is the republican principle underlying the statute, with the latter being seen simply as an emanation of the said principle. In other words, the contents expressed in a statute need to be constantly reaffirmed in order to qualify as contents belonging to a fundamental principle. Thirdly, the norm at stake has to be of a general nature in order to qualify as an expression of a principle. Unless those three requirements are met, the Constitutional Council will not consider a statute as a fundamental principle recognised under the laws of the Republic. So far, it has been predominantly fundamental rights (eg freedom of conscience, right of defence) that were categorised as a fundamental principle recognised under the laws of the Republic. To conclude, the Council’s power to extend the FC may be interpreted as an amendment of the FC – but only to a certain extent, for this power is marked by respect for the written Constitution.

6 Constitutional Council (hereinafter CC), 71-44 DC of 16 July 1971 (Contrat d’association) ECLI:FR:CC:1971:71.44. DC. This landmark ruling can be seen as the starting point in the development of the notion of the block of constitutionality. The Constitutional Council established that the freedom of association belongs to the ‘fundamental principles recognized under the laws of the Republic’. Interestingly, the Constitutional Council held that the contents of the Statute Law of 1 July 1901, which is dedicated to the recognition of the principle of freedom of association, belong to the block of constitutionality. 7 The preamble of the 1958 Constitution reads as follows: ‘The French people solemnly proclaim their attachment to the Rights of Man and the principles of national sovereignty as defined by the Declaration of 1789, confirmed and complemented by the Preamble to the Constitution of 1946’. 8 CC 73-51 DC of 27 December 1973 (Loi de finances pour 1974) ECLI:FR:CC:1973:73.51.DC. This ruling dealt with the issue of equality before the law. Strictly speaking, the Declaration of 1789 was not made during a republican period, as the First Republic started in September 1792. 9 CC 74-54 DC of 15 January 1975 (Loi relative à l’interruption volontaire de la grossesse, the so-called IVG ruling) ECLI:FR:CC:1975:74.54.DC.

308  Rainer Palmstorfer, Laurent Sermet and Erich Vranes

II.  Constitutional Foundations of EU Membership and Closely Related Instruments A.  Constitutional Foundation of European Economic Community Membership in 1957 France was one of the founding members of the European Economic Community (EEC). At that time, the Constitution of 1946 was still in force. Pursuant to the Constitution of 1946, the constitutional procedure for adopting treaties was shared between the Parliament and the President. In the case of the treaties establishing the EEC and the European Atomic Energy Community (EAEC or Euratom), parliamentary approval was necessary (cf Article 27 of the Constitution of 1946). The President had the power to ratify the treaties, after the approval of the Parliament (Article 31 of the Constitution of 1946). With regard to these treaties, the parliamentary vote was favourable.10 Against this background, one has to come back to the Preamble of the Constitution of 1946, considered to be explicitly a part of the block of constitutionality by the Constitutional Council in 1975.11 The Preamble contains a clause dedicated to the Charter of the United Nations. This clause, which is also known as the ‘peace clause’,12 has two functions. On the one hand, it is used to validate French participation in the European integration project. On the other hand, the clause is also seen as a threshold against the giving up of French sovereignty without any limitations.

B.  The Introduction of Title XV Until the Maastricht process (1992), the FC did not contain any special provision for European integration. Instead, general norms were used (Title VI – On treaties and agreements: Articles 52 to 55 FC). As from 1992, a special Title (Title XV – On the EU) was adopted to give a constitutional basis to the French Republic’s membership in the EU. These amendments were prompted by rulings of the Constitutional Council on the compatibility of the Maastricht treaty with the French Constitution. In the context of the Maastricht Treaty, the Constitutional Council handed down three rulings. In its first ruling, it held that the Maastricht treaty contradicted the FC.13 In this ruling, the Constitutional Council found violations of essential conditions for the exercise of sovereignty regarding voting rights and eligibility to the municipal elections of European citizens settled in the territory of a Member State of which they are not nationals, the provisions on the Economic and Monetary Union (EMU), the common visa policy and the use of qualified majority after 1 January 1996 for the votes in the Council. As a response, Constitutional Act 92-554 of 25 June 1992 inserted a new Title in the FC, ‘The European Communities and the European Union’. 10 On 10 July1957, the French National Assembly adopted, by 342 votes against 239, a bill that authorised President René Coty to ratify the treaties establishing the EEC and Euratom. 11 CC 74-54 (n 9). 12 The peace clause reads as follows: ‘Subject to reciprocity, France shall consent to the limitations upon its sovereignty necessary to the organisation and preservation of peace’. The commitment of the Preamble is also stated with another clause: ‘The French Republic, faithful to its traditions, shall respect the rules of public international law. It shall undertake no war aimed at conquest, nor shall it ever employ force against the freedom of any people’ (official English translation, available at www.elysee.fr). 13 CC 92-308 DC of 9 April 1992 (Maastricht I) ECLI:FR:CC:1992:92.308.DC.

France  309 The second ruling (Maastricht II) clarified that the Maastricht treaty was not contrary to the amended Constitution.14 In Maastricht III, the Constitutional Council declared itself incompetent to rule on the constitutionality of the ratification authorisation act passed by a referendum (Law 92-1017), as such laws are considered to be the direct expression of national sovereignty.15 The Maastricht Treaty seems to have been a turning point. While the steps made in EC/EU integration were not reflected in amendments of the FC until 1992, there have since then been several successive modifications of the constitutional provisions. The Maastricht Treaty also marked the beginning of a series of rulings of the Constitutional Council dealing with EU integration issues. These rulings concerned Schengen,16 the Amsterdam Treaty,17 the Constitutional Treaty18 and the Lisbon Treaty.19 It has to be underlined that it was not until recently that French judicial organs have started to more explicitly take into account, in their legal reasoning, those features that are particularly characteristic for EU integration. Only after a considerable period of time, in which international law and EC/EU law were not clearly differentiated as to their effects within the French legal order, have the French supreme judicial organs, in a remarkable turnaround, started to more distinctly stress the spécificité of EU law. In its Constitutional Treaty ruling, the Constitutional Council regarded EU law as a legal order that is ‘integrated into the domestic legal order and distinct from the international legal order’,20 and for example by more explicitly recognising that this body of law may require a differentiated judicial approach in several respects.21 Limitations to French sovereignty which are necessary for international organisation and peace were permissible on the basis of paragraph 15 of the Preamble of the Constitution of 1946. Articles 88-1(ff) FC, inserted in 1992, constitute constitutional provisions that are central to EU integration. Several other provisions are relevant in the context of EU integration. The most important of these will be briefly described in the next section.

C.  The Individual Legal Provisions 1.  Preamble of the Constitution of 1946 and Déclaration of 1789 Together with Article 3 FC, the Preamble of the Constitution of 1946 and the Déclaration of 1789 can be seen as the constitutional foundations of French sovereignty. Although human rights have not figured as barriers to EU integration as prominently as in Germany, for example,22 the 14 CC 92-312 DC of 2 September 1992 (Maastricht II) ECLI:FR:CC:1992:92.312.DC para 46. 15 CC 92-313 DC of 23 September 1992 (Maastricht III) ECLI:FR:CC:1992:92.313.DC. 16 CC 93-325 DC of 13 August 1993 (Loi relative à la maîtrise de l’immigration et aux conditions d’entrée, d’accueil et de séjour des étrangers en France) ECLI:FR:CC:1993:93.325.DC, leading to the introduction of Art 53-1 (Constitutional Law No 93-1256). 17 CC 97-394 DC of 31 December 1997 (Amsterdam) ECLI:FR:CC:1997:97.394.DC. 18 CC 2004-505 DC of 19 November 2004 (Constitutional Treaty) ECLI:FR:CC:2004:2004.505.DC. 19 CC 2007-560 DC of 20 December 2007 (Lisbon) ECLI:FR:CC:2007:2007.560.DC. 20 CC 2004-505 (n 19) para 11 (‘un ordre juridique communautaire intégré à l’ordre juridique interne et distinct de l’ordre juridique international’). 21 Cf eg Council of State (hereinafter CoS), decision 287110 of 8 February 2007 (Arcelor) ECLI:F R:CEASS:2007:287110.20070208, which stresses that the scrutiny of national legal acts implementing EU directives may require adaptations (‘le contrôle de constitutionnalité des actes réglementaires assurant directement cette transposition est appelé à s’exercer selon des modalités particulières dans le cas où sont transposées des dispositions précises et inconditionnelles’). On this decision see below, section III.B.2. 22 Cf eg Franz C Mayer, Edgar Lenski and Mattias Wendel, ‘Der Vorrang des Europarechts in Frankreich’, (2008) 43 Europarecht 63.

310  Rainer Palmstorfer, Laurent Sermet and Erich Vranes Constitutional Council has recognised them as potential limits to integration. For the purposes of EU integration, Paragraph 15 of the preamble of the Constitution of 1946, the so-called ‘peace clause’, again is of particular interest.23 This clause has been regarded as a basis for limiting French sovereignty and transferring competences to the EU in its first Maastricht ruling.24 Since the aforementioned judicial reorientation, which was triggered by the Constitutional Council in 2004 and will be discussed below, Article 88-1, inserted at the occasion of the 1992 FC amendment, is also regarded as a – possibly general – basis for EU integration. Nonetheless, paragraph 15 is still consistently cited as a norm of reference, along with Article 88-1, in the assessment of whether amendments of primary EU law are compatible with the FC.25 The limits to amendments of primary law are, however, derived from the interplay of this clause with other constitutional provisions.

2.  Article 3 FC Article 3 FC stipulates that national sovereignty shall be vested in the people. Similarly, Article 3 of the Déclaration of 1789 proclaims that sovereignty resides with the nation. However, due to the interrelationship with other provisions (in this regard, the Constitutional Council cites in particular paragraphs 1426 and 15 of the Constitution of 1946, Article 53 and Title XV [governing EU integration] of the Constitution of 1958), French sovereignty is not per se seen as a barrier to EU integration. In other words, as there is a special title and a special provision within the Constitution that constitutionalise France’s EU membership, there should, in principle, be no legal argument against the EU legal order.27 Nonetheless, the ‘fundamental conditions of the exercise of national sovereignty’, a notion derived from these provisions by the Constitutional Council, have come to serve as a French constitutional barrier for amendments of primary EU law.

3.  Article 11 FC Apart from substantive requirements,28 the FC also provides for a number of procedural requirements regarding international treaties that need to be taken into account also against the background of EU integration.29 Pursuant to Article 52 FC, the President is competent to ratify treaties. Article 11 FC allows the President to ask the French people for direct approval before ratifying a treaty.30 This provision can have profound political and legal consequences. The President resorted to this procedure with regard of the ratification of the Maastricht Treaty and the Treaty Establishing a Constitution for Europe (TECE). The latter was rejected in a referendum

23 See n 12. 24 CC 92-308 (n 13) paras 9 ff. 25 Cf eg CC 2007-560 (n 19) para 5. 26 According to para 14, France is to comply with public international law. 27 See CC 2004-505 (n 18) para 11. 28 As will be shown, these are, in particular, issues of sovereignty and constitutional identity. 29 As regards the adoption of international treaties these are: Art 10 FC (Promulgation of Acts of Parliament by the President), Art 11 FC (referendum), Art 53 FC (Parliamentary approval of international treaties), Art 52 FC (Ratification of international treaties by the President) and, not least, Art 89 FC (Constitutional Amendment). 30 Art 11 FC reads as follows: ‘The President of the Republic may … submit to a referendum any Government Bill which … provides for authorisation to ratify a treaty which, although not contrary to the Constitution, would affect the functioning of the institutions’.

France  311 held on 29 May 2005, even though the FC had already been amended31 in order to respond to the Constitutional Council’s Constitutional Treaty ruling.32

4.  Article 53 FC The FC requires an Act of Parliament on the approval/ratification of a series of international ­treaties (cf Article 53 FC).33 This provision was/is applicable to the EU Treaties. It was also applied with regard to the ratification of the Treaty establishing the European Stability Mechanism (TESM) and of the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union (TSGC).

5.  Article 54, Article 61(2), Article 61-1 FC It is important, in legal and political terms alike, as there are many ways to seize the Constitutional Council, to note also that the Constitutional Council declines to re-examine provisions of primary EU law.34 This approach appears to provide a shield for the acquis at the level of primary law. Article 54 FC empowers the Constitutional Council to review the constitutionality of ­international undertakings. A referral can be made by the President of the Republic, the Prime Minister, the President of one of the other Houses, or by 60 Members of the National Assembly or 60 Senators. It must take place before ratification or approval of the international undertaking in question. If the Council declares unconstitutionality, the FC has to be amended prior to the ratification. The reasoning behind Article 54 FC is that a duly ratified treaty has to be fully applied.35 Since 1992, Article 54 FC has routinely been used to a priori review amendments of the EU treaties,36 which has led to several amendments of the FC. Thus, after the Constitutional Council’s Maastricht ruling,37 Articles 88-1, 88-2 and 88-3 were inserted into the FC. These provisions were amended due to the Council’s Amsterdam judgment.38 Likewise, the Constitutional Council’s Constitutional Treaty ruling39 and Lisbon ruling40 led to constitutional amendments41 and the insertion of Articles 88-6 and 88-7.42 With regard to the Nice Treaty, the Constitutional Council was not seized as the issue of conformity with the FC was

31 Constitutional Act No 2005-204 of 1 March 2005. 32 CC 2004-505 (n 18). 33 Pursuant to Art 53 FC, ‘Peace Treaties, Trade agreements, treaties or agreements relating to international organization, those committing the finances of the State, those modifying provisions which are the preserve of statute law, those relating to the status of persons, and those involving the ceding, exchanging or acquiring of territory, may be ratified or approved only by an Act of Parliament. They shall not take effect until such ratification or approval has been secured …’. 34 Cf eg CC 2004-505 (n 18) para 8. 35 See § 14 of the Preamble to the Constitution of 1946; also see Art 27 of the Vienna Convention on the Law of Treaties. 36 Cf also Loïc Azoulay and Ronkes Agerbeek, ‘Comment on Conseil Constitutionnel (French Constitutional Court), Decision No. 2004-505 DC of 19 November 2004, on the Treaty establishing a Constitution for Europe’, (2005) 42 CML Rev 871. 37 CC 92-308 (n 13). 38 CC 97-394 (n 17). 39 CC 2004-505 (n 18). 40 CC 2007-560 (n 19) para 5. 41 Cf Constitutional Act No 2005-204 of 1 March 2005 and Constitutional Act No 2008-103 of 4 February 2008. 42 See also Simon-Louis Formery, La constitution commentée, 13th edn (Paris, Hachette Livre, 2010/11) 116 f; Guy Carcassonne, La Constitution introduite et commentée, 9th edn (Montrouge, Editions du Seuil, 2009) 381 ff.

312  Rainer Palmstorfer, Laurent Sermet and Erich Vranes considered uncontroversial. In contrast to the TESM, the TSCG was subject to a procedure under Article 54 FC. The Constitutional Council held that the TSCG does not contain any unconstitutional provisions.43 A second avenue to review international agreements is available through Article 61(2) FC. Under this provision, acts of Parliament may be referred for constitutionality review to the Constitutional Council before promulgation. This provision has been interpreted by the Constitutional Council as empowering it to indirectly review international undertakings via the acts of Parliament ratifying them.44 In other words, if Article 54 FC provides for direct control of international commitments, Article 61(2) FC provides for their indirect control. This control may occur in two cases: with respect to a statute authorising a treaty (Article 53 FC) and in the case of a statute transposing a European directive (Article 88-1 FC). In both cases the Constitutional Council has to review the constitutionality of the respective statute. The case of transposing a directive involves two elements of control. The first element is the Constitutional Council’s constitutionality review of the French statutes transposing a ­directive. The second element is the European Court of Justice’s (ECJ or CJEU) control of the directive itself in light of the Treaties. As will be shown, the Constitutional Council has greatly limited its control of statutes transposing directives.45 The French system of constitutional review has been traditionally characterised by its limitation to a priori scrutiny of legal acts and international undertakings. However, it should be mentioned that a new type of a posteriori review entered into force on 1 March 2010. According to Article 61-1 FC, if, during ongoing judicial proceedings, it is claimed that a statutory provision infringes constitutional rights, the matter may be referred to the Constitutional Council. This new leg of constitutional review, which has become known as the question prioritaire de constitutionnalité, or QPC, has attracted a considerable amount of scholarly attention.46 In particular, it is questionable whether or not the QPC review allows for the control of treaties already in force in the French legal order and what could be the subject of a prior constitutional review? The Council of State (Conseil d’Etat) excluded this possibility on the grounds that the law in question (a law of approbation of a treaty, in the framework of Article 53 FC) escapes the procedure of the QPC.47 The QPC has initiated a judicial dialogue between the French supreme courts, the Constitutional Council and the CJEU. The Court of Cassation (Cour de Cassation) in a 2010 referral to the CJEU took the view that the QPC mechanism may be contrary to EU law.48 According to the Court of Cassation, French law requires French courts seized with a question prioritaire de constitutionnalité to rule as a matter of priority on whether to refer to the Constitutional Council a question on whether a provision of national law is consistent with the FC, when at the same time the conflict of that provision with EU law is at issue.49 Reacting immediately, the Constitutional Council and the Council of State held in obiter dicta in two other 2010 proceedings that the

43 CC 2012-653 DC of 9 August 2012 (TSCG) ECLI:FR:CC:2012:2012.653.DC. 44 On this see eg Pierre Pactet and Ferdinand Mélin-Soucremanien, Droit constitutionnel, 28th edn (Paris, Sirey, 2009) 567 ff with further references; Chloé Charpy, ‘Droit constitutionnel et droit communautaire. Le statut constitutionnel du droit communautaire dans la jurisprudence (récente) du Conseil’, (2009) 80 Revue Française de Droit Constitutionnel 621, 630 ff. 45 See at section III.B.1. 46 See the references in Vranes, ‘Constitutional Foundations’ (n *) 530 n 26. 47 Council of State, decision 312305 of 14 May 2010, ECLI:FR:CESSR:2010:312305.20100514. 48 Court of Cassation, decision 10-40001 of 16 April 2010 (Melki). 49 Cf CJEU 22.06.2010 C-188/10 (Aziz Melki) and C-189/10 (Sélim Abdeli) ECLI:EU:C:2010:363, paras 22 and 31.

France  313 Court of Cassation’s point of view was erroneous, taking a viewpoint close to that of the French Government in the CJEU proceedings triggered by the Court of Cassation.50 In its Solomonian ruling, the CJEU referred the matter back to the Court of Cassation, indicating that if the Court of Cassation’s interpretation of the QPC mechanism was correct (ie priority given to a constitutional review), then there would indeed be an infringement of EU law. However, if by contrast the view maintained by the French Government was appropriate, then the QPC system is compliant with EU law (ie the CJEU following its Simmenthal doctrine considers that the priority of an ‘incidental procedure of constitutional review’ is contrary to EU law only if it prevents national judges from exercising their powers or fulfilling their obligation to refer questions before it for a preliminary ruling).51 The CJEU’s ruling is noteworthy also in another respect. Namely, it appears to indicate that the Constitutional Council’s constant jurisprudence, according to which it regards itself as being prevented from referring matters to the ECJ under Article 267 TFEU, due to the one-month timelimit set for its rulings by Article 61 FC, may be inconsistent with EU law.52 However, the CJEU’s judgment can perhaps also be understood as implying that it belongs to national courts to interpret the relevant national procedural provisions in a manner which guarantees the supremacy of EU law. Nevertheless, the Constitutional Council accepted judicial dialogue in the context of the ex post review,53 by making its first preliminary reference to the CJEU.54 This first case concerned the Framework Decision of 13 June 2002 establishing the European Arrest Warrant.55 The Council considered that there were serious questions as to the scope of Articles 27 and 28 of that Framework Decision and admitted that the constitutional review of French law depended on the interpretation of those decisions. The Council, therefore, considered it necessary for the CJEU to give a preliminary ruling which was then taken into consideration by the Council.

6.  Article 55 FC Article 55 FC has consistently played a crucial role in the relationship between French constitutional law, EU law, and international law more generally.56 In the words of French scholars, this provision has given rise to a ‘saga of three decades’.57 This appears to be due to, inter alia, the fact that Article 55 FC does not explicitly state which judicial organ is called upon to ensure

50 CC 2010-605 DC of 12 May 2010 (Loi relative à l’ouverture à la concurrence et à la régulation du secteur des jeux d’argent et de hasard en ligne) ECLI:FR:CC:2010:2010.605.DC; Council of State, decision 312305 (n 47). 51 CJEU, joined Cases C-188/10 and C-189/10 (Melki) (n 49) paras 40 ff. 52 Ibid, para 56 (‘imposing a strict time-limit on the examination by the national courts cannot prevent the reference for a preliminary ruling on the validity of the directive in question’). 53 CC 2013-314P QPC of 4 April 2013 (Jeremy F) ECLI:FR:CC:2013:2013.314P.QPC. 54 This case concerned the Framework Decision establishing the European Arrest Warrant. The Council considered that there were serious questions as to the scope of Arts 27 and 28 of that Framework Decision and admitted that the constitutional review of French law depended on the interpretation of those decisions. The Council, therefore, considered it necessary for the CJEU to give a preliminary ruling which was then taken into consideration by the Council. 55 2002/584/JHA, Council Framework Decision of 13 June 2002 on the European arrest warrant and the surrender procedures between Member States, [2002] OJ L 190/1. 56 Art 55 is worded as follows: ‘Treaties or agreements duly ratified or approved shall, upon publication, prevail over Acts of Parliament, subject, with respect to each agreement or treaty, to its application by the other party’. 57 Terry Olson and Paul Cassia, Le droit international, le droit européen et la hiérarchie des normes (Paris, Presses Universitaires de France, 2006) 47 (‘saga jurisprudentielle d’une trentaine d’années’); see also Catherine Haguenau-Moizard, ‘Offene Staatlichkeit: Frankreich’, in von Bogdandy et al (eds), Handbuch Ius Publicum Europaeum vol II (Munich, C.F. Müller, 2008) 37; Bertrand Mathieu, ‘Les rapports normatifs entre le droit communautaire et le droit national’, (2007) 72 Revue Française de Droit Constitutionnel 675.

314  Rainer Palmstorfer, Laurent Sermet and Erich Vranes the primacy of international agreements vis-à-vis French acts of Parliament58 and is competent, ­therefore, to review French legal acts as to their compliance with international and/or EU obligations. Since a ruling delivered in 1975,59 the Constitutional Council has persistently taken the view that it is not competent to examine whether French laws are in conformity with French international undertakings. An important exception has, however, been introduced by the Constitutional Council in 2004, when it held that it is obligated, due to Article 88-1 FC, to review whether a French act implementing an EU directive complies with unconditional and precise requirements of the directive in question.60 Since that decision, the Constitutional Council ceased referring to Article 55 FC in its rulings on the relationship between EU and French law. Instead, Article 88-1 FC has been given a cardinal place in the Constitutional Council’s argumentation in such cases.61 By contrast, the Council of State continues to refer to Article 55 FC along with Article 88-1 FC. The underlying legal reason for the latter approach has so far remained obscure.62

7.  Article 88-1 FC The central constitutional provision dedicated to EU integration is Article 88-1 FC. It has been inserted into the FC due to the Constitutional Council’s 1992 Maastricht ruling. Article 88-1 FC was subsequently modified to account for the Lisbon Treaty developments.63 Since the entry into force of the Lisbon Treaty in December 2009, Article 88-1 FC is worded as follows: The Republic shall participate in the European Union constituted by States which have freely chosen to exercise some of their powers in common by virtue of the Treaty on European Union and of the Treaty on the Functioning of the European Union, as they result from the treaty signed in Lisbon on 13 December, 2007.

This clause was inserted against the background of the specificity of the EU legal order as compared to public international law.64 Indeed, since the Maastricht I ruling it was considered that EU law is a ‘distinct legal order which, although it is part of the legal system of the different Member States of the Communities, that does not belong to the French institutional order’.65 Article 88-1 FC had originally been understood as a mere declaratory introductory provision to Title XV of the Constitution.66 It was the landmark ruling of the Constitutional Council in the Économie numérique case, which attributed to this clause a quite different legal status, and a primary role in EU integration. According to this judgment, which dealt with a French Act of Parliament implementing an EU directive, Article 88-1 FC gives rise to a constitutional

58 See also Formery, ‘Constitution’ (n 42) 119. 59 CC 74-54 (n 9). 60 CC 2004-496 DC of 10 June 2004 (Économie numérique) ECLI:FR:CC:2004:2004.496.DC. 61 Cf eg CC 2004-505 (n 18); CC 2007-560 (n 20); on this see eg Charpy, ‘Droit constitutionnel’ (n 44) 626. 62 Charpy, ‘Droit constitutionnel’ (n 44) 626 and 628 ff. 63 Constitutional Act No 2008-103 of 4 February 2008. 64 In particular, this specificity lies in the direct effect and the primacy of EU law. See ECJ 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66; ECJ 05.02.1963 26/62 (Van Gend en Loos/Administratie der Belastingen) ECLI:EU:C:1963:1. 65 CC 92-308 (n 13) para 34. A will be shown below, the Council also stresses the distinctiveness of the EU legal order in CC 2004-505 (n 18) para 11. 66 Cf Jacqueline Dutheil de la Rochère, ‘Comment on Conseil Constitutionnel, Decision No. 2004-496 of June 2004’, (2005) 42 CML Rev 859.

France  315 obligation to transpose EU directives. In the central passage of the ruling, the Constitutional Council, relying on Article 88-1 FC, held: With respect to the French act at issue, the Constitutional Council furthermore found that, given that the act’s litigious clauses restricted themselves to drawing the necessary consequences from unconditional and precise provisions of the directive, it was incompetent to review the French legal provisions.67 In a subsequent decision, the Constitutional Council confirmed that to the extent an implementing act does not restrict itself to drawing the necessary conclusions from unconditional and precise provisions of a directive, it is competent to scrutinize the implementing act under French constitutional law.68

This judgment is remarkable for several reasons. Above all, the obligation at issue – to transpose an EU directive – is regarded also as an obligation under French constitutional law by virtue of Article 88-1 FC. As a consequence, it is for the Constitutional Council also to examine whether a transposing statute is obviously incompatible with the directive to be transposed.69 The Constitutional Council was quick to add, however, that this constitutional obligation exists only to the extent that there is no express contrary provision in the FC. If this is not the case, then Article 88-1 FC is understood, by the Constitutional Council, as a sufficient basis to regard itself as incompetent to review the implementing act and to recognise the ECJ as being solely competent to interpret and assess the validity of the directive under EU law.70 As indicated, this conclusion only applies to the extent that the French legislator has no margin of manoeuvre due to the requirements of the directive (ie to the extent that the implementing merely draws the necessary consequences from precise and unconditional directive requirements). This ruling has been further elaborated upon in a number of subsequent cases,71 culminating in the Constitutional Council’s 2004 Constitutional Treaty judgment. In that judgment, on the basis of Article 88-1 FC, the Constitutional Council for the first time explicitly regarded EU law as a legal order that is ‘integrated into the domestic legal order and distinct from the international legal order’.72 This ruling has also been read as recognising Article 88-1 FC as a general constitutional basis for, and a potential limit to,73 the primacy of EU law.74 It has also been submitted that it follows from Article 88-1 FC, as interpreted in this judgment, that France could not be integrated into a European federal state.75 This new line of case law has set off an intense debate in French academia.76 Nonetheless, there remain several uncertainties. In particular, it is still uncertain whether the Constitutional Council’s approach of regarding the EU obligation of implementing a directive also as an obligation under French constitutional law is transposable to other obligations stemming from EU law.

67 CC 2004-496 (n 60) para 9 (‘Considérant que … les 2 et 3 du I de l’article 6 se bornent à tirer les conséquences n ­ écessaires des dispositions inconditionnelles et précises du 1 de l’article 14 de la directive susvisée sur lesquelles il n’appartient pas au Conseil Constitutionnel de se prononcer; que, par suite, les griefs invoqués par les requérants ne peuvent être utilement présentés devant lui’). 68 CC 2004-497 DC of 1 July 2004 (Loi relative aux communications électroniques et aux services de communication audiovisuelle) ECLI:FR:CC:2004:2004.497.DC para 20. 69 CC 2006-543 of 30 November 2006 (Secteur de l’énergie) ECLI:FR:CC:2006:2006.543.DC para 7. 70 This ruling, thus, can also be seen as a clear commitment of the Constitutional Council to enter in a judicial dialogue with the ECJ. 71 CC 2004-497 (n 68); CC 2004-498 of 29 July 2004 (Bioéthique) ECLI:FR:CC:2004:2004.498.DC. 72 CC 2004-505 (n 18) para 11. 73 The question of the constitutional limits to the primacy of EU law will be discussed below. 74 Cf CC 2004-505 (n 18) paras 9–13; on this see below. 75 Azoulay and Agerbeek, ‘Comment’ (n 36) 884. 76 See the references in Vranes, ‘Constitutional Foundations’ (n *) 534 n 54.

316  Rainer Palmstorfer, Laurent Sermet and Erich Vranes Moreover, it remains to be seen to what extent the Constitutional Council will use Article 88-1 FC as a basis for other EU integration-related constitutional obligations that might be inferred in its future jurisprudence.77 Some authors already see Article 88(1) FC as the general basis for all legal effects of EU law.78 Despite these ambiguities, the spécificité of EU law today seems recognised to a considerably greater extent in decisions of the French judiciary than before the Article 88(1) FC-based judicial turnaround.

8.  Articles 88-2 to 88-7 FC The new provisions of Articles 88-2 and 88-3 FC were drafted to overcome various inconsistencies between the FC and the Maastricht Treaty that were identified by the Constitutional Council (EMU, visas, participation of other European Member States’ citizens in municipal elections). Article 88-2 FC, at the time of its insertion in 1992, served as a constitutional basis for transfer of powers that were necessary in the context of the establishment of the EMU and the EU’s rules on the crossing of external borders. After several amendments,79 it now addresses the European arrest warrant. This reference is due to an opinion of the Council of State, which considered the Framework Decision to be in conflict with the Constitution. Article 88-2 FC reads as follows: Statutes shall determine the rules relating to the European arrest warrant pursuant to acts adopted by the institutions on the European Union.

Article 88-3 FC deals with municipal elections and according to it, Subject to reciprocity and in accordance with the terms of the Treaty on European Union signed on 7 February 1992, the right to vote and stand as a candidate in municipal elections shall be granted only to citizens of the Union residing in France. Such citizens shall neither hold the office of Mayor or Deputy Mayor nor participate in the designation of Senate electors or in the election of Senators. An Institutional Act passed in identical terms by the two Houses shall determine the manner of implementation of this article.

This provision has its roots in the Constitutional Council’s 1992 Maastricht ruling. In that ruling the Council found that the EU rules on municipal elections violated the FC. In particular, it inferred from Article 3 FC (according to which only French citizens are entitled to vote) and Article 24 FC (according to which the Senate, which is indirectly based on municipal elections, participates in the exercise of national sovereignty) that the right to vote and to stand as a candidate for municipal elections was reserved for French citizens only. This restriction due to national sovereignty was overcome by the wording of Article 88-3 FC.80 Article 88-4 FC, inserted in 1992 and subsequently amended in 1999 and 2007, is meant to compensate the French Parliament for its loss of competences in the context of EU integration.81 The provision aims to safeguard the involvement of the French Parliament in the drafting of EU acts. It is worded as follows: The Government shall lay before the National Assembly and the Senate drafts of or proposals for Acts of the European Communities and the European Union containing provisions which are of a statutory nature as soon as they have been transmitted to the Council of the European Union. It may also



77 Charpy,

‘Droit constitutionnel’ (n 44) 626. with further references. 79 Cf eg Formery, ‘Constitution’ (n 42) 162; Carcassonne, ‘Constitution’(n 42) 385. 80 Cf eg Formery, ‘Constitution’ (n 42) 62 ff and 162–63. 81 Cf ibid, 163. 78 Ibid,

France  317 lay before them other drafts of or proposals for Acts or any instrument issuing from a European Union Institution. In the manner laid down by the Rules of Procedure of each House, resolutions may be passed, even if Parliament is not in session, on the drafts, proposals or instruments referred to in the preceding paragraph. A committee in charge of European affairs shall be set up in each of the Houses of Parliament.

In its current wording, the provision takes into account the conclusions of the Constitutional Council’s 2004 ruling on the Constitutional Treaty and its 2007 Lisbon ruling, pursuant to which the French Constitution did not enable the French Parliament to exercise the rights envisaged for national parliaments in the TECE and the Lisbon Treaty, respectively.82 Article 88-5 FC, introduced in 2005 and modified in 2008,83 makes future EU accessions in principle dependent on a positive referendum: Any Government Bill authorizing the ratification of a treaty pertaining to the accession of a state to the European Union shall be submitted to referendum by the President of the Republic. Notwithstanding the foregoing, by passing a motion adopted in identical terms in each House by a three-fifths majority, Parliament may authorise the passing of the Bill according to the procedure provided for in paragraph three of article 89.

This provision envisages a possible accession by Turkey, as becomes clear when it is read in conjunction with Article 47 of the loi constitutionnel of 23 July 2008, by virtue of which Article 88-5 FC is rendered inapplicable to accessions that result from an EU intergovernmental conference whose meeting was decided by the European Council before July 1 2004, a clause which was meant to cover Bulgaria, Romania, and Croatia.84 Articles 88-6 and 88-7 FC, which entered into force on 1 December 2009, were included in the Constitution so as to accommodate the conclusions of the Constitutional Council’s Lisbon ruling. Article 88-6 FC lays the constitutional groundwork for the French Parliament’s involvement in the control of subsidiarity, as envisaged by the Lisbon Treaty: The National Assembly or the Senate may issue a reasoned opinion as to the conformity of a draft proposal for a European Act with the principle of subsidiarity. Said opinion shall be addressed by the President of the House involved to the Presidents of the European Parliament, the Council of the European Union and the European Commission. The Government shall be informed of said opinion. Each House may institute proceedings before the Court of Justice of the European Union against a European Act for non-compliance with the principle of subsidiarity. Such proceedings shall be referred to the Court of Justice of the European Union by the Government. For the purpose of the foregoing, resolutions may be passed, even if Parliament is not in session, in the manner set down by the Rules of Procedure of each House for the tabling and discussion thereof. Such proceedings shall be obligatory upon the request of sixty Members of the National Assembly or sixty Senators.

Article 88-7 FC empowers the Parliament to oppose modifications deriving from the simplified revision procedure and modifications in the field of judicial cooperation on civil matters: Parliament may, by the passing of a motion in identical terms by the National Assembly and the Senate, oppose any modification of the rules governing the passing of Acts of the European Union in cases

82 CC 83 Cf

2004-505 (n 18) paras 37–41 and CC 2007-560 (n 19) paras 28–32. loi constitutionnelle of 1 March 2005 and loi constitutionnelle of 23 July 2008, respectively. also Formery, ‘Constitution’ (n 42) 164 f.

84 See

318  Rainer Palmstorfer, Laurent Sermet and Erich Vranes provided for under the simplified revision procedure for treaties or under judicial cooperation on civil matters, as set forth in the Treaty on European Union and the Treaty on the Functioning of the European Union, as they result from the treaty signed in Lisbon on December 13, 2007.

III.  Constitutional Limits to EU Integration A.  Limits to the (Further) Transfer of Powers to the EU through Treaty Amendments 1.  Introductory Remarks The limits to the (further) transfer of competences through treaty amendments have been clarified, to a considerable extent, in the jurisprudence of the Constitutional Council, whose conclusions are reflected in the FC, as repeatedly modified to take account of these rulings.

2.  The Constitutional Yardstick (Norms of Reference) The yardstick that is regarded as relevant by the Constitutional Council for the constitutional review of amendments of primary EU law has been carved out in its rulings on the Maastricht and Amsterdam Treaties, the TECE and the Lisbon Treaty. In that regard the Constitutional Council explicitly referred to the 1789 Déclaration, as ‘confirmed and completed’ by the Preamble of 1946, the principle of national sovereignty enshrined in Article 3 of the Déclaration and Article 3 FC, paragraph 15 of the Preamble of the Constitution of 1946 which permits limitations to national sovereignty, and Article 53 FC.85 Since its Amsterdam judgment, the Constitutional Council has consistently included Article 88-1 FC in this group of explicit norms of reference.86

3.  The Ensuing Constitutional Barriers a.  The Barriers Abstractly Defined From these provisions, the Constitutional Council inferred in 1992 that national sovereignty does not per se constitute an obstacle preventing France from concluding international agreements ‘for participation in the establishment or development of a permanent international organisation enjoying legal personality and decision-making powers on the basis of transfers of powers decided on by the Member States, subject to reciprocity’.87 After its judicial reorientation in 2004, and apparently on the basis of the new role played by Article 88-1 FC, the Constitutional Council has slightly reworded its conclusion. In particular, it emphasised that ‘texts having constitutional value enable France to participate in the creation and development of a permanent European organization, vested with separate legal personality and powers of decision-making by reason of the transfer of powers agreed to by the Member States’.88 However, in its 2007 Lisbon ruling, the 85 Cf CC 92-308 (n 13) paras 9 ff; CC 97-394 (n 17) paras 1 ff; CC 2004-505 (n 18) paras 1 ff; CC 2007-560 (n 19) paras 3 ff. 86 Cf CC 2004-505 (n 18) para 11. 87 CC 92-308 (n 13) para 13; see also CC 97-394 (n 17) para 6. 88 CC 2004-505 (n 18) para 6.

France  319 Constitutional Council chose a more restrictive formulation, underlining the supreme position of the French Constitution in matters of EU integration: While confirming the place of the Constitution at the summit of the domestic legal order, [the aforementioned] constitutional provisions enable France to participate in the creation and development of a permanent European organisation vested with a separate legal personality and decision-taking powers by reason of the transfer of powers agreed to by the Member States.89

This formulation leads us directly to the question of the legal barriers to changes of primary EU law that are inferred by the Constitutional Council from French constitutional law. From the aforementioned constitutional provisions, the Constitutional Council has concluded that primary EU law must not contain ‘a clause running counter to the Constitution’, must not impact ‘constitutionally guaranteed rights and freedoms’, nor must it ‘adversely affect the fundamental conditions of the exercising of national sovereignty’ (‘conditions essentielles d’exercice de la souveraineté nationale’), unless the Constitution is revised beforehand.90 In terms of legal reasoning, this essential conclusion is arrived at in a fairly sudden and somewhat obscure manner, given that the Constitutional Council confines itself to making this inference from a mere listing of the aforementioned constitutional provisions. Arguably, this is also a reason why it is difficult to assess matters such as the constitutional function(s) that Article 88-1 is playing in EU integration since the Constitutional Council’s new judicial approach. Moreover, in terms of legal content, it is evident that this barrier, abstractly defined as it is, provides rather limited guidance as to the exact constitutional borderlines for future developments in EU treaty law. This is because the notion of ‘fundamental conditions of the exercising of national sovereignty’ clearly depends on case-by-case concretisation in the past and future jurisprudence of the Constitutional Council.91 The next section turns to the guidelines that have so far been developed in the Constitutional Council’s case law. Sections c. and d. that follow it then deal with further barriers to changes of primary EU law. b.  ‘Fundamental Conditions’: Concretisations in Case Law Examining whether the fundamental conditions of the exercising of national sovereignty are met requires addressing three questions. First, does the transferred competence form a constituent element of national sovereignty? Second, is this competence totally or partially transferred? Third, does France have the power to oppose decisions at the international level in the field of jurisdiction transferred? Two indications of the meaning of the clause ‘fundamental conditions of the exercising of national sovereignty’ can be derived already from the Constitutional Council’s Maastricht ruling. The first of them is not, however, very informative as to the underlying legal reasoning. According to it, after an – arguably purely descriptive – overview of the EMU provisions contained in the former EC treaty, the Constitutional Council briefly held that these provisions have the effect of depriving Member States of competences in a field that touches upon conditions that are essential for the exercise of sovereignty.92 The second indication is considerably more helpful. It makes it clear that the ‘fundamental conditions of the exercising of national



89 CC

2007-560 (n 19) para 8. para 9. 91 On this see also Azoulay and Agerbeek, ‘Comment’ (n 36); J Dutheil de la Rochère, ‘Comment’ (n 66). 92 CC 92-308 (n 13) para 43. 90 Ibid,

320  Rainer Palmstorfer, Laurent Sermet and Erich Vranes sovereignty’ are infringed – in casu, in the field of the measures relating to the entry and movement of persons – when the unanimity rule is abandoned in favour of qualified majority voting.93 This barrier has been revisited in a more general manner in later Constitutional Council jurisprudence. In that jurisprudence the Council held that further amendments of the FC will be required for any transfer of competences which jeopardises the fundamental conditions of the exercising of sovereignty either because that transfer (i) does not relate to a transfer already permitted in the FC or because (ii) it is a modification of the exercise of competences already transferred.94 The first barrier follows from the fact that the French constitutional legislator has typically (with the exception of the Nice Treaty) reacted to amendments on the level of primary EU law by authorising transfers of competences only as required by the concrete EU Treaty. Thus, for example, after the Constitutional Council’s Maastricht ruling, Article 88-2 FC was inserted, which explicitly permitted transfers of powers only to the extent that was necessary e.g. for the establishment of EMU ‘in accordance with the Treaty on European Union signed on 7 February 1992’ (the Maastricht Treaty). This type of static reference to a fixed point in time – ie a given EU amendment treaty – necessitated repeated FC amendments that are ‘synchronised’ with changes in primary EU law. Therefore, the Constitutional Council’s decisions on the Amsterdam Treaty, the TECE and the Lisbon Treaty have each indicated a series of provisions in EU amendment treaties that required FC amendments. In the Lisbon judgment, these fields of EU action have been referred to as competences that are ‘inherent to national sovereignty’: The provisions of the Treaty of Lisbon which transfer to the European Union under the ‘ordinary legislative procedure’ powers inherent in the exercising of national sovereignty require a revision of the Constitution. This applies to Article 75 of the Treaty on the Functioning of the European Union as regards the fight against terrorism and related activities, to Article 77 as regards border controls, to d) of paragraph 2 of Article 79 as regards the fight against trafficking in human beings, to Article 81, as regards judicial cooperation on civil matters and to Articles 82 and 83 thereof as regards judicial cooperation in criminal matters, and as regards the powers mentioned in said Articles which do not come under the provisions of Articles 62 and 65 of the Treaty establishing the European Community nor Article 31 and 34 of the Treaty on European Union.95

As regards the second aforementioned clarification (modifications of the exercise of competences already transferred to the EU), the Constitutional Council has underlined that also a loss of the right of Member State initiative in a given field – in conreto: the field of visas, asylum and the free movement of persons96 – can amount to an infringement of the ‘fundamental conditions of the exercising of national sovereignty’. In its Lisbon ruling, the Constitutional Council formulated more generally that in fields of competences that have already been transferred to the EU, but which are ‘inherent’ to national sovereignty, any modification of the EU rules on decision-making – namely changes from unanimity to qualified majority, transfers

93 Ibid, para 49. 94 CC 97-394 (n 17) para 9: ‘It follows that further amendment to the Constitution will be required for the clauses of the Treaty of Amsterdam which transfer powers to the European Community in such a way as to jeopardise the essential conditions for the exercise of national sovereignty, either because these transfers do not relate to European economic and monetary union or the crossing of external borders, or because they lay down conditions not already provided by the Treaty on European Union signed on 7 February 1992 for the exercise of powers the transfer of which was authorised by Article 88-2.’ 95 CC 2007-560 (n 19) para 18. 96 CC 97-394 (n 17) paras 22 ff; on this see also ruling CC 2004-505 (n 18) paras 14 ff.

France  321 of decision-making competences to the EU Parliament and losses of Member State powers of initiative – require FC amendments.97 Likewise, clause passerelle-type provisions are regarded as requiring ­constitutional amendment in fields that are inherent to national sovereignty, as they do not require ratification or approval under French constitutional law, which in turn prevents scrutiny by the Constitutional Council, in line with Article 54 or Article 61(2) FC.98 This reasoning also applies to the general passerelle clause (Article 48[7] TEU), which allows for simplified revision of EU decision-making procedures.99 By contrast, Article 48(6) TEU, which permits simplified revisions of EU internal policy areas, is regarded as being in conformity with French constitutional law. This is because the reference to national constitutional law in Article 48(6) TEU is read by the Constitutional Council as referring to Article 53 FC, which requires parliamentary ratification or approval of international agreements.100 Furthermore, the Constitutional Council has ascribed the protection of fundamental rights to the array of ‘fundamental conditions of the exercise of national sovereignty’. In this respect, it is noteworthy that while the Constitutional Council initially took a rather lenient attitude by declining to see such fundamental conditions as endangered because they ‘are guaranteed by the Court of Justice’.101 The 2004 Constitutional Treaty decision contained the only judicial discussion on the republican tradition and the EU. Was the Charter for Fundamental Rights (the Charter) in conformity with the French principle of secularism?102 The Constitutional Council exercised selective scrutiny of the Charter (eg its eventual impacts on collective rights of groups in the sense of Article 1(3), on the French constitutional principle of secularism, and on restrictions of public access to court hearings).103 The Constitutional Council found that the Charter does not endanger any of these rights. Its reasoning was based above all on frequent references to the explanations of the Praesidium of the EU Convention that drafted the Charter and to the Charter’s general limitation clause, in which the principle of proportionality plays a pivotal role (Article 52 of the Charter; Article II-112 of the Charter as incorporated in the TECE). This line of reasoning arguably risks underestimating the potential for conflict between human rights standards at EU and national level. This is because, on the one hand, the explanations of the Praesidium merely constitute supplementary means of interpretation.104 On the other hand, the risk of conflict arguably is not fully defused by the Charter’s general limitation clause105 and the principle of proportionality enshrined in it. In particular, proportionality decisions are balancing decisions that depend on the concrete circumstances of particular cases and on the standard of deference, which is applied by a tribunal, which may vary from one tribunal to another and between cases.106 Furthermore, the Constitutional Council refers to the case law of

97 CC 2007-560 (n 19) para 20. 98 CC 2004-505 (n 18) paras 33 ff and CC 2007-560 (n 19) paras 23 ff on the role of Art 54 and Art 61 para 2 of the French constitution. 99 CC 2004-505 (n 18) para 35; and CC 2007-560 (n 19) para 27. 100 CC 2007-560 (n 19) para 26. 101 CC 92-308 (n 13) para 17. 102 According to Art 1 FC, ‘France shall be an indivisible, secular, democratic and social Republic’. 103 CC 2004-505 (n 18) paras 14–21. 104 See the introduction to the explanations of the Explanations relating to the CFR [2007] OJ C 303/17. 105 On this cf Stefan Griller, ‘Der Anwendungsbereich der Grundrechtscharta und das Verhältnis zu sonstigen Gemeinschaftsrechten, Rechten aus der EMRK und zu verfassungsgesetzlich gewährleisteten Rechten’, in A Duschanek and S Griller (eds), Grundrechte für Europa. Die Europäische Union nach Nizza (Vienna, Springer, 2002) 131; Erich Vranes, ‘The Final Clauses of the Charter of Fundamental Rights – Stumbling Blocks for the First and Second Convention’, European Integration online Papers no 7 (2003). 106 On this cf Erich Vranes, ‘Der Verhältnismäßigkeitsgrundsatz. Herleitungsalternativen, Rechtsstatus und Funktionen’, (2009) 47 Archiv für Völkerrecht 1.

322  Rainer Palmstorfer, Laurent Sermet and Erich Vranes the European Court of Human Rights, rightly pointing out that the latter recognises the principle of secularism. However, the fact that the Constitutional Council referred to a case that was then still pending was criticised. c. Reciprocity As indicated above, according to paragraph 15 of the Preamble of the Constitution of 1946, limitations to sovereignty are subject to the requirement of reciprocity.107 The reciprocity requirement is considered to oblige France to apply the respective treaty from the moment when the last treaty state has ratified it. This provision gave rise to a – rather brief – examination in the Maastricht judgment, where it was held that the Maastricht treaty fulfilled the reciprocity requirement, as the treaty and its protocols would take effect only after the last instrument of ratification was deposited.108 Hence, there was in particular no scrutiny as to the reciprocity of the substantive contents of the obligations entered into by the various Member States. It appears that this criterion has not been reapplied in later rulings.109 Against this background also Article 55 FC has to be mentioned. This provision makes sure that the reciprocity requirement is continuously safeguarded in the course of the application of a treaty by the other parties. Concerning Article 55 FC, and against the background of Title XV, it is questionable whether this reciprocity requirement also applies to EU Treaties. d.  Limits to the Primacy of EU Law As the FC has been transformed to accommodate the distinctiveness of EU law from its international roots (Title XV as opposed to Title VI), a doctrinal debate developed on the hierarchical level of EU law from a French constitutional law perspective. Since the amendment of Title XV, quoting expressly the Lisbon Treaty, it should be assumed that EU law, within the limits of specific constitutional constraints, has a constitutional rank in the French legal order.110 And yet, as will be shown below, the primacy of EU law is not fully recognised. The EU principle of primacy has been explicitly addressed in the Constitutional Council’s Constitutional Treaty judgment, since the principle of primacy was expressly mentioned in Article I-6 TECE. The Constitutional Council held, first, that this treaty would remain an international agreement, which – according to its Article I-5 – would neither impact the existence of the FC, nor its place ‘at the summit of the domestic legal order’.111 Moreover, from the declaration on the primacy of EU law that was annexed to the TECE, the ruling inferred that the ambit of the principle of primacy was to remain the same as before.112 Against this background and the TECE provisions more generally, and, in particular, in view of the interplay between Articles I-5 and I-6 TECE, the Constitutional Council concluded that the TECE would not modify the scope of the principle of primacy ‘as it results from Article 88-1 of the French constitution’ (‘la portée du

107 On this cf also Delphine Dero, La réciprocité et le droit des Communautés et de l’Union européennes (Brussels, Bruylant, 2006). 108 CC 92-308 (n 13) para 16. 109 Cf Conseil Constitutionnel, Jurisprudence du Conseil Constitutionnel. Tables d’analyses au 18 juin 2010, in particular 980 ff. 110 In this respect, the Lisbon Treaty differs from international agreements under Art 55 FC. See, eg, CC 2017-749 DC of 31 July 2017 (CETA) ECLI:FR:CC:2017:2017.749.DC para 44. 111 CC 2004-505 (n 18) paras 9 f. 112 Ibid, para 12.

France  323 principe de primauté du droit de l’Union telle qu’elle resulte … de l’article 88-1’).113 In other words, Article 88-1 FC is not only seen as the general basis for the effects of EU law, but quite obviously also as a limit to the scope of the principle of primacy.114 (It appears noteworthy in this central respect that the official English translation of the ruling appears misleading, as it seems to imply quite the contrary, namely that the scope of the principle of primacy is derived from EU law and is merely ‘duly acknowledged by Article 88-1 of the Constitution’).115 Hence, it appears that the Constitutional Council does not only insist on the supreme position of the FC.116 It also – in particular, in view of its reference to the interplay between Articles I-5 and I-6 TECE – regards the principle of primacy as being limited by the national identity that is inherent in a Member State’s fundamental political and constitutional structures in line with Article I-5 TECE, and as now incorporated in Article 4(2) TEU. Although the CJEU has to take into account Article 4(2) TEU in its rulings, in recent years it has tended to mitigate the consequences of the primacy of EU law by displaying increased ­deference to domestic constitutional concerns in cases such as Schmidberger and Omega,117 it does not seem, however, that the French constitutional barriers for the principle of primacy are per se in line with EU law.118 This is because the CJEU – despite this judicial rapprochement – has not yet desisted from its view that EU law has primacy also over national constitutional law. What is more, the elaborations of the ECJ in Sayn-Wittgenstein clearly show that even if specific parts of national law qualify as expressions of national identity under Art 4(2) TEU, these parts will have to be weighed against other interests protected by EU law.119 Another recent judicial development should be mentioned in this context, as it can be understood as a variant of the issue of the primacy of EU law. In a famous line of jurisprudence that started with the 1978 Cohn-Bendit ruling,120 the Council of State had consistently declined to follow ECJ’s case law on direct effect of non-transposed directives. Although most practical effects of this line of case law have been mitigated by French tribunals through a series of ­judicial techniques (such as the recognition of state liability or the obligation to interpret French law in accordance with directives), it was not before its 30 October 2009 ruling in Perreux that the Council of State formally followed the ECJ on this issue.121 This approach was confirmed in 2011.122 In sum, while the primacy of EU law over French infra-constitutional law is firmly recognised by the French judiciary,123 EU law has not been recognised as having primacy vis-à-vis the FC. This follows not only from the jurisprudence of the Constitutional Council,124 but also from rulings by the Council of State125 and the Court of Cassation.126 However, the stance taken by

113 Ibid, para 13. 114 See also Charpy, ‘Droit constitutionnel’ (n 44) 626 with further references. 115 Official translation. 116 CC 2004-505 (n 18) paras 9–10. 117 ECJ 12.06.2003 C-112/00 (Schmidberger) ECLI:EU:C:2003:333; ECJ 14.10.2004 C-36/02 (Omega) ECLI:EU:C:2004:614. 118 For a contrary view see Mathieu, ‘Rapports normatifs’ (n 57) 680 f. 119 CJEU 22.12.2010 C-208/09 (Sayn-Wittgenstein) ECLI:EU:C:2010:806, para 83. 120 CoS decision 11604 of 22 December 1978 (Cohn-Bendit) ECLI:FR:CEASS:1978:11604.1978:1222. 121 CoS decision 298348 of 30 October 2009 (Perreux) ECLI:FR:CEASS:2009:298348.20091030. This line of case law and the judicial reorientation have attracted a considerable number of comments in academic writings. See the literature listed in Vranes, ‘Constitutional Foundations’ (n *) 545 in n 99. 122 CoS opinion 345978 and 346612 of 21 March 2011 (MM J and T) ECLI:FR:CESSR:2011:345978.20110321. 123 Cf above, section II.C.4. 124 Ie, the rulings CC 92-308 (n 13); CC 97-394 (n 17); CC 2007-560 (n 19). 125 CoS decisions 200286 of 30 October 1998 (Sarran) ECLI:FR:CEASS:1998:200286.19981030; CoS 287110 (n 21). 126 Court of Cassation, decision 99-60274 of 2 June 2000 (Fraisse).

324  Rainer Palmstorfer, Laurent Sermet and Erich Vranes the French judiciary may have been somewhat moderated by its new approach to secondary EU law, EU directives in particular, which are now reviewed in a more limited manner under French constitutional law. This leads us to the constitutional barriers for secondary EU law, which are addressed in the next section.

B.  Scrutiny of Secondary Legislation, Especially Ultra Vires Doctrine 1.  Scrutiny by the Constitutional Council Until 2004, it was not quite clear whether the Constitutional Council was prepared to constitutionally review secondary EU law.127 It was the aforementioned judicial turnaround starting in 2004 which has shed more light on this issue, as it has also clarified the constitutional barriers for secondary EU law. Thus, it not only follows from the Constitutional Council’s respective rulings, which have dealt with the French acts implementing EU directives, that the duty of implementation constitutes a duty under French constitutional law, but also that there are barriers to that duty stemming from the FC. It ensues from the Constitutional Council’s 2004 jurisprudence – which has already been briefly addressed above and which has to some extent been refined meanwhile (cf in the following text) – that this obligation cannot override an express conflicting constitutional provision (‘disposition expresse contraire de la Constitution’).128 This requirement has since been modified by the Constitutional Council to a contravention of a ‘rule or principle inherent in the constitutional identity of France’.129 If there is no such provision, the ECJ is regarded as solely competent to monitor whether a directive respects the competences set forth in the treaties and the fundamental rights guaranteed at EU level.130 This appears to imply that the Constitutional Council will review directives under the aforementioned constitutional yardstick. It also follows from this jurisprudence that the Constitutional Council does not examine provisions in a national implementing act that is limited to drawing the necessary consequences from unconditional and precise requirements of a directive.131 If, however, the requirements laid down in a directive leave a margin of manoeuvre, the Constitutional Council will review the respective implementing act.132 This is a distinction which, according to the Secretary General of the Constitutional Council is influenced by the jurisprudence of the Austrian Constitutional Court.133 Moreover, the Constitutional Council appears to have made its review dependent on whether a given fundamental right is equally protected at EU level.134 In the course of a vigorous discussion135 of this new line of jurisprudence, it has been pointed out by the Secretary General that a ‘disposition expresse contraire de la Constitution’ (which

127 See eg Jean-Philippe Kovar, ‘Commentaire des decisions du Conseil constitutionnel’ (2004) Revue trimestrielle de droit européen 587. 128 CC 2004-496 (n 60) para 7. 129 CC decision 2006-540 DC of 27 July 2006 (Droit d’auteur) ECLI:FR:CC:2006:2006.540.DC para 18. 130 Ibid. 131 CC 2004-496 (n 60) para 9 and CC 2004-498 (n 71) para 7. 132 CC 2004-497 (n 68) para 20. 133 On this cf Dutheil de la Rochère, ‘Comment’ (n 66) 864. 134 CC 2004-498 (n 71); on this see also Charpy, ‘Droit constitutionnel’ (n 44); Olson and Cassia, Droit international (n 57) 55 ff. 135 See the literature mentioned in Vranes, ‘Constitutional Foundations’ (n *) 547 in n 113.

France  325 will make the Constitutional Council exercise its competence) must be FC-specific. This means that there must be no equivalent in EU fundamental rights and general principles of EU law.136 An example given by a Member of the Constitutional Council is the French principle of secularism.137 This enquiry into whether there exists equivalent protection at EU level arguably appears to be reflected also in the more recent jurisprudence of the Council of State. The Constitutional Council’s line of jurisprudence was further elaborated in 2006 in a ruling on the statute pertaining to copyright and related rights in the information society which implemented Directive 2001/29/EC.138 The Constitutional Council first reconfirmed that the implementation of directives ‘results from a constitutional obligation’,139 which is monitored by the Constitutional Council.140 This control is then, however, subjected to a ‘double limit’. First, the directive’s implementation must not ‘run counter to a rule or principle inherent to the constitutional identity of France, except when the constituting power consents thereto’.141 The notion of constitutional identity is still not clear.142 Secondly, the Constitutional Council may only declare a legislative clause contrary to Article 88-1 FC if it is obviously incompatible to the clause of the directive. It can be said, however, that the judicial control of the Constitutional Council is marked by judicial self-restraint. This first limit may give rise to a review of the EU directive itself under French constitutional law.143 This limit, therefore, is not used in the treaty assessment context (Article 54 FC) but comes into play in the context of the transposition of directives. Secondly, given that the Constitutional Council considers itself unable to request a preliminary ruling from the ECJ due to the timeframe set forth in Article 61 FC, it holds that it can ‘only find a statutory provision unconstitutional under Article 88-1 FC if this provision is obviously incompatible with the Directive which it is intended to transpose’.144 It is, therefore, incumbent on national courts, if need be, to refer a matter to the ECJ for a preliminary ruling.145 Implementing provisions that restrict themselves to drawing necessary conclusions from unconditional and precise provisions of a directive continue not to be reviewed by the Constitutional Council after this 2006 clarification of its case law.146 It has to be pointed out that the Constitutional Council resorted to the concept of constitutional identity also in its ex post jurisdiction.147 All in all, there are only two occasions in which the Constitutional Council would examine the constitutionality of a statute

136 See Dutheil de la Rochère, ‘Comment’ (n 66) 866. 137 Olivier Dutheillet de Lamothe, Le contrôle de conventionnalité (intervention lors de la visite à la Cour constitutionnelle italienne le 9 mai 2008; www.conseil-constitutionnel.fr); see also Dutheil de la Rochère, ‘Comment’ (n 66), who refers to French prohibition on automatic extradition in relation to political crimes as an example. 138 Directive 2001/29/EC of 22 May 2001 on the harmonisation of certain aspects of copyright and related rights in the information society [2001] OJ L 167/10. 139 CC 2006-540 (n 129) para 17. 140 Ibid, para 18. 141 Ibid, para 19. 142 Concerning the concept of constitutional identity see, for example, Monica Claes and Jan-Herman Reestman, ‘The Protection of National Constitutional Identity and the Limits of European Integration at the Occasion of the Gauweiler Case’, (2015) 16 German Law Journal 917, 951 f. 143 Cf CC 2006-540 (n 129) para 28 (‘the abovementioned Directive of May 22nd 2001 does not run counter to any rule or principle inherent to the constitutional identity of France’). 144 Ibid, para 20. 145 Ibid. 146 Ibid, para 35. 147 See CC 2010-79 QPC of 17 December 2010 (Transposition of a directive) ECLI:FR:CC:2010:2010.79.QPC para 3: ‘in the absence of contestation of a rule or principle inherent to the constitutional identity of France, the Constitutional Council is not competent to control the conformity with the rights and freedoms guaranteed by the Constitution of legislative provisions that are limited to the necessary consequences of unconditional and precise provisions of a European Union directive’.

326  Rainer Palmstorfer, Laurent Sermet and Erich Vranes transposing a directive. First, the statute must be ‘obviously incompatible’148 with the directive and, second, the statute overrides an express conflicting constitutional provision. A series of questions arise from this judicial adjustment. On the one hand, there is the issue of the content of identité constitutionnelle. In academic writings, it has been submitted that it encompasses principles such as the indivisibility of the French Republic, secularism, democracy, decentralisation, the welfare state and French participation in EU integration.149 According to Mathieu, the constitutional identity may comprise objective principles (eg secularism) as well as fundamental rights, which raises the question, not yet clarified by the Constitutional Council, as to whether the French constitutional identity is (co-)defined by the mere formal existence of a given fundamental right or by the specific interpretation that it has received in French jurisprudence.150 In this context, one needs to be aware of the fact that the constitutional identity can in any case be modified by the constituting power.151 On the other hand, the question arises of the probable impact of this presumptive widening of the French constitutional reserve, as compared to the afore-described 2004 case law. From a functional perspective, the concept of constitutional identity serves to strike a balance between, on the one hand, the constitutional participation of France in the EU and, on the other hand, the preservation of core elements of French constitutionalism. It may be understood as an expression of French sovereignty. Against the background of constitutional identity, it is questionable whether or not the Constitutional Council is competent to raise this issue in the framework of its ex post review (QPC). The issue of constitutional identity was also touched upon in the Constitutional Council’s ruling in CETA.152 There, the Constitutional Council considered that, in the case of an international treaty conferring exclusive jurisdiction on the Union, its review should be limited to reviewing clauses contrary to the constitutional identity of France.153 Thus, the Constitutional Council applies the same standards as in the case of statutes transposing a directive and, therefore, exercises only a restricted constitutional review.154 In its CETA ruling, the Constitutional Council did not take the opportunity to elaborate on the notion of constitutional identity. By finding no violation of a constitutional provision, the Council concluded that CETA cannot violate the constitutional identity, either.155 For the Council, a closer analysis of the issue of constitutional identity was, therefore, not necessary. By contrast, as regards CETA provisions which fall within a competence shared between the EU and the Member States or within a competence belonging solely to the Member States, the

148 Ibid, para 20. 149 Florence Chaltiel, ‘Droit constitutionnel et droit communautaire. Nouvelle précision sur les rapports entre le droit constitutionnel et le droit communautaire La décision du Conseil constitutionnel du 27 juillet 2006 sur la loi relative aux droits d’auteurs’, (2006) 68 Revue Française de Droit Constitutionnel 837. 150 Mathieu, ‘Rapports normatifs’ (n 57) 690. 151 Cf the proviso ‘except when the constituting power consents thereto’ in para 19 of the Constitutional Council’s ruling. This is also emphasised eg by Mayer et al, ‘Vorrang’ (n 22) 71; and Mathieu, ‘Rapports normatifs’ (n 57) 692. 152 CC 2017-749 DC (n 110). See Denys Simon, ‘La ratification du CETA ne nécessitera pas de révision constitutionnelle: bref propos sur la décision du Conseil constitutionnel du 31 juillet 2017’, (2017) 27 (8-9) Europe. Actualité du Droit de l’Union Européenne 1; Jérôme Roux, ‘Accord économique et commercial global (AECG/CETA)’, 91 La Semaine juridique. Édition générale 1821; Baptiste Bonnet, ‘Le Conseil constitutionnel et le CETA’, (2017) 72 Actualité juridique. Droit administratif 2008; Jérôme Roux, ‘La jurisprudence “IVG” fragilisée par inadvertance’, (2017) 193 Recueil Dalloz 2378; Florence Chaltiel, ‘Le traité CETA devant le juge constitutionnel’, (2018) 386 (73) Les Petites Affiches, 6; Joris Larik, ‘Prêt-à-ratifier: The CETA Decision of the French Conseil constitutionnel of 31 July 2017’, (2017) 13 European Constitutional Law Review 759. 153 CC 2017-749 DC (n 110) para 14. 154 However, the CETA ruling was handed down in a procedure under Art 54 FC, whereas a law transposing a directive can be the subject of procedure under Art 61(2) FC and, respectively, Art 61-1 FC (QPC). 155 CC 2017-749 DC (n 110) para 73.

France  327 Constitutional Council examined whether there was a clause violating the FC.156 In other words, unlike the part of CETA falling under EU’s exclusive competence, the mixed agreement part was subject to an unrestricted constitutional review, its yardstick being the whole FC. Even under that review, however, the Constitutional Council did not find such a violation.157

2.  Scrutiny by the Council of State Shortly after the Constitutional Council’s reorientation, the Council of State, too, was confronted with a French legal act implementing an EU directive, which led to its 2007 landmark Arcelor decision. Relying on Article 55 FC, the Council of State held that international obligations cannot take precedence over French constitutional law. Evidently taking account of the Constitutional Council’s new case law, the Council of State inferred a constitutional obligation from Article 88-1 FC to implement EU directives. However, pursuant to the Council of State, the constitutional review of acts implementing precise and unconditional provisions of a directive has to take place under modified conditions. Accordingly, when seized with a question of constitutionality, the Council of State is to examine whether on the level of EU law a rule or a general principle of EU law exists which, in the way it is currently interpreted by the ECJ, guarantees effective protection of the respective French constitutional rule or principle.158 If this is the case, then it is incumbent upon the Council of State to examine whether the directive complies with the relevant rule or principle of EU law and to reject the complaint, unless there are serious doubts, which would require requesting a preliminary ruling from the ECJ. If, by contrast, such a rule or principle does not exist at the EU level, then the Council of State regards itself as competent to examine directly the constitutionality of the implementing measure in question.159 Like the Constitutional Council’s new line of case law, the Council of State’s Arcelor ruling spurred an intense debate in France that has even spilled over to the wider public.160 It is evident that this ruling involves a considerable number of issues both from the perspective of EU and French law, the latter having received quite controversial assessments. As regards the EU law perspective, it is clearly problematic that the Council of State claims to be competent to determine by itself, and without making a preliminary reference to the CJEU, whether equal protection of a particular French right exists at the level of EU law. Such a stance also deprives the ECJ of the possibility, clearly envisaged by primary EU law, to further develop its case law and the EU law acquis.161 As regards French constitutional law, it has been submitted that the ruling’s reference to Article 55 FC indicates that it appears legally impossible, for the Council of State, to recognise the primacy of EU law vis-à-vis French constitutional law. This would imply that EU law could even receive less favourable treatment than international law, given that complaints based on constitutional law can be raised against EU law.162 In this context, it has also been held that the Council of State has confirmed a hard core of sovereignty.163 Moreover, it has been submitted that the fact that the Council of State chose not to employ the notion of constitutional identity, as introduced

156 Ibid, para 13. 157 Among other issues, the Constitutional Council examined whether CETA touched upon matters inherent to national sovereignty. Cf ibid, para 70. 158 CoS 287110 (n 21). 159 Ibid. 160 Cf Mayer et al, ‘Vorrang’ (n 22) 71 with further references on French literature. 161 Ibid, 85 f. 162 Ibidm, 77 with further references to academic writings. 163 Charpy, ‘Droit constitutionnel’ (n 44) 639 ff.

328  Rainer Palmstorfer, Laurent Sermet and Erich Vranes by the Constitutional Council,164 but opted for inquiring into the existence of equivalent protection at EU level, may suggest that the Council of State may not be prepared to fully concur with the Constitutional Council’s new approach to EU law.165 Nonetheless, it should be noted that the Constitutional Council’s approach, too, has been interpreted, by its Secretary General, as encompassing an examination as to whether there is equivalent protection of fundamental rights at EU level.166 Finally, as in the case of the Constitutional Council’s judicial re-orientation, the question remains as to the extent to which the Council of State’s approach can be transferred to other acts of secondary EU law.

IV.  Concluding Remarks • The opening-up of the French legal order towards EU law has been a gradual process. In a turnaround that has occurred in particular since 2004, the French judiciary has explicitly recognised that EU law is integrated into the domestic legal order and is distinct from the international legal order. French authors submit, however, that the Constitutional Council, the Council of State and the Court of Cassation may still be pursuing partly divergent approaches towards EU law. To some extent, this tends to be illustrated also by the current debate on the compatibility of the question prioritaire de constitutionnalité (QPC) with CJEU case law. • The primacy of EU law vis-à-vis French infra-constitutional law appears to be consistently recognised by the French judiciary. • Limits to EU integration are, however, derived from the FC, which is regarded as being placed at the top of the internal legal order. • With respect to primary EU law, the Constitutional Council has made it clear that EU treaty amendments must neither contain a clause running counter to the Constitution and calling into question constitutionally guaranteed rights and freedoms, nor must they adversely affect the fundamental conditions of the exercising of national sovereignty (‘conditions essentielles d’exercice de la souveraineté nationale’), unless the Constitution is appropriately amended beforehand. • This abstractly defined barrier depends on case-by-case concretisation. Relevant indications can be derived from the Constitutional Council’s jurisprudence. According to it, further amendments of the FC will be required for any transfer of competences which jeopardises the fundamental conditions of the exercising of sovereignty either because that transfer (i) does not relate to those already permitted in the constitution or because (ii) modifies the exercise of already transferred competences. • With respect to the first barrier, which appears to be defined in terms of the policy fields affected by a (new) transfer of competences, the Constitutional Council has, in its most recent judgment (concerning the Lisbon treaty) referred to a series of treaty provisions which it regards as ‘inherent to national sovereignty’.



164 Cf

the preceding section. et al, ‘Vorrang’ (n 22) 71. the preceding section.

165 Mayer 166 Cf

France  329 • With respect to the second barrier, which relates to modifications of the exercise of competences already transferred to the EU, the Constitutional Council has emphasised that, as regards such competences which are ‘inherent to national sovereignty’, any modification of the EU rules on decision-making require FC amendments. This modification comprises changes from unanimity to qualified majority, transfers of decision-making competences to the EU Parliament, losses of Member State powers of initiative, and clause passerelle-type provisions including the general passerelle clause (Article 48(7) TEU). • Furthermore, the Constitutional Council has subsumed the protection of fundamental rights under the notion of conditions essentielles d’exercice de la souveraineté nationale. • The principle of primacy of EU law has been regarded as being limited by its traditional scope and by the national identity that is inherent in a Member State’s fundamental political and constitutional structures in line with Article I-5 TECE, a principle that is now incorporated in Article 4(2) TEU. • With respect to the scrutiny of secondary EU law, important developments have taken place in the context of the judicial re-orientation that was started in 2004. • While the Constitutional Council now regards the EU law obligation to implement a directive as an obligation under French constitutional law, it underlined in 2004 that this obligation cannot override an express conflicting constitutional provision. • In 2006, the Constitutional Council reformulated this limit so that the implementation of a directive must not run counter to a rule or principle that is inherent to the constitutional identity of France. The constituting power can, however, consent to necessary amendments of French constitutional law. Moreover, the Constitutional Council restricted its constitutional review of a national act implementing a directive to provisions that are manifestly incompatible with the directive being transposed. It is incumbent on national courts, if need be, to refer a matter to the CJEU for a preliminary ruling. • Since 2007, the implementation of directives is also regarded as a constitutional obligation by the Council of State. According to it, the constitutional review of acts implementing precise and unconditional provisions of a directive has to take place under modified conditions. When seized with a question of constitutionality, the Council of State proceeds to examine whether, on the level of EU law, a rule or a general principle of EU law exists which, in the way it is currently interpreted by the CJEU, guarantees effective protection of the respective French constitutional rule or principle. If this is the case, then the Council of State considers itself competent to examine whether the directive complies with the relevant rule or principle of EU law and to reject the complaint, unless there are serious doubts, requesting a preliminary ruling from the CJEU. If, by contrast, such a rule or principle does not exist at the EU level, the Council of State considers itself competent to examine directly the constitutionality of the implementing measure in question. • This judicial reorientation has spurred an intense legal debate. In this debate it has been submitted inter alia that the Constitutional Council’s approach of curtailing the EU law obligation to implement EU directives through its review under French constitutional law risks infringing EU law. Similarly, it has been held as problematic in terms of EU law that the Council of State claims to be competent to determine by itself, and without making a preliminary reference to the CJEU, whether equal protection of a given FC right exists at the level of EU law. Such a stance also deprives the CJEU of the possibility to further develop its case law and the EU law acquis.

330  Rainer Palmstorfer, Laurent Sermet and Erich Vranes • This judicial reorientation also raises a series of questions under French constitutional law. This is particularly so, given that it has not yet become clear eg which rules and principles comprise the French constitutional identity, nor whether the Council of State’s approach to EU directives is fully in line with that of the Constitutional Council. It is also not yet clear if, and to which extent, their approaches can be transposed from directives to other acts of secondary EU law.

References L Azoulay and R Agerbeek, ‘Comment on Conseil Constitutionnel (French Constitutional Court), Decision No. 2004-505 DC of 19 November 2004, on the Treaty establishing a Constitution for Europe’, (2005) 42 CML Rev 871. B Bonnet, ‘Le Conseil constitutionnel et le CETA’, (2017) 72 Actualité juridique. Droit administratif 2008. G Carcassonne, La Constitution introduite et commentée, 9th edn (Montrouge, Editions du Seuil, 2009). F Chaltiel, ‘Droit constitutionnel et droit communautaire. Nouvelle précision sur les rapports entre le droit constitutionnel et le droit communautaire La décision du Conseil constitutionnel du 27 juillet 2006 sur la loi relative aux droits d’auteurs’, (2006) 68 Revue Française de Droit Constitutionnel 837. F Chaltiel, ‘Le traité CETA devant le juge constitutionnel’, (2018) 386 (73) Les Petites Affiches, 6. C Charpy, ‘Droit constitutionnel et droit communautaire. Le statut constitutionnel du droit communautaire dans la jurisprudence (récente) du Conseil’, (2009) 80 Revue Française de Droit Constitutionnel 621. M Claes and JH Reestman, ‘The Protection of National Constitutional Identity and the Limits of European Integration at the Occasion of the Gauweiler Case’, (2015) 16 German Law Journal 917. D Dero, La réciprocité et le droit des Communautés et de l’Union européennes (Brussels, Bruylant, 2006). J Dutheil de la Rochère, ‘Comment on Conseil Constitutionnel, Decision No. 2004-496 of June 2004’, (2005) 42 CML Rev 859. SL Formery, La constitution commentée, 13th edn (Paris, Hachette Livre, 2010/11). S Griller, ‘Der Anwendungsbereich der Grundrechtscharta und das Verhältnis zu sonstigen Gemeinschaftsrechten, Rechten aus der EMRK und zu verfassungsgesetzlich gewährleisteten Rechten’, in A Duschanek and S Griller (eds), Grundrechte für Europa. Die Europäische Union nach Nizza (Vienna, Springer, 2002) 131. C Haguenau-Moizard, ‘Offene Staatlichkeit: Frankreich’, in von Bogdandy et al (eds), Handbuch Ius Publicum Europaeum vol II (Munich, CF Müller, 2008) 37. JP Kovar, ‘Commentaire des decisions du Conseil constitutionnel’ (2004) Revue trimestrielle de droit européen 587. J Larik, ‘Prêt-à-ratifier: The CETA Decision of the French Conseil constitutionnel of 31 July 2017’, (2017) 13 European Constitutional Law Review 759. B Mathieu, ‘Les rapports normatifs entre le droit communautaire et le droit national’, (2007) 72 Revue Française de Droit Constitutionnel 675. FC Mayer, E Lenski and M Wendel, ‘Der Vorrang des Europarechts in Frankreich’, (2008) 43 Europarecht 63. T Olson and P Cassia, Le droit international, le droit européen et la hiérarchie des normes (Paris, Presses Universitaires de France, 2006). P Pactet and F Mélin-Soucremanien, Droit constitutionnel, 28th edn (Paris, Sirey, 2009). M Rogoff, ‘Fifty Years of Constitutional Evolution in France: The 2008 Amendments and Beyond’, in M Rogoff et al (eds), The Financial Crisis of 2008. French and American Responses. Proceedings of the 2010 Franco-American Legal Seminar (Portland ME, University of Maine School of Law, 2011) 1. J Roux, ‘Accord économique et commercial global (AECG/CETA)’, (2017) 91 La Semaine juridique. Édition générale 1821. J Roux, ‘La jurisprudence “IVG” fragilisée par inadvertance’, (2017) 193 Recueil Dalloz 2378.

France  331 D Simon, ‘La ratification du CETA ne nécessitera pas de révision constitutionnelle: bref propos sur la décision du Conseil constitutionnel du 31 juillet 2017’, (2017) 27 (8-9) Europe. Actualité du Droit de l’Union Européenne, 1. E Vranes, ‘The Final Clauses of the Charter of Fundamental Rights – Stumbling Blocks for the First and Second Convention’, (2003) 7 European Integration online Papers. E Vranes, ‘Der Verhältnismäßigkeitsgrundsatz. Herleitungsalternativen, Rechtsstatus und Funktionen’, (2009) 47 Archiv für Völkerrecht 1. E Vranes, ‘Constitutional Foundations of, and Limitations to, EU Integration in France’, (2013) 19 European Public Law 535.

332

11 Croatia TAMARA ĆAPETA*

I.  Main Characteristics of the National Constitutional System Croatia is a newly independent state. It adopted its first Constitution on 22 December 1990.1 This Constitution served as a basis for separating from Yugoslavia.2 As it was adopted just before the holidays, it is often referred to as the Christmas Constitution. This Constitution aimed at distinguishing the country from the previous, socialist, political regime. It stressed a democratic, pluralistic orientation based on the rule of law, and respect for fundamental rights. These basic values in essence have remained unchanged since the adoption of the Constitution. The first Article of the Constitution defines Croatia as a ‘democratic and social state’. One of the introductory provisions (Article 3) enumerates as the highest values of the Croatian constitutional order: ‘freedom, equal rights, national and gender equality, peace-making, social justice, respect for human rights, inviolability of ownership, conservation of nature and the environment, the rule of law and a democratic multiparty system’. Chapter III of the Constitution, containing 57 articles, lists human rights and fundamental freedoms guaranteed in the Croatian legal order. The Constitution has undergone several amendments.3 From the point of view of European Union (EU) membership, the most important were the ones of 2010.4 They enabled the success of the membership referendum and introduced a new EU Chapter that entered into force on the day of Croatia’s EU accession, 1 July 2013. It is necessary to note that currently there are two versions of the numeration of the Articles of the Constitution. This is due to the two different methods of post-amendment consolidation

* Jean Monnet Professor of EU Law, Head of the EU Law Department at the Faculty of Law of the University of Zagreb. The author would like to thank Lovorka Kušan for her comments on a previous draft of this text. 1 The Official Journal in Croatia is called Narodne novine (hereinafter NN). The 1990 Constitution was published in NN 56/90. It is necessary to note that, as one of the republics of the Socialist Federal Republic of Yugoslavia, Croatia already had its own Constitution even before the decision on independence. 2 Following the May 1991 referendum on separation from Yugoslavia, the separation was accomplished on the basis of the Constitutional decision on the sovereignty and independence of the Republic of Croatia of the Croatian Parliament of (NN 31/91) and the Decisions on the separation from Yugoslavia of 8 October 1991 (NN 53/91). 3 First amendments were introduced in December 1997 (NN 135/97); the second in November 2000 (NN 113/2000); the third in March 2001 (NN 28/2001); the fourth in July 2010 (NN 76/2010) and finally, the most recent amendment was introduced by the popular referendum on December 2013 (NN 5/2014). 4 Amendments were published in NN 76/2010. The Constitution was amended in accordance with the regular amendments procedure. The decision to amend the Constitution has to be made by a two-thirds majority of all members of the Parliament. Such a decision was indeed adopted on 16 June 2010. 133 MPs voted for the amendment, four voted against it, and there was one abstention.

334  Tamara Ćapeta of the Constitution used by the Constitutional Court (CC)5 and the parliamentary committee for legislation, in their respective versions.6 The CC considered in 2011 that the consolidation of different texts of constitutional significance had been implemented in an unconstitutional way.7 Subsequently, the CC produced its own version of the Constitution and demanded that the parliamentary committee correct their version as published in the Official Gazette (NN). This, however, has not (yet) happened. This causes problems in quoting articles of the Constitution. I have decided to quote both numbers if they differ – first the one from the CC’s version, followed by the one from the Official Gazette in brackets.8 As with constitutions of many other new post-socialist democracies, the Constitution also envisages that the sovereignty of the Republic is ‘inalienable, indivisible and non-transferable’ (Article 2). The same Article explains that this does not mean that Croatia cannot participate in associations of states. It must, however, be able to dispose of its sovereignty and be able to withdraw from the association in which it participates. Thus, the Lisbon Treaty provision on the possibility of leaving the EU9 facilitated Croatia’s accession. Under the Constitution, the power is vested in the people, and is exercised either directly by them or through elected representatives. Direct exercise of this power via referenda is a recent practice. The EU membership referendum of 22 January 2012 was the first since the 1991 referendum on Croatian independence and separation from Yugoslavia.10 Thus, a referendum was a modality for deciding on transformative constitutional events. Since joining the EU, however, there have been several citizens’ initiatives for convening referenda that were indirectly connected to the EU, as will be explained below. They sought to achieve legislative and constitutional amendments.11 One of them led to a constitutional amendment inserting a clause that declares marriage a union between a man and a woman.12 Although only 37.9 per cent of citizens participated in the referendum, it still resulted in a constitutional amendment. This result prompted the government to legislate on referenda through a statute, a process that is still ongoing. In the meantime, almost all referendum initiatives found their way to the CC, which had to start developing case law relating to this issue. As I will explain below, some of these CC decisions are also important for the relationship between EU and domestic law. They are important because the CC started developing the ‘constitutional identity’ doctrine in these cases.

5 The version of consolidated text of the Constitution as published by the Constitutional Court of Croatia can be found at www.usud.hr/. 6 NN 85/10. This version is published also on the website of the Croatian Parliament, www.sabor.hr/ustavrepublike-hrvatske. 7 Izvješće U-X-80/2005 od 23/3/2011 o ustavnopravno neprihvatljivim učincima pročišćenih tekstova Ustava Republike Hrvatske, ustavnih zakona, zakona, drugih propisa i općih akata [Report U-X-80/2005 of 23 March 2011 on the constitutionally unacceptable effects of the consolidated versions of the Constitution of the Republic of Croatia, constitutional laws, laws and other acts]. 8 The translation of the Constitution into English exists in both versions. The English version of the text as used by the CC can be found at www.usud.hr/sites/default/files/dokumenti/The_consolidated_text_of_the_Constitution_of_the_ Republic_of_Croatia_as_of_15_January_2014.pdf. The English version, as published by the Croatian Parliament is at www.sabor.hr/en/constitution-republic-croatia-consolidated-text. 9 Art 50 TEU. 10 The referendum was held on the 19 May 1991. The turnout was very high: 83.56%. 11 Constitutional referendum on the definition of marriage, initiative for a legislative referendum on Labour Law; initiative for a constitutional referendum on regulation of elections; legislative referendum on management of roads; legislative referendum on outsourcing, initiative for legislative referendum on Referendum Law, and the latest two initiatives, one for the referendum on the amendments in the electoral law, and the other against the ratification of the Istanbul Convention. 12 Initiated by the conservative citizens’ initiative and backed up by the Catholic Church, a sufficient number of signatures was collected to oblige the Government to organise a referendum. It was held on 1 December 2013 and with a 65.87% majority and a turnout of 37.9% the new clause found its way into the Constitution.

Croatia  335 According to the Constitution, the organisation of the Government is based on the principle of separation of powers. The three branches named in the Constitution are the legislative, the executive, and the judicial. Interestingly, the same provision (Article 4) that introduces the separation of powers principle also explains that this principle does not exclude cooperation between the branches of government. In particular, it ‘encompasses forms of mutual cooperation’ as well as ‘reciprocal checks and balances’. Croatia is defined by the Constitution as a unitary state. Nevertheless, the Constitution guarantees to citizens the right to local and regional self-government. It even enumerates policy areas that may be regulated and administered at lower levels of government, such as the organisation of localities and housing, zoning and urban planning, public utilities, child care, social welfare, primary health services, education and primary schools, culture, physical education and sports, consumer protection, protection and improvement of the environment, fire protection, and civil defence.13 With respect to local decision-making, the Constitution provides the same rights to EU citizens and to nationals, which must be exercised ‘in compliance with law and the EU acquis communautaire’.14 Croatia has been a parliamentary democracy since the constitutional amendments of 2000, when the semi-presidential system was abandoned after the first president, Franjo Tuđman, passed away. The Croatian Parliament, called Hrvatski sabor, is a unicameral body15 composed, according to the Constitution, of between 100 and 160 members elected by direct elections every four years. The Law on Election of Representatives to the Croatian Parliament set this number at 151.16 The Parliament, as a rule, adopts statutes by simple majority of the MPs present, provided that at least half of the MPs are present. In some cases, however, the Constitution demands higher majorities.17 The Parliament may delegate to the Government its legislative powers, except those for which some type of qualified majority is envisaged by the Constitution. This enables the Government to decide by way of legislative decrees. The Croatian President is directly elected for five years and can be re-elected once. The Constitution does not give significant powers to the President. The President calls the parliamentary elections, gives the mandate for the formation of the Government, promulgates laws, may initiate ex-post constitutionality review, represents the country abroad, and is the commanderin-chief of the Croatian army. The independence of the judiciary is guaranteed by the Constitution. The Constitution only regulates the role of the Supreme Court, whereas the organisation of the rest of judiciary is regulated by statute. The Supreme Court has the role of ensuring uniform application and equality of everyone before the law. The President of the Supreme Court is nominated by the Croatian Parliament on the proposal of the President for a period of four years. Other judges are appointed, promoted, and dismissed by an independent body, called the National Judicial Council. It is composed of 11 people, of whom seven are judges, two are university law professors, and two are MPs, one of which is a member of the opposition. The tenure of the judges lasts until retirement age. 13 Art 129a (134) of the Croatian Constitution (hereinafter HRC). 14 Art 128 (132) HRC. 15 This is so since constitutional amendments of 2001. Until then, the Parliament was bi-cameral, with the other house representing regional communities. 16 140 MPs are elected in 10 electoral units, eight are elected by the national minorities and three MPs are elected by Croatian citizens living abroad. 17 The laws regulating the rights of national minorities, or laws altering the borders are to be adopted by a two-thirds majority of all MPs whereas the laws elaborating human rights and those altering the electoral system require the majority of all MPs.

336  Tamara Ćapeta Courts are divided into the general and specialised courts. General courts at the first instance are Municipal Courts and on appeal County Courts (for some cases County Courts act as first instance). All cases that are not attributed to the jurisdiction of specialised courts are decided by general courts. There are three branches of specialised courts – commercial courts, administrative courts, and misdemeanour courts. All of them are organised on two levels, with a single High Commercial, High Administrative, and High Misdemeanour Court, respectively. At the top of the hierarchy of these courts is the Supreme Court of Croatia. The adjudication culture has undergone significant changes during the last decade. The old, strictly formalistic view of law based on few sources and on textual interpretation18 is gradually being replaced by a culture that recognises the importance of interpretation in law and recognises sources other than statutes. This was reflected in the 2010 constitutional amendments which introduced a new sentence describing the judicial function. It reads as follows: ‘Courts shall administer justice according to the Constitution, law, international treaties and other valid sources of law.’19 Hitherto, the Constitution mentioned only the Constitution and statutes as relevant sources to be used by the judges. According to the Constitution, international treaties were always part of the Croatian legal order, and if duly ratified, stood higher than statutes in the hierarchy of norms. However, in practice, they were rarely used in adjudication, with the exception of the European Convention on Fundamental Rights and Freedoms (ECHR), which has been increasingly applied by judges in recent years. Initially, treaties were not listed as a source of law with the exception of being mentioned in the chapter of the Constitution dealing with international relations. Today, the consolidated Constitution mentions, as explained, treaties also among the sources to be used by judges. The expansion of the sources of law to be applied by judges was, among other reasons, intended to facilitate the application of EU law. Another novelty in the judiciary introduced by the 2010 constitutional amendments was vesting judges with their judicial duties personally. This increased the personal responsibility of judges for their decisions and might in time influence the judicial culture. The Constitution also defines the Office of the State Attorney as an independent judicial body. The State Attorney is appointed by the Parliament on the basis of the Government’s proposal, whereas his/her deputies are appointed by the National State Attorney Council, an independent body similar to the National Judicial Council. The Constitution also features a Constitutional Court (CC). However, the CC is not part of the judiciary, and is often referred to as the ‘fourth branch of government’. Nevertheless, the CC understands itself as ‘court or tribunal’ in the sense of Article 267 TFEU,20 even though it has not yet made its first reference. The organisation and powers of the CC are regulated by the Constitution and a Constitutional Act on the Constitutional Court.21 In terms of constitutionality review, the CC has jurisdiction to exercise abstract and concrete constitutionality review of statutes and by-laws. The review can, in principle, be initiated only after a law enters

18 An excellent description of the legal cultures of new democracies was given by Zdenek Kühn, ‘Worlds Apart. Western and Central European Judicial Culture at the Onset of the European Enlargement’, (2004) 52 American Journal of Comparative Law 531; see also Zdenek Kühn, ‘European Law in the Empires of Mechanical Jurisprudence: The Judicial Application of European Law in Central European Candidate Countries’, (2005) 1 Croatian Yearbook of European Law and Policy 55; Tamara Ćapeta, ‘Courts, Legal Culture and the Enlargement of the EU’, (2005) 1 Croatian Yearbook of European Law and Policy 23; Alan Uzelac, ‘Survival of the Third Legal Tradition?’, (2010) 49 Supreme Court Law Review 377. 19 Art 115 (117) HRC. 20 CC 24.06.2019, Decision U-III-970/2019 (Ljubljanska banka), para 14. 21 Consolidated version published in NN 49/02.

Croatia  337 into force. Abstract review is initiated by a request submitted by privileged applicants.22 These include courts, if the question of constitutionality arises in proceedings pending before them. Non-privileged applicants, such as any natural person or a company, may propose to the CC the opening of a constitutionality review procedure, which the CC may refuse. The CC, however, as a matter of its judicial policy, is deciding in all such cases. This policy makes the abstract constitutional review a kind of actio popularis in Croatia. The concrete control is initiated by constitutional complaint against an individualised decision, usually the judicial decision, by which the applicant alleges the violation of his/her constitutionally protected human rights. The Constitution does not provide for the anterior or posterior review of international agreements and the CC has not sought to develop such jurisdiction. Based on the statistical data available at the CC’s website, 11.4 per cent of all cases referred to the CC in 2017 were constitutionality review proceedings of laws or by-laws. The overwhelming majority of the cases (87.2 per cent) were constitutional complaints on the violation of human rights and fundamental freedoms guaranteed by the Constitution. The Constitution can be amended on a proposal of one-fifth of the MPs, the President, and the Government. Draft amendments are adopted by a simple majority in Parliament and amendments are adopted by a two-thirds majority. The Constitution can also be amended by a referendum.

II.  Constitutional Culture The heritage of the socialist system, when the Constitution was not perceived as a normative act but rather as a broad political programme,23 has left its mark on the current constitutional culture in Croatia. Although the Constitution, and with it also the CC, are gaining importance in the public political sphere, the understanding that the Constitution is not applicable law is still reflected in everyday adjudication by courts. The Constitution simply did not find its way into the case law (except that of the CC), either as a source of rights or as an interpretative tool. Only rarely have judges initiated constitutional review procedures in front of the CC in order to challenge the validity of a statute.24 The CC’s case law is sometimes contradictory in that regard. In some cases the CC has strongly reproached the courts for not referring,25 whereas in others, it has acted discouragingly.26 Despite the fact that courts rarely apply the Constitution, its importance becomes more visible in the political process. Political actors often invoke the Constitution and use the CC to check the constitutionality of their decisions or challenge the decisions of other actors. As a result of these developments the CC is gaining importance; this is also reflected in the latest Constitutional

22 According to Art 35 of the Constitutional Act on the Constitutional Court, the request for constitutional review may be presented by ‘one fifth of the members of the Croatian Parliament, a committee of the Croatian Parliament, the President of the Republic of Croatia, the Government of the Republic of Croatia to review the constitutionality and legality of regulations, the Supreme Court of the Republic of Croatia (hereinafter SCHR) or another court of justice, if the issue of constitutionality and legality has arisen in proceedings conducted before that particular court of justice, the People’s Ombudsman in proceedings provided by Article 92 of the Constitution of the Republic of Croatia’. 23 Cf Branko Smerdel, ‘Ustav RH nakon ustavnih promjena 2010. godine’ [‘The Constitution of the Republic of Croatia after constitutional amendments of 2010’], (2010) Hrvatska pravna revija 1, 6. 24 According to the search engine of the CC, there were four such requests in 2013, three in 2014, two in 2015, none in 2016 and only one in 2017. 25 CC 08.07.2013, Decision U-III-736/2005. 26 CC 27.08.2013, Decision U-I-4175/2013. But, see also CC 27.09.2016, Decision U-II-2944/2012 which may be read as the change in the CC’s position.

338  Tamara Ćapeta amendments, which increased the majority necessary for the nomination of constitutional justices. The required majority in the Parliament was increased to two-thirds majority.27 Given that the parliamentary majority will rarely win two-thirds of the seats, the new arrangement requires the agreement to be reached between the governing party and the opposition on the appointment of a constitutional judge. The new procedure was tested in 2016, when the Parliament appointed 10 (out of 13) members of the CC (some new, some reappointed). Faced with the possible blockage of constitutional adjudication, as the mandate of a number of judges was to expire, the parliamentary parties reached an agreement and divided the vacant seats at the CC – one half for the majority, the other for the opposition. Even if many reproached the politicisation of the selection process, the political agreement that was reached also demonstrates the importance that political actors attach to the CC. The possibility of the CC becoming dysfunctional was seen as unacceptable and damaging for democracy, making a political agreement indispensable. The CC is still searching its own understanding of the constitutional order. Sometimes still caught in formalistic legal interpretive logic, its methods of constitutional adjudication are also strongly influenced by other systems, most notably the ECHR system, to which Croatia acceded in 1997. By October 2013, the CC has relied on the case law of the ECtHR in more than a thousand decisions.28 The CC often refers to the decisions of the German Federal Constitutional Court, and sometimes performs comparative analyses of different constitutional solutions.29 The CC is thus a court open to external influences, under which it has taken the move towards more contextual interpretation, embracing certain previously unfamiliar concepts, such as, for instance, the principle of proportionality.30 EU membership will certainly influence further the method and understanding of the law by the CC. The CC has already developed its case law by understanding the Constitution as a living document. Taking into consideration the important place of the German Federal Constitutional Court in the CC decisions, it is to be expected that its jurisprudence in relation to constitutional issues resulting from the EU membership will be heavily influenced by German doctrine.

III.  Constitutional Foundations of EU-Membership A.  The New EU Chapter In order to prepare the country for accession, the 2010 constitutional amendments added a new Chapter (VIII), entitled ‘European Union’. The drafting of the EU chapter was done by a group of experts and then proposed to the political decision-making. The new text envisaged that this Chapter would enter into force at the moment of accession to the EU. Thus, it became binding only as of 1 July 2013. It contains four Articles. The first states the constitutional basis for EU membership and is relevant for future amendments of the EU Treaties. The second describes the roles of Croatian institutions at the EU level and provides a constitutional basis for the enactment of a law governing the relations between the institutions of government in EU affairs. The third 27 Art 122/1 (126/1) HRC. 28 Answers to the Questionnaire for the XVIth Congress of the Conference of European Constitutional Courts. Vienna, 12–14 May 2014 (Zagreb, 2 October 2013) 18. 29 Ibid 25. 30 Snježana Bagić, Načelo razmjernosti u praksi europskih sudova i hrvatskog Ustavnog suda [The principle of proportionality in the case law of European courts and the Croatian Constitutional Court] (Zagreb, Pravni fakultet Sveučilišta u Zagrebu, 2016).

Croatia  339 article concerns the relationship between EU and domestic law. The fourth article restates the EU-based citizenship rights. Before we explain these provisions in more detail, it is necessary to give some background to the 2010 constitutional amendments, when the EU Chapter was introduced: The 2010 constitutional amendments were adopted during Croatia’s EU membership negotiations. The negotiations were officially opened on 3 October 2005 and closed on 30 June 2011, when all negotiating Chapters were closed. However, membership expectations had started building in the Croatian population ever since the opening of negotiations for the Stabilisation and Association Agreement (SAA) in the year 2000.31 During the 13 years passing until accession, the enthusiasm and eagerness of Croatian citizens for EU membership steadily decreased. This decrease was due to the long and troublesome negotiations and serious political and economic crises that developed in the EU. The EU, thus, ceased being seen as the promised land of plenty. Nevertheless, the pro-European political consensus of all parliamentary and most non-parliamentary political parties existed during the entire process of negotiations. There was also a consensus that EU membership necessitated citizens’ support to be expressed through a referendum. The Constitution, at that time, provided for rather strict referendum rules. The general rule required at least 50 per cent voter turnout for a referendum to be valid and at least 50 per cent of the votes cast should be in favour for the referendum to be successful.32 An even stricter majority was required by the provision which was decided to serve as the basis for EU membership (‘association and dissociation’ clause).33 It required 50 per cent of all voters (not only of those who participated) to vote in favour for the referendum to be successful. The high support threshold and the voters’ fading enthusiasm were the first motive for the 2010 amendments. The political elites could not afford the possible failure of a referendum after almost a decade and a half of negotiations. Faced with a high probability of not collecting enough supporting votes, it was necessary to loosen the referendum requirements, which required constitutional amendments. The second motive for constitutional amendments was EU conditionality.34 Namely, during negotiations it became clear that certain constitutional amendments would be necessary in order to satisfy membership requirements. Constitutional amendments, thus, became also a conditio sine qua non for the conclusion of accession negotiations. Most of these changes proved not to be controversial.35 31 The SAA was signed in the end of 2001 but entered into force only in 2005. Official membership requirement was put by the Croatian government on 21 February 2003. 32 Art 86 HRC, pre-2010 version. 33 Art 141 HRC, pre-2010 version. There were scholars, at the time, who argued that the named provision is not necessarily a proper constitutional basis for membership. According to Siniša Rodin, historically the main purpose of that provision was dissociation from former Yugoslavia. Even if the entire provision is drafted as if it regulates association, and the dissociation is mentioned only in the last paragraph, this was so because of a political moment of the time. The high referendum threshold envisaged was easily surmounted in the referendum on dissociation from Yugoslavia. Rodin proposed an alternative provision that allowed the transfer of powers to an international organisation or an association by an international agreement, provided such an agreement is ratified by two-thirds parliamentary majority. The referendum was not mandated by that provision, but it was possible. Cf Siniša Rodin, ‘Regulatorna autonomija država članica i ustavna osnova za pristupanje Republike Hrvatske Europskoj uniji’ [‘Regulatory autonomy of the Member States and the constitutional basis for the accession of the Republic of Croatia to the European Union’], in HAZU (ed), Pristupanje Republike Hrvatske Europskoj uniji [Accession of the Republic of Croatia to the European Union] (Zagreb, HAZU, 2007) 23; Siniša Rodin, ‘Ustavna osnova za članstvo RH u EU’ [‘The constitutional basis for the membership of the Republic of Croatia in the EU’], Informator 24.09.2008. 34 On conditionality see Dimitry Kochenov, EU enlargement and the failure of conditionality (Alphen aan den Rijn, Kluwer Law International, 2008). 35 The Constitution had to be amended, for instance, to accommodate the European Arrest Warrant, as it prohibited, prior to amendments, the extradition of Croatian citizens. Membership also required strengthening of the independence of the Central Bank and the State Audit Office; enabling EU citizens living in Croatia to participate in elections for European Parliament and in local elections.

340  Tamara Ćapeta Thus, initiating the constitutional amendment process became indispensable if Croatia was to join the EU. The legal basis for the EU referendum was amended to render a referendum successful if a majority of the votes cast was in favour. The general provision on referenda was also amended by removing the 50 per cent voter turnout requirement for validity. These constitutional amendments were agreed exclusively to respond to particular needs of the time. The aim was achieved – on 22 January 2012, 43.51 per cent of Croatian voters participated in a referendum on EU membership, and 66.27 per cent of those voting approved Croatia’s future EU membership. However, such adjustments to the Constitution to suit the political needs of the moment deprive the Constitution of much of its value. Additionally, they have also opened the path for future, not necessarily desirable, developments. On the one hand, the amendments made referenda much easier. These indeed lead, as explained in the introduction, to the increase in the number of referenda, a development which could, on its own, be welcomed. However, they also opened the possibility for a referendum, even one resulting in constitutional amendments, in which a minority of the voting population can impose its values on the rest of society. Thus, in 2013, the Constitution was for the first time amended through a referendum. 65.87 per cent of the 37.9 per cent of voters who voted in a referendum introduced a change to the Constitution reserving marriage to a union between man and woman only. This was a development contrary to a growing trend in European values36 and which was not welcomed by many in Croatia.37 It is surprising that the removal of the strict referendum requirement for association with other states was not resisted. The ‘association and dissociation’ provision was an important part of the Constitution. It reflected fears of Croatian citizens of entering into unions with other states, resulting from negative experiences with the participation in the Yugoslav federation. The same provision contains a paragraph that prohibits the participation of Croatia in Yugoslavia-like unions.38 For other associations, Croatians clearly chose in the original version of the Constitution a higher threshold of a majority of all (voting) citizens. This choice was too easily given up in the 2010 amendments. Accordingly, the EU Chapter was not the primary reason for the 2010 constitutional amendments. It was rather an important by-product of a process which was initiated for different reasons.39 This might explain the lack of any public discourse, be it general or professional, about this aspect of the 2010 constitutional amendments. It seems that the constitutional conditions under which Croatia will participate in the EU and will implement its policies were not as important. What mattered was that the move to membership is accomplished successfully. The membership modalities were of secondary importance.

36 In May 2015, 62% of Irish voters (with a turnout of 60.5%) upheld the possibility of same-sex marriages. 37 Tina Oršolić Dalessio commented: ‘Ironically, what indirectly facilitated the passage of this referendum was the legal process and strategy Croatia took in becoming a member of the EU’. Tina Oršolić Dalessio, ‘The Interplay of Direct and Indirect Democracy at Work – Croatia’s Battle Over the Rights of Same-Sex Couples’, JURIST – Forum, 23.01.2014, http:// jurist.org/forum/2014/01/tina-dalessio-croatia-referendum.php. 38 Second paragraph of Art 135 (142) HRC on ‘association and dissociation’ reads: ‘Any procedure for the association of the Republic of Croatia into alliances with other states, if such association leads, or may lead, to a renewal of a South Slavic state union or to any form of consolidated Balkan state is hereby prohibited.’ 39 I have argued elsewhere that even if an EU Chapter in the Constitution is welcome, it was not indispensable. Croatia could successfully participate in the EU, embracing its most important constitutional aspects, such as direct effect or supremacy, even without constitutional amendments. Tamara Ćapeta, ‘Nacionalni Ustav i nadređenost prava EU u eri pravnog pluralizma’ [‘National Constitution and supremacy of EU law in the era of legal pluralism’], (2009) 59 (1) Zbornik Pravnog fakulteta u Zagrebu, 63; Tamara Ćapeta, ‘Ustavni acquis communautaire i ustavni poredak RH’ [‘Constitutional acquis communautaire and the constitutional order of the Republic of Croatia’], in H Kačer et al (eds), Liber amicorum Jadranko Crnić (Zagreb, Novi Informator, 2009) 221.

Croatia  341 There were also other, non-EU-related constitutional amendments, which ‘happened’ because the process of amending the Constitution was open.40 Some of those amendments attracted much more public interest than the EU-related ones.41 As there were no public discussions about the new EU Chapter, this aspect of the constitutional amendments proceeded more or less unnoticed. This may be the explanation for the fact that today, seven years after these amendments, the EU Chapter is rarely invoked in practice. It is not referred to in judicial decisions, and it is also missing from the public political discourse. It was only very recently that the CC started to refer to the provisions of the EU Chapter of the Constitution.42

B.  The Constitutional Basis for EU Membership and for Future Treaty Amendments Be it as it was, the ‘association and dissociation’ provision, with loosened referendum criteria, became the legal basis for the move into the EU membership. This was restated in first paragraph of Article 141a (143), the opening provision of the new EU Chapter: Pursuant to Article 142 [‘association and dissociation’ provision] of the Constitution, the Republic of Croatia shall, as a Member State of the European Union, participate in the creation of European unity in order to ensure, together with other European states, lasting peace, liberty, security and prosperity, and to attain other common objectives in keeping with the founding principles and values of the European Union.

This provision not only restated the legal basis for joining the EU, but also added some substantive limits to EU membership. Thus, Croatia only participates in an EU which is committed to peace, liberty, security and prosperity. It is too early to say whether this will have any influence and, if so, on which, of the future EU Treaty amendments. Future Treaty amendments are not dependent on the ‘association and dissociation’ provision. Even if this does not flow clearly from the text of the Constitution, preparatory documents43 indicate that such amendments will not follow the ‘association and dissociation’ procedure, but rather, the procedure for signing international treaties would suffice. This procedure is referred to in paragraph 2 of Article 141a (143): ‘Pursuant to Articles 14044 and 14145 of the Constitution, the Republic of Croatia shall confer upon the institutions of the European Union

40 For the enumeration of all constitutional amendments in 2010 see Branko Smerdel, Unexpected degradation: Croatian constitutionalism since the full EU membership, working paper, www.pravo.unizg.hr/USTP/en/course/conlaw. 41 For example, the removal of the statute of limitation for prosecuting war-related crimes; or votes accorded to ethnic minorities in parliamentary elections. 42 CC 10.07.2018, Decision U-III-208/2018. Cf also CC 02.05.2018, Decision U-I-1694/2017 et al; CC 16.10.2018, U-I-2735/2008. 43 Prijedlog za utvrđivanje nacrta promjene Ustava Republike Hrvatske (Draft Proposal for the Constitutional amendments 2010), https://sabor.hr/sites/default/files/uploads/sabor/2019-01-18/080049/NACRT_PROMJENE_USTAVA.pdf. The same explanation is repeated in the position paper submitted by the opposition members of Parliament at the time of the 2010 Constitutional amendments. 44 This provision regulates the process of ratification of international treaties. Those treaties which ‘require the adoption of amendment to laws, international treaties of military and political nature, and international Treaties which give rise to financial commitments for the Republic of Croatia’ are adopted by Parliament by simple majority. However, agreements which transfer constitutional powers to another organisation are ratified by two-thirds majority of all members of the Parliament. 45 This provision provides for ratified international treaties to become part of the Croatian legal system and have primacy over domestic law.

342  Tamara Ćapeta the powers necessary for the enjoyment of rights and fulfilment of obligations ensuing from membership’. On the one hand, this provision clarified that once the requirements imposed by the ‘association and dissociation’ provision are satisfied (primarily after the positive outcome of a referendum), the Accession Treaty is to be concluded in accordance with the procedure for the ratification of international treaties. As the Constitution demands a two-thirds majority for the transfer of powers by an international treaty, the Accession Treaty was ratified by such a majority in the Parliament.46 The drafters of the constitutional amendments thus opted to use the same rules as the ones envisaged for the ratification of international treaties for the ratification of any future Treaty amendments. The necessary majority for adopting future Treaty amendments will depend on the type of the amendments. If no new powers are transferred to the EU, a simple majority in Parliament might suffice. However, if Treaty amendments grant new powers to the EU, a higher majority would be required – two-thirds of all MPs. But, unlike under the ‘association and dissociation’ provision, no referendum is required. This is so even if there is new transfer of powers. Future Treaty amendments do not require a referendum. However, the Constitution does not exclude such a possibility. The Parliament or the President, on the proposal of the Government and with the approval of the Prime Minister, may call for a referendum. Likewise, one-tenth of all Croatian voters can also obligate the Parliament to call a referendum. Thus, the procedural limit to further integration in the EU is the qualified majority in the Parliament, with a possibility, but not the obligation, to call for a referendum. A question which may arise is whether the same is true in the case of the amendments of the Treaties in order to accept new Member States. The answer will, in the end, depend on the CC’s position regarding the question whether the enlargement of the ‘association of states’ in which Croatia participates requires new popular approval.

C.  ESM and Fiscal Compact Croatia has not signed the ESM Treaty since it is not (yet) a eurozone member. Nevertheless, in March 2018, Croatia joined the Fiscal Compact.47 However, in accordance with Article 14(5) of the Compact, Croatia decided to opt out of Titles III and IV. This enabled the ratification of this agreement in Parliament by simple majority. It was not discussed whether ratification of Titles III or IV would require enhanced majority in Parliament. In any case, ratification of any international agreement, which might, as those adopted during the euro crisis, serve the purpose of furthering EU integration by means others than Treaty amendments, will have to be signed and ratified in accordance with the procedure for the adoption of international treaties. Whether this will involve the Parliament, and whether the necessary majority will be simple or qualified, will depend on the assessment whether by such an agreement Croatia transfers powers flowing from the Constitution.

46 On 9 March 2012, the Croatian Parliament unanimously (136 votes in favour, none against) ratified the Accession Treaty. 47 The Law on the Ratification of the TSCG is published in NN 1/2018.

Croatia  343

IV.  Constitutional Limits to EU Integration A.  Substantive Limits to Future Treaty Amendments 1.  Possible Constitutional Grounds for Imposing Substantive Limits Apart from the requirement that Croatia participates in an EU that is committed to peace, liberty, security and prosperity, no substantive limits to further EU integration are set explicitly in the Constitution. However, it is to be expected that the CC will, in time, develop some boundaries through interpreting the Constitution. However, as Croatia joined the EU only recently, no such case law exists. So far, a developed doctrine considering parts of the Constitution as being its unchangeable ‘core’ does not exist. However, recently, and again in relation to referenda, the CC started developing a concept of ‘constitutional identity’. The necessity for this arose due to the increase in the number of referenda after the 2010 constitutional amendments, which relaxed the referendum requirements, coupled with inadequate referendum legislation. Thus, the task of determining whether there are questions which cannot be decided by a referendum (and which they are) fell to the CC. The first mention of the concept of ‘constitutional identity’ was in the notification of the CC relating to the referendum on marriage (already mentioned above).48 Lacking parliamentary initiative for the review of the constitutionality of that referendum, the CC has decided to act on its own initiative, by a public notification. To justify such a move, it explained that it can interfere on its own motion if it considers that there existed a serious threat to ‘the structural characteristics of Croatian constitutional state, or in other words of its constitutional identity, including the highest values of the constitutional order of the Republic of Croatia (Article 1 and Article 3 of the Constitution)’.49 The CC did not find such a serious threat to the Croatian constitutional identity in the proposal of the referendum on marriage. However, it has clearly explained that the limits to the possibility of amendments of the Constitution by a referendum exist and are based in a concept of ‘constitutional identity’. The same sentence was subsequently repeated in several decisions.50 The latest relevant case in the development of the concept of ‘constitutional identity’ was a 2017 decision, relating to the constitutionality of a decision of local authorities by which a street was named after the date of the establishment of the fascist Croatian state during World War II.51 Justifying both its jurisdiction in the case and its position that such a decision was unconstitutional, the CC considered that ‘[t]he Constitutional principles determine the structure and the essence of the Croatian state. The Republic of Croatia can remain what it is only if none of its structural principles is annulled or amended.’52

48 Notification SuS – 1/2013 of the CC of 14 November 2013 about the popular constitutional referendum on the ­definition of marriage. 49 Ibid, para 5. Translation by the author. 50 See, for instance, CC 13.01.2014, Decision U-VIIR-164/14; or CC 08.04.2015, Decision U-VIIR-1159/2015 on the application by the Croatian Parliament to the CC to determine whether the question proposed to be put at the national referendum on outsourcing is in conformity with the Constitution, para 33.4. 51 CC 10.10.2017, Decision U-II-6111/2013. 52 Ibid, para 19.1.

344  Tamara Ćapeta Even if this case dealt with the constitutionality of a decision and not with the possibility of constitutional amendments, one may conclude that those aspects of the Constitution without which Croatia cannot exist cannot be constitutionally amended. Thus, these core principles represent the unamendable part of the Constitution. The concept of ‘constitutional identity’ has not yet been applied within the EU context. However, taking into consideration the willingness of the CC to follow and apply the practices of other courts of constitutional jurisdiction in Europe, I believe that this is just a matter of time. Constitutional scholars also suggest this course of action.53 Thus, it is highly likely that ‘constitutional identity’ will become a limit to further integration. The content of that concept will have to be elaborated in case law. The first hint that the CC might reject unconditional EU membership was given in two decisions, both relating to the constitutionality of referenda.54 In the last paragraphs of both decisions, the CC announced that the Constitution is supreme to EU law, without elaborating on it any further. These cases will be analysed below. It suffices here to say that with such a statement, which was unnecessary for deciding the case at hand, the CC made a first step towards imposing constitutional limits on EU membership. The other possible provision on the basis of which the CC could impose constitutional limits on further EU integration is the sovereignty clause. The Constitution determines that Croatia is a sovereign state, describing sovereignty as inalienable, indivisible, and non-transferable (Article 2). However, whatever these characterisations mean, they do not prevent the transfer of regulatory powers, nor impose any clear limits on such transfer. The very same article allows for entering into alliances with other states, and other constitutional provisions provide for a procedure for transfer of powers. The only limit imposed is the existence of a possibility of withdrawal, which, as already explained in the introduction, is definitively settled by the Lisbon Treaty provision on withdrawal from the EU. Having in mind the current development in the relationship between constitutional courts of EU Member States and the CJEU, it is more likely that the CC will use the ‘constitutional identity’ doctrine and not the sovereignty clause to impose limits on integration. This is even more probable, given the influence of German constitutional doctrine. Finally, there is also a possibility of using the Kompetenz-Kompetenz argument, given that the Constitution contains an explicit provision on the transfer of competences, on the basis of which Croatia is an EU member. This may serve as a justification for examining whether an exercised competence could have been transferred in the first place under the Constitution.

2.  The Content of the ‘Constitutional Identity’ Concept As already mentioned, a ‘constitutional identity’ doctrine is currently being developed in the CC’s practice. As for its content, for now the CC mentioned two Articles of the Constitution as expressing Croatia’s ‘constitutional identity’ – Articles 1 and 3. 53 Branko Smerdel, ‘In Quest of a Doctrine: Croatian Constitutional Identity in the European Union’, (2014) 64 (4) Zbornik pravnog fakulteta u Zagrebu, 513; Ana Horvat Vuković, ‘U ime Ustava – materijalne granice promjene Ustava’ [‘In the name of the Constitution – substantive limits to constitutional amendments’] (2015) 65 Zbornik Pravnog fakulteta u Zagrebu 481. 54 CC 08.04.2015, Decision U-VIIR-1159/2015 on the application by the Croatian Parliament to the Constitutional Court to determine whether the question proposed to be put at the national referendum on outsourcing is in conformity with the Constitution, para 60; and CC 21.04.2015, Decision U-VIIR-1158/2015 on the application by the Croatian Parliament to the CC to determine whether the question proposed to be put at the national referendum on giving ­highways in concession is in conformity with the Constitution, para 45.

Croatia  345 Article 1 reads as follows: The Republic of Croatia is a unitary and indivisible democratic welfare state. Power in the Republic of Croatia derives from the people and rests with the people as a community of free and equal citizens. The people exercise this power through the election of representatives and through direct decision-making.

Article 3 lists basic constitutional values: Freedom, equal rights, national and gender equality, peace-making, social justice, respect for human rights, inviolability of ownership, conservation of nature and the environment, the rule of law and a democratic multiparty system are the highest values of the constitutional order of the Republic of Croatia.

The CC repeated in 201755 that structural constitutional principles without which the Croatian state, as envisaged by the Constitution, cannot exist embrace the rule of law, freedom, equality, national equality, peace-making and respect for human rights. The two constitutional provisions invoked by the CC, as well as the enumerated constitutional principles, list a number of values, the meaning of which is highly undetermined and is, therefore, prone to different interpretations. It is sufficient to ask, for instance, what it means that Croatia is a ‘democratic welfare state’. If the idea of constitutional identity will ever be used to impose limits to European integration, the meaning of those principles in the context of a common endeavour, such as the EU, will have to be further elaborated. Unlike some other constitutional courts in Europe, Croatia’s CC’s influence in imposing limits to EU integration through concepts such as ‘constitutional identity’ will be more difficult and indirect. One of the main reasons is that the Constitution does not envisage the jurisdiction of CC to perform either anterior or posterior constitutional review of treaties.56

B.  Supremacy over the Constitution and Judicial Review of EU Law 1.  The Declaration of Supremacy of the Constitution and its Effects for the Scrutiny of Secondary EU Law In two decisions relating to referenda, decided on the same day, the CC declared supremacy of the Constitution over EU law. Neither of the two decisions concerned EU law, let alone its conformity with the Constitution. As the declaration on the supremacy of the Constitution was not necessary for deciding on the cases, it might be seen as an intentional effort by the CC to open room for its participation in the constitutional discourse in the EU. That way the CC created conditions for the constitutional review of secondary EU law, if necessary. The first decision57 concerned the constitutionality of a referendum proposal of a law to prevent the concessioning of Croatian motorways. The second decision58 concerned the 55 CC 27.10.2017, Decision U-II-6111/2013 para 51. 56 Jasna Omejec, ‘Legal Framework and Case-Law of the Constitutional Court Of Croatia in Deciding on the Conformity of Laws With International Treaties’ Report, European Commission for Democracy Through Law (Venice Commission), The Constitutional Court of Montenegro and OSCE (2009) CDL-JU(2009)035, http://www.venice.coe.int/webforms/ documents/default.aspx?pdffile=CDL-JU(2009)035-e. 57 CC 21.04.2015, Decision U-VIIR-1158/2015 para 54. 58 CC 08.04.2015, Decision U-VIIR-1159/2015 para 54.

346  Tamara Ćapeta constitutionality of a referendum proposal to amend a law aiming to prevent the Government from outsourcing to the private sector certain ancillary services (like cleaning) within different state bodies.59 The CC found both referenda contrary to certain values embodied in the Croatian Constitution, and therefore found them unconstitutional. In its written pleadings in both cases, the Parliament submitted long and elaborated arguments that the laws that would be a result of the referenda at issue are not only contrary to the Constitution but also to fundamental freedoms of the EU internal market. Responding to these arguments, after declaring the inconsistency with the Constitution, the CC stated in this constitutional procedure it is not necessary to examine substantive conformity of referendum question with the law of the European Union, because the Constitution is, in its legal force, superior to the law of the European Union.

This declaration of the supremacy of the Constitution over EU law was, first of all, not necessary for deciding the cases at hand. The CC could have simply concluded that it is not necessary to deal with other arguments as the proposed law would in any case be contrary to the Constitution. But it did not. The conclusion to be drawn from it is that the statement at issue, albeit obiter dictum, was placed there intentionally. What could the intention be? The most plausible explanation is that the CC decided to engage in the pluralistic game already played by other courts having constitutional jurisdiction in Europe with the CJEU. In a way, this is the only path one could expect from the court tasked with the protection of the Constitution of a country. The statement on the supremacy of the Constitution, however, is not (yet) qualified. Obviously, imposing unilateral and intolerant supremacy would not be in accordance with the way EU pluralism is lived. Thus, it is to be expected that in the future the CC will develop a more refined concept, following the positions of other constitutional courts in Europe, and limiting the supremacy to the crucial aspects of the Constitution. The declaration of the supremacy of the Constitution, opens, of course, the possibility to scrutinise EU law in light of the Constitution. Given the readiness of the CC to take into consideration positions of other European constitutional courts, it is to be expected that it will, in the good tradition of constitutional tolerance already existing in Europe, also limit its control over EU law only to exceptional cases. These may be linked to the concept of constitutional identity, which the CC has started developing.

2.  Scrutiny of EU Law against EU Law Itself The Constitution does not currently provide, nor does it seem likely that the CC will construe it so, the CC with jurisdiction over direct challenges of EU secondary law for incompatibility with the Treaties or the EU Charter of Fundamental Rights (the Charter). However, such questions

59 Apart from being relevant for EU law, the cases were also constitutionally important because they opened the opportunity for the CC to review the constitutionality of a draft law, for which it normally does not have jurisdiction under the Constitution. Namely, the referenda proposed the text of the law about which the referendum question would be asked. Thus, judging the constitutionality of a referendum effectively judged the constitutionality of the proposed future law.

Croatia  347 can easily come up incidentally before the CC. For instance, in a case where the CC assesses the constitutionality of legislation transposing EU law, it may be claimed that the EU norm is invalid for its incompatibility with, for instance, the Charter.60 In order to assess the validity of the transposing legislation, the CC would have to respond also to the allegation of invalidity of EU law. The CC sees control over the hierarchy of legal norms as part of its constitutional role.61 The Constitution even textually empowers the CC not only for constitutionality review, but also for legality review, ie review of the compatibility of by-laws with laws. As EU law became part of the domestic legal order and as the CC declared supremacy of the Constitution over EU law, the CC might interpret the question of compatibility between a rule of secondary EU law and the Treaty or the Charter as an example of a sub-constitutional conflict over which it has jurisdiction. The CC may, on the other hand, qualify a possible conflict between secondary EU law and the Treaties as an issue of constitutionality, via the concept of the rule of law.62 In the third scenario, the CC could characterise the scrutiny of secondary EU law with the Treaties or the Charter as a quasi-constitutional review. This conclusion seems possible as the CC has already characterised certain treaties, especially the ECHR, as having quasi-constitutional status.63 If the Treaties and the Charter are so characterised, the scrutiny of secondary EU law may be understood as a question of constitutionality review. In all scenarios, the question arises whether the CC will be willing to consult with the CJEU on the question of the validity of EU law, and whether it would apply the Foto-Frost doctrine.64 In a recent case,65 the CC confirmed for the first time that it considers itself a ‘court or tribunal’ in the sense of Article 267 TFEU. However, the CC has not (yet) referred to the CJEU either the questions of validity or of interpretation of EU law. Following the Foto-Frost doctrine, the CC should decide that it does not have jurisdiction to invalidate EU legal rules for incompatibility with the Treaties/Charter. The Foto-Frost rule, of course, allows the national court to decide that an EU rule is valid without the reference. However, unless the validity fits the acte clair criteria of CILFIT,66 the CC would have to refer even if it considers that the EU rule is valid, being the court against whose decisions there is no judicial remedy. Thus, the CC might decide always to refer to the CJEU any allegations of invalidity of EU law for conflicts with the Treaties or the Charter. This still would not close the possibility to invalidate an EU act on limited constitutional grounds as discussed above should the CJEU declare its validity within the framework of EU law.

60 This was the case in the Data Retention Directive cases, in which the constitutionality of domestic measures transposing the Directive was challenged in domestic Courts, but opened the question of validity of the Directive itself. The latter issue was resolved by the CJEU, which found the Directive invalid in CJEU 08.04.2014 C-293/12 and C-594/12 (Digital Rights Ireland Ltd v Minister for Communications, Marine and Natural Resources and Others and Kärntner Landesregierung and Others) ECLI:EU:C:2014:238. 61 CC 08.11.2000, Decision U-I-745/1999. 62 The CC used similar arguments in order to establish its jurisdiction to assess validity of domestic legislation in the light of international treaties. Cf text accompanying nn 80 and 81. 63 CC 15.07.1998, Decisions U-I-920/1995 and U-I-950/1996; and CC 08.11.2000, Decision U-I-745/1999. See also, the report by Jasna Omejec, former president of the CC cited in n 56. 64 ECJ 22.10.1987 Case 314/85 (Foto-Frost v Hauptzollamt Lübeck-Ost) ECLI:EU:C:1987:452. 65 CC U-III-970/2019 (n 20). 66 ECJ 06.10.1982 Case 283/81 (CILFIT v Ministero della Sanità) ECLI:EU:C:1982:335.

348  Tamara Ćapeta

C.  Effect of EU Law in the Domestic Legal Order 1.  The Text of the Constitution Article 141c (145) of the Constitution, entitled ‘European Union law’ was inserted as part of the EU Chapter in order to regulate the status of EU law in the Croatian constitutional order. It reads as follows: The exercise of the rights ensuing from the European Union acquis communautaire shall be made equal to the exercise of rights under Croatian law. All the legal acts and decisions accepted by the Republic of Croatia in European Union institutions shall be applied in the Republic of Croatia in accordance with the European Union acquis communautaire. Croatian courts shall protect subjective rights based on the European Union acquis communautaire. Governmental agencies, bodies of local and regional self-government and legal persons vested with public authority shall apply European Union law directly.

The purpose of the first paragraph is not entirely clear. The preparatory instruments, especially the explanatory part of the draft proposal for constitutional amendments 2010, interpret the first and the second paragraphs as expressing the EU principle of effectiveness ‘according to which national law cannot make the exercise of EU rights excessively difficult or virtually impossible’.67 If anything, having in mind its wording, the first paragraph possibly reflects the principle of equivalence (and not the principle of effectiveness), according to which the protection afforded to EU rights has to be at least of equal value to the protection afforded to equivalent domestic rights. Why precisely that principle would be embodied in the text of the Constitution is not self-explanatory. Another possible reading of the first paragraph is that it expresses the concept of direct effect, understood as the capacity of EU law to bestow rights directly. Such restatement of direct effect is further worked out in paragraph 3 in relation to courts and in paragraph 4 in relation to other public bodies. It is, however, not clear why direct effect is expressed in terms of rights in relation to courts, and in terms of the obligation of direct applications in relation to public bodies. The three paragraphs offer enough elements to read the concept of direct effect as being accepted by the Constitution. The second paragraph can also be read as the confirmation of such an interpretation. It provides that EU law is applied in Croatia in accordance with EU law, which itself imposes direct effect. The problem of the second paragraph, which would have to be removed by interpretation, is that it refers only to legal acts ‘accepted’ by Croatia in EU institutions. This could serve as the basis for the argument that legal acts against which Croatia voted in the Council are not binding. This would, of course, run contrary to the qualified majority voting envisaged by the Treaty and could not have been intentionally so framed. The second problem is that this paragraph mentions only ‘legal acts and decisions’ and is thus not referring to primary law or international agreements concluded by the EU. The latter was implicitly resolved by a July 2018 decision in which the CC considered that Article 141c (145) introduces into the Croatian legal order agreements concluded by the EU, as they are part of the EU legal order.68 This was also the first case in which the CC decided to apply and interpret one of the EU Chapter provisions of the Constitution. The procedure was initiated



67 Draft 68 CC

Proposal (n 43) 19 f. U-III-208/2018 (n 42) para 21.

Croatia  349 by a constitutional complaint in a case relating to the extradition of a Turkish citizen to Turkey at the request of a Turkish criminal court. The courts, including the Supreme Court, allowed the extradition. They considered that the fact that the Turkish citizen at issue had been granted refugee status by the Swiss Confederation was irrelevant, as Switzerland is not a member of the EU. The CC, however, explained that Switzerland participates in the Schengen and Dublin acquis on the basis of international agreements which it has signed with the EU. These agreements are part of the EU acquis and are, therefore, on the basis of Article 141c (145) of the Constitution, relevant and binding on Croatian courts. The courts, therefore, by neglecting these international agreements, violated Article 141c (145) of the Constitution. The CC consequently annulled the Supreme Court’s decision and returned the case back to the judiciary. The constitutional complaint alleged the violation of several provisions of the Constitution, including the right to fair trial (which is granted by Article 29 of Croatian Constitution). The complainant, however, never invoked Article 141c (145). Thus, it was the CC’s decision to apply that provision. The case left unresolved the issue whether any alleged violation of Article 141c (145) by a court represents at the same time also the violation of the right to fair trial. Such a wide interpretation might bring a number of cases to the CC, turning that Court into an additional judicial instance. This case has factually arisen from a situation in which courts refused to apply EU law, not the situation in which they misapplied it. Thus, the CC still has multiple possibilities of explaining the relation between constitutional complaint and the violation of Article 141c (145). The second paragraph of Article 141c (145) refers back to the effects of EU law as developed within the EU legal system. It can, thus, be used not only for establishing direct effect but also the supremacy of EU law. There is no explicit mentioning of EU law supremacy anywhere in the Constitution. However, for ordinary supremacy, ie the supremacy of EU law over legislation (not the Constitution), the quoted constitutional provision on EU law as well as other constitutional provisions may serve as a legal basis. Thus, as I have argued in more length elsewhere,69 the provision which envisages that international treaties ratified in accordance with the constitutional procedures become part of Croatian law and have primacy over domestic laws,70 can serve as a constitutional basis justifying both direct effect and supremacy domestically. Given the circularity of paragraph 2 of Article 141c (145), as it refers for the effects of EU law back to the EU law itself, one may ask whether the Constitution or EU law is the legal basis for the validity and effects of EU law in Croatia. The case law, described below, seems to suggest the latter.

2.  Direct Effect, Supremacy and other EU Concepts in Ordinary Courts While I am not aware of any important judgment that was decided by the direct application of EU law, it seems that ordinary courts do not have a problem with the direct effect or supremacy of EU law. In a long-awaited decision on the credits pegged to the Swiss franc,71 the Croatian Supreme Court gave the following statement, even though EU law was not applicable in the case (because the facts arose before the country’s EU membership): The Republic of Croatia has become a member of the European Union on the 1st July 2013, since when the law of the European Union makes part of its legal order and has to be applied; more than that, it has



69 Ćapeta, 70 Art

‘Nacionalni Ustav’ (fn 39) and Ćapeta, Constitutional acquis communautaire (n 39). 134 (142) HRC. 09.04.2015 Judgement and Order No Revt 249/14-2 (Franak).

71 SCHR

350  Tamara Ćapeta primacy over national law. Such obligation to apply the law of the European Union exists in respect of all legal relations that are in the scope of application of the law of the European Union, and which arose after the Republic of Croatia became the member of the European Union.72

Thus, the Supreme Court announced its willingness not only to directly apply EU law, but also to accord it primacy. Interestingly, however, the Supreme Court never mentioned the Constitution as a justification for such powers. Thus, it seems as if the direct effect and supremacy of EU law flow from EU law itself. This is unsurprising, given that Croatian courts, as already explained, rarely apply or refer to the Constitution. On the other hand, this can also be interpreted as being entirely in line with the CC’s position on the legal basis for the validity of EU law in Croatia. In the two decisions in which the CC declared the primacy of the Constitution over EU law it explained that there were no reasons to examine the conformity of the law which would result from the referendum in relation to ‘the law which is valid in Croatia on the basis of the Founding Treaties in the light of Article 141c of the Constitution’.73 Thus, the CC has also pronounced that the legal basis for the validity and the effects of EU law lies in the Treaties with the permission of the Constitution. Therefore, it is to be expected that the case law developed by the CJEU in relation to direct effect and ordinary supremacy will be applied in practice. Some hurdles might lie in the way of practical implementation of supremacy. Most notably, Croatia has rules very similar to Italy at the time of the Simmenthal II judgment.74 Thus, the constitutional law of the CC obliges courts to initiate the constitutional review procedure before the CC in cases of possible incompatibility of the Statute with the Constitution.75 Whether this provision will represent a hurdle will depend on the interpretation, ultimately given by the CC, whether the question of conformity of domestic law with EU law is a question of constitutionality or not. If it is not, there is no reason why ordinary courts could not set aside the domestic law which violates EU law in the case at hand. Thus, the CC has the option of interpreting the Constitution in conformity with EU law requirements, which, given the obligation of conformity interpretation resulting from the loyalty obligation, it should choose. However, we cannot rule out the possibility of the CC reading the Constitution differently. The courts have also seen themselves bound by the concept of interpretive effects. The Supreme Court, in a case related to the European arrest warrant, used the interpretive obligation in order to establish the right of a person to lodge an appeal.76 Namely, not being a party to a dispute, a wife of a victim of a crime for which surrender was demanded, had no right to appeal under Croatian law. The Supreme Court, however, considered: … such textual interpretation would bring into question the realisation of the purpose of European law, as expressed in the mentioned Framework decision of the Council on the European arrest warrant and the surrender procedures between Member States, but also of other rules by which European law is regulated (directives and framework decisions), which refer to the protection of and the rights of victim in criminal and other proceedings. Therefore, in order to achieve the objectives and respect the principles expressed in EU law, national courts are under an obligation to apply national law in the light of the text and purpose of EU law. This means that national law in its application has to be interpreted, as much as possible, in the light of the text and the purpose of relevant framework decisions and directives,

72 Ibid. Unfortunately, paragraphs in decisions of the SCHR are not numerated. The translation into English is the author’s. 73 See n 54. 74 ECJ 09.03.1977 Case 106/77 (Amministrazione delle Finanze dello Stato v Simmenthal SpA) ECLI:EU:C:1978:49. 75 Art 37 of the Constitutional law on Constitutional Court of the Republic of Croatia. 76 SCHR 06.03.2014 Order No Kž-eun 5/14-4 and Kž-eun 14/14-4 (Mustač).

Croatia  351 in order to achieve the result which such framework decisions and directives want to achieve … By becoming a member of the EU, the Republic of Croatia has also undertaken an obligation to act in that way.77

The Supreme Court did not explain wherefrom the legal basis for conform interpretation flows. However, and given that it has quoted the Pupino judgment,78 arguably it also flows directly from the EU legal order. The Supreme Court also considered that the obligation of conform interpretation exists in relation to cases which arose prior to the Croatia’s EU membership. The legal basis for such an obligation is, according to the Supreme Court, the Stabilization and Association Agreement that regulated the relationship of Croatia with the EU prior to accession.79

3.  Scrutiny of Domestic Law against EU Law itself (Especially the Charter) by the Constitutional Court As described, it seems that the Constitution and the practice of courts allow for the daily application of the combination of direct effect and supremacy as demanded from courts by the Simmenthal mandate.80 The question which is still unanswered is whether the CC has jurisdiction to assess the validity of domestic law in light of EU law and whether such power would not deprive regular courts of their Simmenthal power. With the two referendum-related decisions declaring the supremacy of the Constitution over EU law, the rank of the EU law is to a certain degree equalised to that of international treaties. Therefore, the question of whether the CC can assess the compatibility of domestic law with EU law can be answered by making a parallel to its powers to check the compatibility of domestic law with international treaties. Namely, according to the Constitution, ratified treaties form part of the Croatian legal order and prevail over domestic law, but not over the Constitution. The text of the Constitution does not vest the CC with jurisdiction to review the compatibility of domestic legislation with international treaties. However, in a series of judgments,81 the CC has established such jurisdiction. It has used the constitutional ‘rule of law’ principle as justification for the newly established power.82 Thus, the CC sees itself as a guardian of the rule of law understood as respect for the hierarchical order of legal rules. As the Constitution introduces EU law into the domestic legal order and gives it supremacy over domestic law, the CC could, using the same justification as in relation to treaties, establish its jurisdiction to review the compatibility of domestic law with EU law. This can already be read from the two referendum-related decisions. Although the CC did not review the compatibility of laws, which would follow from the proposed referenda, with EU law, it did not declare itself incompetent to exercise such review. Quite to the contrary, the reason the CC gave for not undertaking such review was that the referenda were already found unconstitutional and a compatibility review with EU law was not necessary. By extension, should the CC have found no unconstitutionality it would have proceeded to review compatibility with EU law.

77 Ibid, translation by the author. 78 ECJ 16.06.2005 C-105/03 (Criminal proceedings against Maria Pupino) ECLI:EU:C:2005:386. 79 SCHR Franak (n 71). 80 The expression ‘Simmenthal mandate’ is borrowed from Monica Claes, The National Courts’ Mandate in the European Constitution (Oxford, Hart Publishing, 2006). 81 CC U-I-920/1995 and U-I-950/1996 (n 63). See also CC 03.02.2000, Decision U-I-884/1997; CC 05.07.2000, Decision U-I-1026/1995; CC 20.09.2000, Decision U-I-275/2000; CC 30.04.2008, Decisions U-I-1201, 2866 and 3021/2006. 82 Ćapeta, ‘Constitutional acquis communautaire’ (n 39) 239–44.

352  Tamara Ćapeta Pending the confirmation of this conclusion, a step further could be taken to ask whether such jurisdiction of the CC would be in conflict with the ‘Simmenthal’ obligation of regular courts to perform such review and set aside incompatible domestic legislation. Can both regular courts and the CC, in parallel, have jurisdiction to review the compatibility of domestic law with EU law? One reason to answer in the affirmative is the fact that the CC already enjoys certain jurisdiction running in parallel with that of regular courts. Thus, the CC has, according to the Constitution, not only the power of reviewing the constitutionality of laws, but also of reviewing legality. Regular courts are also empowered to review legality and set aside by-laws without prior referral to CC. It is, thus, possible to conclude that the power of the CC can run in parallel with the ‘Simmenthal’ obligation of regular courts. This is even more so, as the legal consequences of the legality review of regular courts and the CC differ. In particular, regular courts only have the power to set aside the incompatible national rule in the case at hand. The CC, however, can invalidate the incompatible law with erga omnes effect. If the parallelism of power of scrutiny of domestic law for its compatibility with EU law is accepted, Croatia will not face the problem in the Austrian legal order after incorporating the EU Charter in the constitutional order with the Data Retention Directive Case.83 Once the check of domestic legislation with the Charter has become a matter of constitutionality in Austria, regular courts were deprived of the jurisdiction provided by EU law to review and disapply domestic provisions violating the Charter. In Croatia, notwithstanding the ‘incorporation’ of the Charter, the outcome need not be depriving regular courts of that power. However, it must be repeated that the issue is yet to be decided by the CC. The obligation of conform interpretation, which also binds the CC, would strongly suggest the interpretation of the Constitution which would be compatible with the case law of the CJEU in cases such as Melki and Abdeli84 or the recent A v B and Others.85 However, the CC’s case law, according to which some treaties, such as the ECHR, have quasi-constitutional status, might lead its case law in a different direction and cause a problem similar to the one existing in Austria. Finally, on a more practical level, given that anyone can initiate abstract constitutional review before the CC, if the latter is understood as encompassing the compatibility of Croatian law with EU law, the number of abstract review cases initiated before the CC might seriously increase. This may, therefore, become a serious reason for the CC to find another solution.

D.  The ‘Citizenship’ Provision in the EU Chapter of the Constitution The Constitution contains Article 141 d (146) entitled ‘Rights of European Union Citizens’. This provision is probably a novelty in comparison with other EU member states’ constitutions, and therefore deserves to be mentioned. It reads as follows: Citizens of the Republic of Croatia shall be European Union citizens and shall enjoy the rights guaranteed by the European Union acquis communautaire, and in particular: • freedom of movement and residence in the territory of all Member States, • active and passive voting rights in European parliamentary elections and in local elections in another Member State, in accordance with that Member State’s law,

83 For

explanation see CJEU 11.09.2014 C-112/13 (A v B and Others) ECLI:EU:C:2014:2195 para 24. 22.06.2010 Joined Cases C-188/10 (Aziz Melki) and C-189/10 (Sélim Abdeli) ECLI: EU:C:2010:363. C-112/13 (A v B and Others) (fn 83).

84 CJEU 85 CJEU

Croatia  353 • the right to the diplomatic and consular protection of any Member State which is equal to the protection provided to own citizens when present in a third country where the Republic of Croatia has no diplomatic-consular representation, • the right to submit petitions to the European Parliament, complaints to the European Ombudsman and the right to apply to European Union institutions and advisory bodies in the Croatian language, as well as in all the other official languages of the European Union, and to receive a reply in the same language. All rights shall be exercised in compliance with the conditions and limitations laid down in the founding treaties of the European Union and the measures undertaken pursuant to such treaties. In the Republic of Croatia, all rights guaranteed by the European Union acquis communautaire shall be enjoyed by all citizens of the European Union.

This provision enumerates the EU citizenship rights as provided by the Treaties, and grants them to Croatian citizens in other EU states. The curious side of this provision is that the Constitution purports to grant rights to Croatians in other EU Member States. The most probable reason for inserting this provision was a ‘marketing’ one. By repeating these rights, the provision might have had the purpose of persuading Croatians to vote for EU membership.86 The last paragraph of the provision could also have some practical purpose. It guarantees all Treaty-based rights to the nationals of other EU Member States in Croatia. This is an additional legal argument for claiming EU rights in Croatian courts, and it goes beyond the citizenship rights, as it refers to all rights guaranteed by the acquis.

V.  Constitutional Rules and/or Practice on Implementing EU Law The Constitution also contains, in the EU Chapter, a provision on the role of the governmental institutions in the EU decision-making process. Article 141 b (144) entitled ‘Participation in European Union institutions’ reads as follows: The citizens of the Republic of Croatia shall be directly represented in the European Parliament where they shall, through their elected representatives, decide upon matters falling within their purview. The Croatian Parliament shall participate in the European legislative process as regulated in the founding treaties of the European Union. The Government of the Republic of Croatia shall report to the Croatian Parliament on the draft regulations and decisions in the adoption of which it participates in the institutions of the European Union. In respect of such draft regulations and decisions, the Croatian Parliament may adopt conclusions which shall provide the basis on for the Government’s actions in European Union institutions. Parliamentary oversight by the Croatian Parliament of the actions of the Government of the Republic of Croatia in European Union institutions shall be regulated by law. The Republic of Croatia shall be represented in the Council and the European Council by the Government and the President of the Republic of Croatia in accordance with their respective constitutional powers.

86 See more in Tina Oršolić Dalessio, ‘Constitutional provision on EU citizenship – the case of Croatia’, (2011) 2 Dutch Constitutional Law Review.

354  Tamara Ćapeta This article served as the basis for adopting the Law on Cooperation between the Croatian Parliament and the Government of the Republic of Croatia in European Affairs.87 The Parliament discusses the general legislative programme that is published annually by the European Commission, discusses the positions which the Government plans to adopt in the Council (or the European Council), and may even, in the form of Conclusions, bind the Government to a particular voting position at the EU level. An important role in that regard is given to the parliamentary European Affairs Committee (EAC).88 This committee is composed of MPs but also includes outside experts. Formally, the committee’s opinions are merely advisory to the Parliament. The EAC is also in charge of conducting subsidiarity checks. The process may be initiated by any MP, parliamentary committee, parliamentary party group, or the Government. In 2014, the EAC issued a reasoned opinion in the context of a national parliamentary subsidiarity check procedure. In this opinion, issued on 6 October 2014, the EAC considered that the Proposal for amending the Directive on Waste89 did not comply with the principle of subsidiarity.90 Both the EAC and the Environmental Protection Committee considered that the Proposal did not take into consideration the existing differences among national systems of waste management, which prejudiced a balanced development of European regions. However, not enough reasoned opinions were issued to initiate the ‘yellow card’ procedure (only the Austrian, Croatian, and Czech parliaments issued such opinions). Nevertheless, the Proposal was eventually withdrawn.91 In May 2016, the Parliament joined 10 other EU member states’ parliaments in initiating a ‘yellow card’ procedure on the proposal for amending the Posted Workers Directive.92 Even if the Commission rejected the subsidiarity complaints, it nevertheless expressed willingness to continue the dialogue with domestic parliaments on the proposal.93 Thus, the Parliament showed willingness and ability to use its constitutional mandate to participate in decision-making at the EU level.

VI.  Conclusion – The Relationship between the Croatian Constitution and the EU Legal Order Being the youngest EU member, Croatia could not claim not to have been aware that it had joined a constitutionalised, federation-like, legal order. Thus, its Constitution, even textually, confirms that constitutional concepts as developed in the EU legal order are accepted in Croatia. 87 Zakon o suradnji Hrvatskog Sabora i Vlade Republike Hrvatske u europskim poslovima [Law on Cooperation between the Croatian Sabor and the Government of the Republic of Croatia in European Affairs] NN 81/13. 88 See the Report on the work of the EAC for 2014 at www.sabor.hr/fgs.axd?id=42829. 89 Proposal for a Directive amending Directives 2008/98/EC on waste, 94/62/EC on packaging and packaging waste, 1999/31/EC on the landfill of waste, 2000/53/EC on end-of-life vehicles, 2006/66/EC on batteries and accumulators and waste batteries and accumulators, and 2012/19/EU on waste electrical and electronic equipment, COM(2014) 397 final. 90 Reasoned opinion of the Croatian Parliament on COM (2014)397, www.ipex.eu/IPEXL-WEB/dossier/files/downloa d/082dbcc548cd77e10148e4f3bf0f181a.do. 91 Commission response at www.ipex.eu/IPEXL-WEB/dossier/files/download/082dbcc54d4a5c3c014d4cc47c6b0237.do. 92 Proposal for a Directive of the European Parliament and of the Council amending Directive 96/71/EC of The European Parliament and of the Council of 16 December 1996 concerning the posting of workers in the framework of the provision of services, COM(2016) 128 final, 8 March 2016. 93 Communication from the Commission to the European Parliament, the Council and the National Parliaments on the proposal for a Directive amending the Posting of Workers Directive, with regard to the principle of subsidiarity, in accordance with Protocol No 2, COM(2016) 505 final, 20 July 2016.

Croatia  355 Additionally, even if not expressly stated in the text, the EU constitutional concepts, such as direct effect and supremacy of EU law over ordinary legislation, could easily be accommodated within the constitutional text. On the other hand, the constitutional relations that are not settled within the EU legal order are not settled by the Constitution either. Primarily, these are the constitutional limits to further EU integration and the constitutional review of EU law resulting from the ultimate supremacy of the domestic Constitution and not EU constitutional law. As Croatia has been an EU member for just over seven years at the time of final revision of this chapter, these issues have not yet come up in a judicial context. However, by recently declaring, in an obiter dictum, the supremacy of the Constitution over EU law, the CC set the stage for its own participation in the discursive EU constitution-making. Being a court open to doctrinal influences of other European legal systems, it can be expected that the CC will learn from other constitutional courts’ experiences and develop a similar position.94 Contemporary EU constitutional developments seem to base the limits to integration on the concept of constitutional identity. This was further reinforced by the entry into force of the Lisbon Treaty, which obliges the EU to respect the ‘national identities, inherent in their fundamental structures, political and constitutional’95 of each Member State. The meaning of the concepts of national identity and constitutional identity is far from clear. It is, therefore, not only subject to scholarly analyses ever since the Lisbon Treaty,96 but is has entered the constitutional dialogue between domestic constitutional courts and the CJEU.97 It is a matter of time for this meaning to receive sharper contours in the context of that dialogue. The actual effect of this new provision for now is strengthening the position, or at least, the attitude, of domestic constitutional courts. This strengthening is evident in recent statements of these constitutional courts warning the CJEU that they will enforce constitutional supremacy if the CJEU finds an interpretation of EU law that contradicts their Constitutions.98 There are strong signs that the same concepts will be used in the similar way by the CC. Admittedly, the Constitution does not explicitly introduce the concept of constitutional identity and CC has not yet applied it within the EU context. However, the CC has already started to develop that concept in its case-law. Instances of constitutional conflicts in EU law occur rarely. Thus, from a more pragmatic perspective, for the daily application of EU law it is more important that regular courts do not seem to have any trouble with accepting the direct effect and ordinary supremacy of EU law.

94 By which I do not want to say that different Member States’ constitutional positions are the same. In that respect see, Monica Claes and Jan-Herman Reestman, ‘The Protection of National Constitutional Identity and the Limits of European Integration at the Occasion of the Gauweiler Case’, (2015) 16 German Law Journal 917. 95 Art 4.2. TEU. 96 For example Siniša Rodin, ‘National Identity and Market Freedoms after the Treaty of Lisbon’, (2011) 7 Croatian Yearbook of European Law and Policy 11. 97 The recent example is CJEU 05.12.2017 C-42/17 (Criminal proceedings against M.A.S. and M.B.) ECLI:EU:C:2017:936, initiated by the Italian CC, after the CJEU’s judgment in C-105/14 (Taricco and Others) ECLI:EU:C:2015:555. In Taricco, the CJEU considered that the Italian constitutional rule which prohibits retroactive change in the statute of limitation for prosecution of criminal offences is contrary to EU law and has, therefore, demanded the court to dis-apply it. In M.A.S. and M.B. the Italian CC challenges such interpretation, among other grounds, as running contrary to Italian constitutional identity, thus opening the possibility to trigger its constitutional supremacy doctrine, the controlimiti. 98 The well-known examples are CJEU 26.02.2013 C-399/11 (Stefano Melloni v Ministerio Fiscal) ECLI:EU:C:2013:107, as well as the preliminary reference of the German Federal Constitutional Court which resulted in the judgment in CJEU 16.06.2015 C-62/14 (Peter Gauweiler and Others) ECLI:EU:C:2015:400. The other recent example is the already mentioned reference of the Italian CC CJEU C-42/17 (M.A.S and M.B.) (n 95). As explained in CJEU 18.07.2017 C-42/17 Opinion AG Bot ‘[i]n its order for reference, the Corte costituzionale not only submits those questions for a preliminary ruling to the Court, but also advises the Court on the answer that should be given in order to avoid initiating what is known as the “counter-limits” procedure’ (ECLI:EU:C:2017:564 para 10).

356  Tamara Ćapeta

References S Bagić, Načelo razmjernosti u praksi europskih sudova i hrvatskog Ustavnog suda [The principle of proportionality in the case-law of European courts and the Croatian Constitutional Court] (Zagreb, Pravni fakultet Sveučilišta u Zagrebu, 2016) (). T Ćapeta, ‘Courts, Legal Culture and the Enlargement of the EU’, (2005) 1 Croatian Yearbook of European Law and Policy 23. T Ćapeta, ‘National Constitution and supremacy of EU law in the era of legal pluralism’, (2009) 59 (1) Zbornik Pravnog fakulteta u Zagrebu, 63 (Nacionalni Ustav i nadređenost prava EU u eri pravnog pluralizma). T Ćapeta, ‘Constitutional acquis communautaire and the constitutional order of the Republic of Croatia’, in H Kačer et al (eds), Liber amicorum Jadranko Crnić (Zagreb, Novi Informator, 2009) 221 (Ustavni acquis communautaire i ustavni poredak RH). M Claes, The National Courts’ Mandate in the European Constitution (Oxford, Hart Publishing, 2006). M Claes and J-H Reestman, ‘The Protection of National Constitutional Identity and the Limits of European Integration at the Occasion of the Gauweiler Case’, (2015) 16 German Law Journal 917. A Horvat Vuković, ‘U ime Ustava – materijalne granice promjene Ustava’ [‘In the name of the Constitution – substantive limits to constitutional amendments’], (2015) 65 Zbornik Pravnog fakulteta u Zagrebu 481. D Kochenov, EU enlargement and the failure of conditionality (Alphen aan den Rijn, Kluwer Law International, 2008). Z Kühn, ‘Worlds Apart. Western and Central European Judicial Culture at the Onset of the European Enlargement’, (2004) 52 American Journal of Comparative Law 531. Z Kühn, ‘European Law in the Empires of Mechanical Jurisprudence: The Judicial Application of European Law in Central European Candidate Countries’, (2005) 1 Croatian Yearbook of European Law and Policy 55. J Omejec, ‘Legal Framework and Case-Law of the Constitutional Court Of Croatia in Deciding on the Conformity of Laws With International Treaties’, Report, European Commission for Democracy Through Law (Venice Commission), The Constitutional Court of Montenegro and OSCE (2009). T Oršolić Dalessio, ‘Constitutional provision on EU citizenship – the case of Croatia’, (2011) 2 Dutch Constitutional Law Review. T Oršolić Dalessio, ‘The Interplay of Direct and Indirect Democracy at Work – Croatia’s Battle Over the Rights of Same-Sex Couples’, JURIST – Forum, 23.01.2014. S Rodin, ‘Regulatorna autonomija država članica i ustavna osnova za pristupanje Republike Hrvatske Europskoj uniji’ [‘Regulatory autonomy of the Member States and the constitutional basis for the accession of the Republic of Croatia to the European Union’], in HAZU (ed), Pristupanje Republike Hrvatske Europskoj uniji’ (Zagreb, HAZU, 2007) 23. S Rodin, ‘Ustavna osnova za članstvo RH u EU’ [‘The constitutional basis for the membership of the Republic of Croatia in the EU’], Informator 24.09.2008. S Rodin, ‘National Identity and Market Freedoms after the Treaty of Lisbon’, (2011) 7 Croatian Yearbook of European Law and Policy 11. B Smerdel, ‘Ustav RH nakon ustavnih promjena 2010. godine’ [‘The Constitution of the Republic of Croatia after constitutional amendments of 2010’], (2010) Hrvatska pravna revija 1. B Smerdel, Unexpected degradation: Croatian constitutionalism since the full EU membership, working paper, www.pravo.unizg.hr/USTP/en/course/conlaw. B Smerdel, ‘In Quest of a Doctrine: Croatian Constitutional Identity in the European Union’, (2014) 64 (4) Zbornik pravnog fakulteta u Zagrebu, 513. A Uzelac, ‘Survival of the Third Legal Tradition?’, (2010) 49 Supreme Court Law Review 377.

12 Italy MARTA CARTABIA AND BARBARA GUASTAFERRO*

I.  Main Characteristics of the National Constitutional System A.  The Italian Constitution and its Significance in the Legal System The Italian Constitution came into force on 1 January 1948, two years after the institutional referendum that turned the constitutional monarchy into a democratic republic. This constitutional transformation was the result of a political compromise among the three principal forces that acted together as the engine of change: the liberal, the Christian democratic and the socialist-communist traditions. After World War II, these leading forces formed the National Liberation Committee (CLN) in the name of their shared anti-fascist beliefs1 and were eventually all represented in the Constituent Assembly, an elected body entitled to draft a new Constitution. While differing in their cultural and political backgrounds, they found their ‘glue’ in the memory of the past and promoted a ‘never again’ Constitution, intended to avoid the major mistakes of the totalitarian regime. In contrast to the rhetoric of war adopted by the Fascist Government, the constitutional provisions governing the foreign relations of the Italian Republic, namely Article 11, emphasise the principle of pacifism. In this respect, the Italian Constitution belongs to the group of Constitutions that Costantino Mortati has called ‘constitutions born from the Resistance’.2 The anti-totalitarian origins of the Italian Constitution lie at the heart of many of its basic features which, to a certain extent, are shared by other Constitutions stemming from similar historical backgrounds, such as Germany’s.3 The first of these basic features is the rigidity of the Constitution. The Italian Constitution is a written document, which is at the apex of the hierarchical order of legal sources. It cannot be amended by the ordinary legislative procedure, which * Within the framework of a joint reflection Marta Cartabia authored sections III, IV and VI and Barbara Guastaferro sections I, II, and V. 1 Massimo Luciani, ‘Antifascismo e nascita della costituzione’, (1991) 22 Politica del diritto 183, 191. 2 Costatino Mortati, Lezioni sulle forme di governo (Padua, CEDAM, 1973) 222. By ‘constitutions born from the Resistance’, Mortati also referred to other documents, for instance, the French (Fourth Republic) and German ­Constitutions. See Giuseppe Martinico, Barbara Guastaferro, and Oreste Pollicino ‘The Constitution of Italy: Axiological Continuity between the Domestic and International Levels of Governance?’, in A Albi and S Bardutzky (eds), The Role of National Constitutions in the European and Global Governance (The Hague, TMC Asser Press, 2019). For a sound a­ nalysis of the main features of the Italian legal order, see Giuliano Amato, Le istituzioni della democrazia. Un viaggio lungo cinquant’anni (Bologna, Il Mulino, 2015). 3 For a comparative survey on both the similar and dissimilar features of the system of constitutional adjudication in Italy, Germany and France see John Ferejohn and Pasquale Pasquino, ‘Constitutional Adjudication, Italian Style’, in T Ginsburg (ed), Comparative Constitutional Design (Cambridge, Cambridge University Press, 2012) 294.

358  Marta Cartabia and Barbara Guastaferro would allow a simple majority to be unconstrained and free to change the basic document of the polity.4 In order to avoid the possible ‘tyranny of the majority’, specific procedures entailing an overarching consensus among several representations within the Parliament are required to amend the Constitution.5 Moreover, the Italian Constitution sets explicit limits to the amendment procedure. According to Article 139, the ‘Republican form’ is impossible to amend. Besides this explicit limit – stated expressly by the Constitution – the Italian Constitutional Court carves out other implicit limits to constitutional revision. They were defined either by giving an extensive interpretation to the formula ‘Republican form’ – which ended up including the fundamental rights and values necessary to define the legal order as ‘democratic’, such as personal freedom, equal vote, etc – or by emphasising that Article 2 of the Constitution, in defining the rights of the person as ‘inviolable’, would insulate the rights and freedoms enshrined in the first part of the Constitution from constitutional revision.6 These supreme principles thus represent the ‘centre of gravity’ and the unamendable core of our Constitution.7 The second basic feature of the Italian legal order is the constitutionalisation of the so-called ‘Kelsenian’ model of constitutional adjudication,8 which requires a specialised Constitutional Court entitled with the constitutional review of legislation. Far from fearing the gouvernement des juges – as it happened in the French constitutional tradition, which refused for a long time to scrutinise the laws enacted by the sovereign Parliament – and far from embracing an orthodox and absolute view of parliamentary sovereignty espoused by the UK constitutional tradition, reluctant to abandon the idea that ‘the Parliament can do no wrong’ – the Italian constitutional tradition contemplates the possibility that the legislator can act in an unconstitutional way. Needless to say, also behind this choice might have been the experience of totalitarian regimes: the creation of a judicial body, independent from the executive branch9 was intended to prevent possible majoritarian political choices from infringing upon fundamental rights.10 4 This was for example the case with the ‘Statuto Albertino’, namely the Charter granted by the sovereign before the republican era, which did not contemplate the idea that the ordinary law enacted by the Parliament could be subordinated to a higher norm. 5 Those procedures are envisaged by Art 138 of the Constitution (hereinafter Cost): ‘Laws amending the Constitution and other constitutional laws shall be adopted by each House after two successive debates at intervals of not less than three months, and shall be approved by an absolute majority of the members of each House in the second voting. Said laws are submitted to a popular referendum when, within three months of their publication, such request is made by one-fifth of the members of a House or five hundred thousand voters or five Regional Councils … A referendum shall not be held if the law has been approved in the second voting by each of the Houses by a majority of two-thirds of the members’. 6 See, among others, Constitutional Court (hereinafter CCIT) 15.12.1988, Sentenza 1146/1988, ECLI:IT:COST:1988:1146; CCIT 22.01.1982, Sentenza 18/1982, ECLI:IT:COST:1982:18; and CCIT 13.04.1989, Sentenza 232/1989, ECLI:IT:COST:1989:232 (on the principle of judicial protection); CCIT 11.07.1991, Sentenza 366/1991, ECLI:IT:COST:1991:366 on the right to free and confidential communication etc. Moreover, the Court insulated from amendment also the ‘supreme principles’ of the Constitution that initially it intended to protect from the ­interference of ‘external orders’ (see for example CCIT 24.02.1971, Sentenze no 30 and 31/1971, ECLI:IT:COST:1971:30 and ECLI:IT:COST:1971:31 respectively; and CCIT 13.01.1982, Sentenza 16, ECLI:IT:COST:1982:16; and CCIT 22.01.1982, Sentenza 18/1982, ECLI:IT:COST:1982:18). For an analysis on how these supreme principles were eventually opposed also to European integration see Marta Cartabia, Principi inviolabili e integrazione europea (Milan, Giuffrè, 1995). 7 See the similarity with the ‘eternity clause’ of the German Constitution as per Art 79.3. 8 For a comparative analysis endeavouring to identify Europe’s common and divers constitutional traditions of constitutional review, see Maartje de Visser, Constitutional Review in Europe. A Comparative Analysis (Oxford, Hart, 2015). Specifically on the Italian Constitutional Court, V Barsotti, P Carozza, M Cartabia and A Simoncini (eds), Italian Constitutional Justice in Global Context (Oxford, OUP, 2015). 9 The independence of the Constitutional Court from the Governmental branch is guaranteed by its composition. According to Art 135 Cost, the Court is composed of 15 judges, a third nominated by the President of the Republic, a third by Parliament in joint sitting and a third by the ordinary and administrative Supreme Courts. The judges of the Constitutional Court shall be chosen from judges of the ordinary and administrative higher Courts, from full professors of law and lawyers with at least 20 years’ practice. Judges of the Constitutional Court shall be nominated for nine years, and they may not be reappointed, which is another condition guaranteeing their independence from political mechanisms. 10 It is worth recalling that this kind of historical rationale lying behind constitutional review has been recently contested. Among others, a strong criticism on how constitutional reform throughout the word has transferred an

Italy  359

B.  Vertical and Horizontal Separation of Powers Our Constitution qualifies the Italian state as republican and democratic in that, as per Article 1, Italy is a republic where sovereignty belongs to the people. The Constitution envisages both a horizontal and vertical separation of powers.11 As to the first, the most characteristic feature of the Italian legal order is the parliamentary form of government, where the Parliament can grant and revoke confidence in the Government through a reasoned and voted motion. The Parliament has a pivotal role in the constitutional system since, being directly elected by the citizens, it represents popular will. It exercises the legislative function and is composed of the House of Deputies and of the Senate of the Republic, which are almost equal in their functions and powers. Accordingly, the Italian legal order can be described as a system of ‘perfect bicameralism’. The Government of the Republic – made up of the President of the Council and of the Ministers – holds the executive function. Taking into account the results of the elections, the President of the Republic appoints the President of the Council, who conducts and holds responsibility for the general policy of the Government, and ensures the coherence of political and administrative policies. The Italian Constitution is somewhat laconic on the role and functions of the Government, which have significantly varied over the time. To give an example, the dynamics of the parliamentary form of government, and – consequently – the stability of the Government and its position vis-à-vis the Parliament, were inevitably affected by the change of the electoral system, from a proportional to an almost majoritarian one, in 1994. The political landscape later changed again when the electoral system was modified in order to comply with the Constitutional Court decisions no 1 of 2014 and no 35 of 2017.12 At present the electoral system is based on the proportional model, with some minor corrections. The Constitution defines in general terms the judicial branch. The Court system is dual in that it is based on the separation of ‘ordinary courts’ and ‘administrative courts’. Ordinary courts administer civil and criminal justice through judicial bodies (both bodies of first instance and appeal court) and prosecutors.13 Administrative courts have the power to annul public administration’s acts infringing upon individuals’ legitimate interests in a fair and lawful administrative action.14 As to the rights of the individuals in front of the judiciary, besides qualifying the right of defence as ‘inviolable at every stage and level of the proceedings’, the Constitution also states that ‘[n]o one may be withheld from the jurisdiction previously ascertained by law’ (Article 25), thus ensuring that the tribunal established by law is independent and impartial. Moreover, in 1999,

unprecedented amount of power from representative institutions to judiciaries is in Ran Hirschl, Towards Juristocracy. The Origins and Consequences of the New Constitutionalism (Cambridge MA, Harvard University Press, 2007). More generally it has been argued that the excessive ‘tyrannophobia’ is merely a misperception of risk which unduly constrains the executive and has little social utility (see Eric Posner and Adrian Vermeule, ‘Tyrannophobia’, in T Ginsburg (ed), Comparative Constitutional Design (Cambridge, Cambridge University Press, 2012) 317, 345 f). 11 For an analysis of the main features of Italian constitutional system, see, among others, Tania Groppi and Andrea Simoncini, Introduzione allo studio del diritto pubblico e delle sue fonti, 2nd edn (Turin, Giappichelli, 2013); Roberto Bin and Giovanni Pitruzzella, Diritto costituzionale, 14th edn (Turin, Giappichelli, 2013); Augusto Barbera and Carlo Fusaro, Corso di diritto pubblico, 7th edn (Bologna, Il Mulino, 2012); Paolo Caretti and Ugo de Siervo, Istituzioni di diritto pubblico, 10th edn (Turin, Giappichelli, 2010). 12 An in-depth overview can be found in Pasquale Costanzo, Antonio Ruggeri, Lara Trucco (eds), ‘Elezioni politiche 2018: come funziona il Rosatellum bis. Il sistema elettorale spiegato dagli esperti’, (2018) Giurisprudenza italiana. Gli speciali. 13 At the top of the ordinary system of courts is the Supreme Court of Cassation (Corte di Cassazione). 14 Administrative courts of first instance are organised on a regional basis (Tribunale amministrativo regionale) while the last instance court is the Council of State (Consiglio di Stato).

360  Marta Cartabia and Barbara Guastaferro Article 111 of the Constitution was amended to bolster the general principles of due process of law. Nevertheless, the Strasbourg Court has on several occasions condemned Italy for in absentia judgments that do not comply with the procedural guarantees provided for in Article 6 of the ECHR, as well as for the unreasonable duration of the trials. Although belonging to the judicial branch, the Constitutional Court is an ad hoc body entrusted with the constitutional review of legislation.15 Besides this basic function, the Court also behaves as an arbiter in conflicts related to the horizontal and vertical separation of powers.16 In this respect, it is called upon to resolve competence conflicts between the state and regions, and to ensure that laws governing the allocation of power among institutional bodies are observed.17 It is important to recall that only in constitutional controversies related to the horizontal and vertical separation of power, such as the case in which the state challenges a regional law, the Court can be directly appealed. In all other cases, there is only indirect access to the Constitutional Court, in the sense that allegedly unconstitutional laws can be directly challenged before the Court only by judicial authorities rather than a party or individual. Consequently, when the former must resolve a dispute that requires the application of a (possibly unconstitutional) legal provision they need to suspend the case and refer to the Constitutional Court, who is the sole body capable of judging the constitutional or unconstitutional nature of a statutory provision.18 It is worth pointing out that the Constitutional Court was created only in 1956, almost 10 years after the entry into force of the republican Constitution, and that the tension between a positivistic legal culture – emphasising the supremacy of the law enacted by the Parliament in the hierarchy of norms – and the novelty of constitutional review of legislation has always been very strong, thus triggering an early body of case-law that was very deferential to legislative choices.19 To conclude, an important function is carried out by the President of the Republic, elected by two-thirds of the Assembly composed by the Parliament in joint session plus some delegates representing Italian regions. The President of the Republic is the Head of State and represents national unity. Among his/her basic functions, the President shall promulgate the laws, appoints the President of the Council of Ministers and, on his proposal, the Ministers, and may dissolve Parliament or even only one House.20 As to the vertical division of power, Italy can be qualified as a unitary state organised in regional autonomies. Article 5 of the Constitution, while acknowledging that the Republic is ‘one and indivisible’, states that the Republic shall recognise and promote local autonomies.

15 As per Art 134, the Court can scrutinise both Parliament’s ordinary legislation and ‘acts having the force of law’, which include also the legislative decrees and the law decrees, namely those acts issued by the Government when exercising the legislative functions normally attached to the Parliament. Constitutional review of legislation can only occur after their entry into force. In this respect, Italian model of constitutional adjudication can be qualified as ­centralised and a posteriori. 16 Also these functions are envisaged in Art 134 Cost. In addition to these, Constitutional Law no 1/1953 conferred another power upon the Court, which must pass judgment on the permissibility of any referendum requested, pursuant to Art 75 Cost. 17 A conflict may arise, for example, between a judicial organ and a House of Parliament, regarding the immunity guaranteed to Members of Parliament by the Constitution; between the Minister of Justice and the Consiglio Superiore della Magistratura regarding their respective powers over magistrates; between the Government and a public prosecutor regarding the use of classified documents (segreto di Stato). 18 For an outline on the competences, functions and composition of the Italian Constitutional Court see The Italian Constitutional Court, available at www.cortecostituzionale.it/documenti/download/pdf/The_Italian_Constitutional_ Court.pdf. 19 See Maurizio Fioravanti, Costituzionalismo. Percorsi della storia e tendenze attuali (Rome/Bari, Laterza, 2009), ch 3, ‘Costituzionalismo e positivismo giuridico’, 98–104. 20 See, also for the other functions, Arts 87 and 88 Cost.

Italy  361 As a matter of fact, the establishment of the regions only occurred in the 1970s, namely over 20 years after the drafting of the republican Constitution. Nevertheless, starting from the 1990s, the legal order experienced a process of decentralisation from the state to lower levels of government (regions, provinces, and municipalities) in the name of the principle of subsidiarity, aimed at bringing public administration closer to the citizens. The process of decentralisation culminated with the constitutional reform of 2001, which granted more powers to the regions and to local bodies including the power of income and expenditure related to fiscal autonomy.21 Just to give an example, the legislative function was reallocated between the state and the regions in a way that favoured the latter. While the 1948 Constitution attached to the state a general legislative function, enumerating the (few) competences of the regions (to be shared with the state), the current version of Article 117 enumerates the subject areas in which the state holds an exclusive legislative competences and the areas in which the state and the regions share legislative competences, leaving to the regions the power to legislate in all the remaining subject areas. The 2001 amendment process also eliminated some of the procedural devices that qualified the regions as subordinate bodies vis-à-vis the state, such as the Government’s pre-enactment scrutiny of regional laws, which significantly constrained the legislative discretion of regional autonomy.

C.  Prospects for Constitutional Reform A proposal to amend the Constitution was presented by the Renzi Government on 8 April 2014. If enacted, it would have affected two pivotal features of Italian legal order, specifically the ‘perfect bicameralism’22 and the division of legislative, administrative and regulatory powers between central government and the regions. As to the first aspect, the Senate of the Republic could constitute a ‘Senate of the Autonomies’ where Members were no longer directly elected by citizens but by the legislative assemblies of the Regions (Regional Councils) they are meant to represent. Consistently with one of the functions performed by Second Chambers in other constitutional systems,23 the Senate should have represented territorial institutions. Accordingly, only the House of Deputies, still representing the nation, was called to give and revoke confidence to the Executive and to behave as the principal actor of the legislative process (notwithstanding the possibility for the Senate to have a say, particularly when issues concerning the regions were at stake24). Aside from overcoming the identity of functions between the two Houses of Parliament entailed by ‘perfect bicameralism’, the constitutional reform aimed to reshuffle the vertical allocation of power, by abolishing the possibility for the state and the regions to share legislative powers in the same subject areas and by transferring many of the shared competences into the list of the exclusive powers of the state. This reform raised some criticism because of its ‘centripetal’ taste, which seemed to bring back significant powers to the Government at the expense of territorial

21 Constitutional Law no 3/2001. 22 On this aspect of the reform, see Giacinto della Cananea, ‘The End of (Symmetric) Bicameralism or a Novus Ordo?’; Lorenza Violini, ‘The Reform of Italian Bicameralism: Current Issues’; Giulio Enea Vigevani, ‘The Reform of Italian Bicameralism: the First Step’, both in (2014) 6 Italian Journal of Public Law 1, 33, 55. 23 Stefano Mannoni, ‘The “Second Chamber”: a Historical and Comparative Sketch’, (2014) 6 Italian Journal of Public Law 8. 24 Sandro Staiano, ‘Le leggi monocamerali (o più esattamente bicamerali asimmetriche)’, (2016) 1 Rivista AIC, 1.

362  Marta Cartabia and Barbara Guastaferro autonomies,25 and to weaken the role of the Parliament. In the end, although the Parliament voted in favour of the constitutional revision proposed by the Government (after significantly adjusting it), the constitutional reform did not survive the popular vote. With a turnout of 65.5 per cent of electors, 59.1 per cent of citizens rejected the reform through a referendum held on 4 December 2016. On the contrary, a constitutional referendum significantly backed a constitutional law approved on 19 October 2020 with a large majority within the Parliament (with only 14 votes against out of 569) and aimed to amend Articles 56, 57, and 59 of the Italian Constitution in order to reduce the number of the members of the Italian Parliament (from the current 630 to 400 Deputies and from the current 315 to 200 Senators). The objective of the constitutional amendment was to favour an improvement in the decision-making process of the Chambers, and, at the same time, to decrease the costs of politics. Indeed, before the reform, Italy was the country with the highest number of parliamentarians directly elected by the people (945), followed by Germany, Great Britain and France.

II.  Constitutional Culture The formal features of the Constitution analysed thus far have at times been strengthened and at others been undermined by the concrete implementation of the same Constitution that has itself been significantly affected by the constitutional culture of the country. In contrast to the important role attached to the Parliament – which holds the legislative function because it represents the general will of the people – the constitutional provisions which delegate, in exceptional circumstances, the legislative function to the Government have been to a certain extent abused. Indeed, as per Article 76 of the Constitution, legislative competences can be delegated to the Government provided detailed principles and criteria have been established by the Chambers in the law of delegation (and only for a limited time and for specified purposes). Moreover, as per Article 77, the Government can adopt provisional measures having the force of statutes, which must be confirmed by the Parliament within 60 days from their publication. Although the ‘formal Constitution’ provides that this can happen only in ‘extraordinary cases of necessity and urgency’, the ‘material Constitution’ witnessed an expansion of this form of legislation. By way of contrast, the other basic feature of the Constitution, namely its rigidity, has been respected. Indeed, there have been few constitutional amendments and this sort of ‘conservative’ approach to the constitutional text is possibly due to two technical reasons. The first reason is the complex procedure to revise our Constitution as per Article 138, which requires a large consent among major political parties and, in some cases, the approval of the people through a constitutional referendum. Interestingly enough, a quite consolidated trend sees the success of detailed and narrow reforms of the Constitution and the failure of major constitutional reform which would significantly change the kind of institutional balance imagined by the ‘Founding Fathers’. The second reason is what one of the most influential members of the Italian Constituent Assembly, Piero Calamandrei, calls the ‘far-sighted’26 nature of many constitutional provisions: our Constitution is rich in programmatic provisions that, while safeguarding the

25 Raffaele Bifulco, ‘A New Senate? A First Look to the Draft Constitutional Bill’, (2014) 6 Italian Journal of Public Law 46. 26 Piero Calamandrei, Questa nostra Costituzione, Introduction by A Garrone (Milan, Giuffrè, 1955).

Italy  363 constitutional ‘tension towards eternity’, have guaranteed the ‘adaptability’ of our constitutional text to the dynamic developments of facts.27 Needless to say, the generic nature of some of the constitutional provisions, which are sometimes framed as declarations of principles, produced also some side-effects. The realisation of the constitutional compact needs the concrete action of the legislator which, if negligent, does not implement basic provisions of the Constitution. It also requires that the interpretative function is carried out by the relevant constitutional actors, such as judges, in a way which is consistent with the Constitution.28 As a matter of fact, the so-called revolutions29 have sometimes remained solely on paper, as bitterly acknowledged by Calamandrei with regard to many provisions of the Italian Constitution a few years after it came into force.30 Nevertheless, since the very beginning of its activity, the Constitutional Court has tried to clarify that the programmatic and generically framed nature of some constitutional provision do not prevent the Constitution from being directly applicable and justiciable in concrete cases.31 The Constitutional Court also accompanied the evolution of another basic feature of Italian Constitution, namely its openness towards international law, a feature which is common to other Constitutions born after World War II and founded upon the rejection of the totalitarian experiences.32 For example – although espousing a ‘dualistic’ reading of the relationship between the domestic and international legal order – Article 10 of the Italian Constitution allows for a permanent accommodation of international customary law within the domestic legal order.33 As we will show in the following paragraphs, the Italian Constitutional Court gradually tempered this bold openness towards international law, by defending the core principles of the Constitution especially when faced by the several challenges posed by EU law. Indeed, Italy’s membership in the European Union has, over time, reshaped some major principles of the Italian constitutional system and deeply affected the model of constitutional adjudication as originally designed by the Italian Constitution of 1948. On the one hand, the 27 On the tension between the conservative nature of the Constitution and necessity of its evolution see Massimo Luciani, ‘Dottrina del moto delle costituzioni e vicende della costituzione repubblicana’, (2013) 1 Rivista AIC, 1. 28 On the obligation upon the common judge to interpret the laws in light of the Constitution see, among others, Elisabetta Lamarque, Corte Costituzionale e giudici nell’Italia repubblicana (Rome/Bari, Laterza, 2012); Giusi Sorrenti, L’interpretazione conforme a Costituzione (Milan, Giuffrè, 2006). 29 Piero Calamandrei, ‘Cenni introduttivi sulla Costituente e i suoi lavori’, in P Calamandrei (ed), Scritti e discorsi politici vol II (Florence, La Nuova Italia, 1966) 421. 30 Piero Calamandrei, ‘La Costituzione e le leggi per attuarla’, in P Calamandrei (ed), Opere giuridiche vol III (Naples, Morano, 1965) 553. 31 See in particular the first judgment of the Constitutional Court (CCIT 05.06.1956, Sentenza 1/56, ECLI:IT:COST:1956:1) and Valerio Onida, ‘L’attuazione della Costituzione tra magiustratura e Corte costituzionale’, in Aspetti e tendenze del diritto costituzionale. Scritti in onore di C. Mortati, IV (Milan, Giuffrè, 1977); and Sergio Bartole, Interpretazioni e trasformazioni della Costituzione repubblicana (Bologna, Il Mulino, 2004). 32 It is possible to find the roots of this phenomenon even earlier, looking back at what, in the 1930s, Mirkine-Guetzévitch called the ‘internationalization of modern constitutions’ (see Boris Mirkine-Guetzévitch, Les nouvelles tendances du droit constitutionnel (Paris, Marcel Giard, 1931) 48, cf Giuseppe Martinico, ‘Constitutionalism, Resistance and Openness: Comparative Law Reflections on Constitutionalism in Postnational Governance’, (2016) 35 Yearbook of European Law 318). 33 Nevertheless, with Judgment 238/2014, the Italian Constitutional Court deemed that the international customary law of state immunity from the civil jurisdiction of another state does not have any effect in the Italian legal order if ­inconsistent with the fundamental right to defence provided by the Italian Constitution (CCIT 22.10.2014, Sentenza 238/2014, ECLI:IT:COST:2014:238). As an application of the ‘counter-limits doctrine’ also to international customary law, the judgment received a great echo. See, among others, Anne Peters, ‘Let Not Triepel Triumph – How To Make the Best Out of Sentenza No 238 of the Italian Constitutional Court for a Global Legal Order’, EJIL:Talk 22.12.2014, www.­ejiltalk.org; Nico Krisch, ‘The Backlash against International Courts’, Verfassungsblog 16.12.2014, www.verfassungsblog.de; Christian Tomuschat, ‘The National Constitutional Court Trumps International Law’, (2014) 6 Italian Journal of Public Law 189; Elisabetta Lamarque, ‘Some WH Questions About the Italian Constitutional Court’s Judgement on the Rights of the Victims of the Nazi Crimes’, (2014) 6 Italian Journal of Public Law 197.

364  Marta Cartabia and Barbara Guastaferro principle of supremacy has in fact introduced a new source of higher law, which competes with the national constitutional principles as a benchmark for judicial review of national legislation. On the other hand, direct effect entrusts all lower courts, rather than only the Constitutional Court, with the power of performing judicial review of legislation in relation to European law. Put another way, one could say that participation in the European Union has introduced a ‘second judicial review of legislation’, parallel to (and potentially competing with) the original one. It is no surprise, then, that the Italian Constitutional Court – like higher courts in other Member States – firmly resisted the principles of supremacy and direct effect of European law, as we will see in section IV.

III.  Constitutional Foundations of EU Membership and Closely Related Instruments Italy is a country characterised by a dualist tradition in which the international and national legal systems are each conceived as autonomous. There are two fundamental provisions in the Italian Constitution regarding the relationship between international and domestic law: Articles 10 and 11. These articles draw an important distinction between general rules of international law and international treaties. According to Article 10, the former, which correspond to customary international law, are directly incorporated into the Italian system and enjoy the same rank as constitutional provisions; although they are required to abide by the supreme fundamental principles of the Constitution, as stated by the Constitutional Court in Judgment 238/2014. In contrast, international treaties must be incorporated into national law, in order to be binding within the domestic system. According to Articles 80 and 87 of the Constitution, ratification is by the President of the Republic, but in some enumerated cases – which cover a wide spectrum of treaties – a previous Act of Parliament is required. In practice, most international treaties become part of the Italian system through a legislative Act of Parliament, which confers on the treaty provisions the same binding value of ordinary legislative statutes that are notoriously subordinate to the Constitution. As a consequence, international treaties are usually considered of lesser authority than constitutional law in the hierarchy of sources of law. Originally, both the Founding Treaties of the European Union and the European Convention of Human Rights were assimilated to every other international treaty. Eventually, their position was differentiated. Since the 1970s, EU legislation was considered to fall within the scope of Article 11: ‘Italy agrees on conditions of equality with other States, to the limitation of sovereignty that may be necessary to a world order ensuring peace and justice among the Nations.’ Consequently, and thanks to the gradual acknowledgment of the principle of supremacy by the case law of the Italian Constitutional Court, EU legislation now enjoys a higher rank and supersedes domestic ordinary law in cases of conflict. Even if Article 11 was not framed in view of the European Community – but in view of international organisations promoting peace, such as the United Nations – it has always been construed as the constitutional foundation of Italian membership to the EU. The ‘limitation of sovereignty’ clause provided by Article 11 was considered to be broad enough to provide a constitutional foundation to the transfer of powers contemplated by all EC and EU Treaty amendments. Besides Article 11, which has been in the Constitution since its inception, a new relevant article was introduced by a constitutional amendment in 2001, representing the very first explicit constitutional reference to EU law. In this regard, Article 117, paragraph 1, states ‘Legislative

Italy  365 powers shall be vested in the State and the Regions in compliance with the Constitution and with the constraints deriving from EU legislation and international obligations.’ This provision provoked a significant doctrinal debate and divided scholars.34 According to some, the novel formulation of Article 117 entailed a sort of constitutional codification of the primacy of EU law. Through this sort of ‘subordination’ of national legislation to EU law, Article 117 was assumed to pave the way for a ‘monist’ interpretation of the relationship between legal orders.35 By way of contrast, other scholars did not attach this explicit acknowledgment of EU law within the Constitution to any possible monistic revirement in conceptualising the relationship between the domestic and the European legal orders. According to this reading, rebuffing the possible discontinuity of the constitutional amendment, Article 117 did not trigger any further limitation of national sovereignty.36 The latter position seemed to be endorsed by the Constitutional Court, which clarified that it is Article 11 of the Constitution – allowing the Republic to limit its sovereignty in favour of international organisations – to represent the ‘sound’ constitutional foundation of EU membership. Article 117 would in this respect merely codify one of the obligations already arising from Italian participation to the EU: namely, that national legislator is constrained by the respect of EU law while exercising its legislative function.37 Using Monica Claes’ categorisation of the European clauses provided by the Constitutions of the Member States of the Union, we might argue that Article 11 is the only ‘allez provision’,38 namely the only constitutional provision legitimising EU membership and allowing for a transfer of power and sovereignty to the EU. Nevertheless, this does not mean that the novelty provided for by Article 117 was without impact. First of all, and thanks to the codification of Article 117, at present both EU legislation and ECtHR judgments, although distinct from each other, enjoy a special status.39 Second, this article has strengthened the possibility of European and ECHR law acting as a basis for the constitutional review of legislation.40 Indeed, if the Constitutional Court is entitled to 34 For an overview, see Roberto Chieppa, ‘Nuove prospettive per il controllo di compatibilità comunitaria da parte della Corte costituzionale’, (2007) 12 Il Diritto dell’Unione Europea 493, 493 f; Barbara Guastaferro, Legalità sovranazionale e legalità costituzionale. Tensioni constitutive e giunture ordinamentali (Turin, Giappichelli, 2013). 35 Francesco Paterniti, ‘La riforma dell’art. 117, 1° co. della Costituzione e le nuove prospettive dei rapporti tra ­ordinamento giuridico nazionale e Unione Europea’, (2004) 49 Giurisprudenza Costituzionale 2101, 2106; Luisa Torchia, ‘I vincoli derivanti dall’ordinamento comunitario nel nuovo Titolo V della Costituzione’, (2001) 29 Le Regioni 1203. 36 Federico Sorrentino, ‘Regioni, diritto internazionale e diritto comunitario. Nuovi profili costituzionali dei rapporti tra diritto interno e diritto internazionale e comunitario’, paper presented at the conference ‘Regioni, diritto internazionale e diritto comunitario’, Genoa 23.03.2002, www.associazionedeicostituzionalisti.it; Cesare Pinelli, ‘I limiti generali alla potestà legislativa statale e regionale e i rapporti con l’ordinamento comunitario’, (2001) 126 Il Foro italiano 194. 37 See in particular CCIT 05.11.2008, Sentenza 368/2008, ECLI:IT:COST:2008:368; and CCIT 21.06.2010, Sentenza 227/2010, ECLI:IT:COST:2010:227. 38 Monica Claes, ‘Le “clausole europee” nelle costituzioni nazionali’, (2005) 2 Quaderni costituzionali, 283, 286 f. 39 More recently the same special status was recognised in relation to the European Social Charter (see CCIT 11.04.2018, Sentenza 120/2018, ECLI:IT:COST:2018:120). Although this report is not focused on the ECHR, it is worth summarising the state of the art outlined by the Constitutional Court in relation to the status of the ECHR in the domestic legal order. The ECHR has a supra-legislative value (ie its normative ranking is half-way between statutes and constitutional norms) and can serve as an ‘interposed parameter’ for reviewing the constitutionality of primary laws, since the conflict between them and the ECHR can result in an indirect violation of the Constitution (Art 117 Cost). The constitutional preferential treatment accorded to the ECHR implies the necessity to interpret national law in light of ECHR provisions, but it still requires ECHR to respect the Constitution. Last but not least, the ECHR cannot be treated domestically in the same way as EU law, in the sense that common judges cannot set aside national legislation inconsistent with ECHR but must necessarily refer the question to the Constitutional Court. See Marta Cartabia, ‘National Constitutional Adjudication in the European Space’, in Barsotti et al, Italian Constitutional Justice (n 8) 205, 219–26. In contrast to Italy, other countries give the same legal status to EU law and the ECHR law. For a comparative study on the domestic impact of both EU law and ECHR, see G Martinico and O Pollicino (eds), The Interaction between Europe’s Legal Systems. Judicial Dialogue and the Creation of Supranational Laws (London, Edward Elgar, 2012). 40 It is worth specifying this as in other countries, such as France, the Constitutional Council restricts its review to the conformity of laws solely with the Constitution. Treaties and other international or European commitments are therefore

366  Marta Cartabia and Barbara Guastaferro scrutinise the compliance of statutory law with the Constitution, and the Constitution itself wants national legislation to be constrained by international obligations and EU legislation, these ‘external’ sources can more easily enter the constitutional adjudication process. As a matter of fact, after the constitutional amendment of Article 117, paragraph 1, many actions were brought before the Court to challenge national legislation for inconsistency with Article 117 paragraph 1 and therefore with EU law.41 It can be argued that this development is one of the factors that favoured the recent openness of our Constitutional Court towards the European Court of Justice (CJEU). As we will see in the conclusions, the Court was forced to drop its ‘agnostic’ attitude towards EU law, and, eventually, to refer a preliminary ruling to the CJEU after several years of ‘self-marginalisation’ from the constitutional dialogue with the CJEU. It is important to underline that thanks to the ‘overstretching’ of our ‘allez provision’, Article 11 allows for limitation of sovereignty in favour of international organisations. And for a long time there has been no ‘retour provision’42 in our Constitution, namely no provision ­representing a response of national Constitutions to the European integration process. No specific constitutional amendment was required to accommodate, for example, the extension of electoral rights to non-Italian citizens.43 Against this backdrop, the 2012 constitutional amendment, which changed Article 81 of the Italian Constitution in order to comply with EU guidelines on balanced budget required by the so-called ‘Fiscal Compact’,44 is quite a novelty for our political praxis. According to its novel formulation, Article 81 states that ‘The State shall balance revenue and expenditure in its budget, taking account of the adverse and favourable phases of the economic cycle.’ Moreover, explicit references to EU law were introduced in Articles 97 and 119, which require, respectively, public administrations and lower levels of government (such as Regions, municipalities etc.) to ensure compliance with the economic and financial constraints deriving from the European Union.45 Some authors have tried to argue that the Fiscal Compact is invalid, but in light of its presumed conflict with the EU Treaties.46 Some others have criticised the constitutional reform for imposing too many constraints on the budgetary autonomy of local government and, most importantly, on social rights.47 Others have doubted the decision to no parameters of constitutional review. This is why the first preliminary ruling from the French Conseil Constitutionnel has been qualified an ‘unlikely’ one, see François-Xavier Millet, ‘How much lenience for how much cooperation? On preliminary reference of the French Constitutional Council to the Court of Justice’, (2014) 51 CMLRev 195, 198. 41 See for example CCIT 24.10.2005, Sentenza 406/2005, ECLI:IT:COST:2005:406, where the Court declares a regional law unconstitutional because it is incompatible with the provision of an EU directive (and then enacted in violation of Art 117 of the Constitution which requires national legislation to be enacted in harmony with EU legislation). For a survey of the case law in which the Court takes into account EU law in the constitutional review of legislation, see Guastaferro, Legalità sovranazionale (n 34) 210–16. 42 According to the above-mentioned Monica Claes tassonomy. 43 This was for example the case in other Member States, as noted in the European Parliament comparative study ­written by Leonard Besselink, Monica Claes, Šejla Imamović, Jan Herman Reestman, National Constitutional Avenues for further European Integration (Brussels, European Parliament, 2014) 147–58. 44 Constitutional Law No 1 of 20 April 2012. 45 Art 97 Cost: ‘General government entities, in accordance with EU law, shall ensure balanced budgets and the ­sustainability of public debt’. Art 119 Cost subjects the budgetary autonomy of municipalities, provinces and metropolitan cities to the ‘obligation to balance their budget’ and requires them to ‘contribute to securing compliance with the economic and financial constraints imposed under European Union law’. Also Art 117 dedicated to the allocation of powers between the state and the regions, has been amended to include the harmonisation of public budgets among the fields in which the state holds an exclusive legislative power. 46 For instance Giuseppe Guarino, Cittadini europei e crisi dell’Euro (Naples, Editoriale Scientifica, 2014). For a strong criticism more based on economic arguments see Marcello Degni and Paolo De Ioanna, Il Vincolo stupido (Rome, Castelvecchi, 2012). 47 Massimo Luciani, ‘Costituzione, bilancio, diritti e doveri dei cittadini’, (2013) 174 Astrid Rassegna; Silvio Gambino and Walter Nocito, ‘Crisi dello Stato, governo dell’economia e diritti fondamentali: note costituzionali alla luce della

Italy  367 amend the Constitution, since the EU did not expressly require it, but left the Member States discretionary power on the modality through which the principle is introduced to the domestic legal order.48 Nevertheless, despite the attention of scholarly literature, the parliamentary debate on the domestic impact of European economic governance arising from the financial crisis was very poor. As was pointed out in both Chambers and in the relevant Parliamentary Committees, the procedure for authorisation of the Fiscal Compact was combined with other procedures related to the Treaty amendment in Article 136(3) TFEU and the ESM Treaty.49 Moreover, there has been no constitutional litigation on the validity of these measures and this ‘seems to be explained first of all by the way the system of constitutional adjudication is designed’.50 In other words, because of the absence of something like the Verfassungsbeschwerde or the possibility for parliamentary minorities to bring a case before the Constitutional Court, the latter did not have the chance to be involved in this issue. On the contrary, the Constitutional Court is currently referring to these amended provisions (in particular to Articles 81 and 97) to review national and regional ­legislation which do not comply with the balanced budget principle. The poor parliamentary debate which characterised the approval of the last treaties stemming from the eurocrisis, which at least raised some criticism,51 is a constant in Italian institutional history. This is remarkable if we consider that, unlike other countries, the Italian legal order does not require a referendum to ratify international treaties. In this respect, if the direct popular will is not consulted, one would expect a lively debate in important representative institutions such as the Parliament. By way of contrast, Italy proved its adaptability to the EU legal order in ratifying very quickly and with large majorities almost all EU treaties. Just to give an example, the Italian Parliament was the first to ratify the Treaty establishing a Constitution for Europe in 2005.52 Moreover, it approved the Lisbon Treaty in 2008 with unanimous votes both in the House of Deputies and the Senate.

IV.  Constitutional Limits to European Integration53 A.  The ‘European Journey’ of the Constitutional Court towards the Acceptance of the Supremacy Principle Aware of the significant impact of EU law on the system of constitutional adjudication, the Constitutional Court had a long way to go before fully complying with the European requirements. crisi finanziaria in atto’, (2012) 162 Astrid Rassegna; Nicola Lupo and Guido Rivosecchi, ‘Quando l’equilibrio di bilancio ­prevale sulle politiche sanitarie regionali’, (2012) 5/6 Le Regioni, 1062; Daniela Mone, ‘La costituzionalizzazione del ­pareggio di bilancio ed il potenziale vulnus alla teoria dei controlimiti’, (2014) 3 Rivista AIC, 1. 48 For the EU law and constitutional law aspects of the reform see, respectively, Gian Luigi Tosato, ‘La riforma ­costituzionale del 2012 alla luce della normativa dell’Unione: l’interazione fra i livelli europeo e interno’, and Massimo Luciani, ‘L’equilibrio di bilancio e i principi fondamentali: la prospettiva del controllo di costituzionalità’, both presented at the workshop ‘Il principio del pareggio di bilancio secondo la riforma costituzionale del 2012’, 22 dicembre 2013, ITCC. 49 Leonardo Pierdominici, Constitutional Change through Euro Crisis Law. Report on Italy (Florence, EUI, 2014) http:// eurocrisislaw.eui.eu/. 50 Cristina Fasone, The reaction of Constitutional Courts towards the Euro-crisis law. The protection of social rights in Italy, Portugal and Spain, EUI Working paper MWP 2014/25 (Florence, EUI, 2014), http://cadmus.eui.eu. 51 The Fiscal Compact obtained 368 votes in favour and 68 votes against it in the House of Deputies, coming from two political parties (Italia dei valori and Lega Nord). 52 The House of Deputies approved the draft TECE on 25 January 2005, with only 28 opposing votes, and the Senate voted on 6 April 2005 with only 16 votes against. 53 This section revisits Cartabia, ‘National Constitutional Adjudication’ (n 39).

368  Marta Cartabia and Barbara Guastaferro The great distance between the Italian understanding of judicial review of legislation and the model adopted by the European Union required the Court to undertake a ‘European Journey’.54 The journey can be divided into three stages: first, constitutional resistance to the supremacy of European legislation; second, limitations of sovereignty; and third, constitutional reception of the supremacy of EU legislation.

1.  Constitutional Resistance to the Supremacy of European Legislation In the first stage of its European journey the Italian Constitutional Court was at odds with the doctrine of the Court of Justice: being international treaties incorporated into domestic law by an Act of Parliament, the Italian Court endowed them with the same rank as any other piece of ordinary legislation. The result was that they took precedence over earlier legislative acts, but could be overturned by later ordinary legislative acts according to the rule lex posterior derogat priori. Alternatively put, due to its dualistic tradition Italy was not prepared to recognise the supremacy of European law. In fact, the contrast between the two courts could not have been more explicit, and exploded when both were required to intervene in the same proceedings, concerning the Italian law nationalising the electricity industry. Within the context of the same judicial controversy, the courts reached two opposite conclusions. For the ECJ that was the very occasion to affirm the supremacy of European law: By contrast with ordinary international treaties, the [EEC Treaty] 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 … The executive force of Community law cannot vary from one State to another in deference to subsequent domestic laws, without jeopardizing the attainment of the objectives of the Treaty.55

By way of contrast, dealing with the same legal question, the Italian Constitutional Court, in its Judgment 14/1964, affirmed that European Treaties should be treated as any other international treaty. They were given the same value as ordinary laws of the Italian Parliament and therefore cannot prevail over more recent pieces of legislation. The disagreement with the ECJ was complete.56 It might be interesting to note that, similarly, the supremacy of Community law was also not initially welcomed in other Member States, like France and the United Kingdom, because it did not fit with the traditional constitutional principle of the sovereignty of the Parliament.

2.  Limitations of Sovereignty In the following years, however, the Italian Constitutional Court incrementally tempered its attitude of opposition to the supremacy of European law and distinguished the status of European Treaties from all other international norms. The pivot of this new line of reasoning was the idea of ‘limitations of sovereignty’, mentioned in Article 11 of the Constitution. This new trend in the case law of the Constitutional Court appeared for the first time in the Frontini case.57 The Court was asked to decide whether a crucial provision of the European Treaties, Article 189 of the EEC



54 Paolo

Barile, ‘Il cammino comunitario della Corte’, (1973) 18 (1) Giurisprudenza costituzionale, 2406. 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66. 56 CCIT 24.02.1964, Sentenza 14/1994, ECLI:IT:COST:1964:14. 57 CCIT 18.12.1973, Sentenza 183/1973, ECLI:IT:COST:1973:183. 55 ECJ

Italy  369 Treaty (now Article 288 TFEU) that confers on European institutions the power to issue normative acts, like regulations and directives, could be read as compatible with the Constitution. It is in this context that the idea of limitations of sovereignty plays a decisive role: This formula [limitations of sovereignty] legitimates the limitation of the powers of the State with respect to its legislative, executive, and judicial functions, which was necessary for the institution of a Community among the European States … Italy and the other sponsoring States conferred upon and recognised in the Economic Community … certain sovereign powers, constituting it as an institution characterised by an autonomous and independent legal order. In particular, with Art. 189 of the Founding Treaty, the Council and the Commission of the Community were given the power to issue rules with general application – that is … acts having a general normative content equal to State laws, having obligatory efficacy in all their parts, and directly applicable in each of the Member States, or in other words immediately binding on the States and their citizens, without the need for internal measures of adaptation of reception.

Here the Constitutional Court abandoned the framework of classical international law, because European integration has special features and cannot be adequately understood through the lens of the general principles of international law. With membership in the European Communities, the Italian institutions – Parliament, Government, judiciary, regions and even the Constitutional Court – have limited their sovereign powers and surrendered some to the European institutions. Insofar as the European institutions perform their activities within the boundaries of competences established by the Founding Treaties, their acts – regulations and directives – enjoy a special status, since they can be directly relevant within the domestic legal order, with no need to be ‘transplanted’ into domestic legislation. The Court thus paved the way for the principle of supremacy and direct effect of European legislation within the limit of competences established by the Treaties. However, even after this important step, some hidden ambiguities still remained unresolved and were bound to come to light, sooner or later, in the subsequent case law of the Italian Constitutional Court.

3.  Constitutional Reception of the Supremacy of EU Legislation Once the Constitutional Court accepted the supremacy of European law on the basis of Article 11 of the Italian Constitution, a new problem arose: who was the guardian of the supremacy of European Law within the national legal system? On this point, the ECJ and the Constitutional Court again took opposite positions at the outset. In contrast with the Italian tradition of centralised judicial review, where only the Constitutional Court is vested with the exclusive power of judging the laws, in the Simmenthal Judgment of 1978, the Court of Justice stated that: Every national court must, in a case within its jurisdiction, apply Community law in its entirety and protect rights which the latter confers on individuals and must accordingly set aside any provision of national law which may conflict with it, whether prior or subsequent to the Community rule.58

This conferral of powers on ‘every national court’ to set aside national legislation was a shock for the Italian system, and brought about a true innovation, both from a theoretical point of view and from a practical one. Within the doctrinal framework of the Constitutional Court, each national provision conflicting with European legislation was to be considered contrary to Article 11 of the Constitution. Consequently, each conflict between national and European norms fell within

58 ECJ

09.03.1978 106/77 (Amministrazione delle Finanze dello Stato v Simmenthal SpA) ECLI:EU:C:1978:49, para 21.

370  Marta Cartabia and Barbara Guastaferro the jurisdiction of the Constitutional Court. In contrast, the Court of Justice dealt with conflicts between European laws and national legislation with a different perspective, as if the two sets of rules were part of the same legal system with the national courts acting as European judges and guardians of European legislation. It is no wonder, then, that it took a number of years for Italian legal scholarship and the Constitutional Court to accept the ‘silent revolution’ brought about by the Court of Justice in Simmenthal. Only in 1984, with the Granital judgment, did the Constitutional Court overrule its previous case law, affirming that European Community law must always be applied, whether it precedes or follows the ordinary laws incompatible with it: and the national judge, vested with the respective application may benefit from the assistance offered by the preliminary question of interpretation, in the sense of Article 177 [now Article 267 TFEU] of the Treaty.59

As the Constitutional Court explained in Granital: The effect of the [European legislation] is not that of annulling, in the proper meaning of the term, the incompatible internal norm, but of preventing that norm from being applied for the resolution of the controversy before the national judge … For the very reason of the distinction between the two legal orders, the priority of the regulation adopted by the [European Community] means … that the [national] law does not interfere in the sphere which the [European] act occupies, which is completely pre-empted by the Community law … Outside of the material domain and the temporal limits in which the Community discipline is in force, the national rule preserves, intact, its proper legal authority and keeps its legal efficacy.

After this decision, the task of directly applying European law was shifted to the lower courts with the Constitutional Court preserving some jurisdiction only in residual and marginal cases. The Constitutional Court has now therefore reached a ‘practical concordance’ with the Court of Justice, even though important differences still remain at the theoretical level, insofar as the Court of Justice’s case law is still based on a monist principle, whereas the Italian Constitutional Court still adheres to a dualist doctrine. More recently, the Constitutional Court has further specified the division of labour between lower courts and the Constitutional court itself in the field of the implementation of EU legislation, expressing its willingness to have a direct say especially in the judicial review of national legislation infringing upon fundamental rights as protected by both the Italian Constitution and the Charter of Fundamental Rights of the EU (CFR). The Constitutional Court based its reasoning on the following principles. First of all, since the violation of an individual right might violate, at once, both the guarantees enshrined in the Italian Constitution and those codified by the CFR,60 ordinary judges might ask the Constitutional Court about possible conflicts between national law and the Charter. Should this arise, in assessing whether the contested provision violates both the Constitution and the CFR, the Court might submit a request for a preliminary ruling to the EU Court of Justice whenever that proves necessary to clarify the meaning and the effects of the Charter’s rules. Second, the Court specified that, at the outcome of that assessment, the Court might invalidate with erga omnes effects the provision deemed to be unconstitutional, thus removing it from the national legal system: this without affecting the power of ordinary courts to refer a preliminary ruling on similar issues to the Court of Justice of the European Union,61 and without affecting their duty not to apply

59 ITCC 60 ITCC 61 ITCC

05.06.1984, Sentenza 170/1984, ECLI:IT:COST:1984:170. 20.12.2017, Sentenza 269/2017, ECLI:IT:COST:2017:269, para 5.2 Considerato in diritto. 23.01.2019, Sentenza 20/2019, ECLI:IT:COST:2019:20, para 2.3 Considerato in diritto.

Italy  371 national provisions inconsistent with the rights laid down in the Charter where appropriate.62 The Court has further clarified, in this regard, that the guarantees set down by the Charter enrich the tools for protecting fundamental rights fostering a plurality of legal remedies which should be informed by the principle of loyal cooperation and should be finalised to assure the maximum protection of rights in light of Article 53 CFR.63 Such a plurality of remedies actually allows the Constitutional Court to give effect to the possibility, discussed in Article 6 of the Treaty on European Union (TEU) … that the corresponding fundamental rights guaranteed by European law, and in particular by the Charter, are interpreted in harmony with the constitutional traditions common to the Member States, also mentioned by Article 52(4) of the CFR as relevant sources.64

The result of this new course is a sort of concurrent judicial review of national legislation when the European law is involved: the ordinary court can set aside the national legislation in contrast with a provision of the EU law having direct effect; however, if they consider more appropriate to involve the Constitutional Court, in particular in the field of protection of fundamental rights, the latter maintains its jurisdiction and can decide the case with general effect declaring the invalidity of the Italian law incompatible with the European rights, norms and principles.

B.  Substantive Limits to European Integration: The Doctrine of ‘Counter-Limits’ The assent to the position of the ECJ by the Constitutional Court analysed in the previous ­paragraph was accompanied by some reservations. The Italian Constitutional Court, like many other national constitutional courts – starting with the German Bundesverfassungsgericht (hereafter BVerfG) – has accepted the supremacy of European law ‘subject to conditions’. While European legislation in general prevails over national legislation (including the national Constitution, as required by the ECJ65), European norms are not allowed to override the core principles of the Italian Constitution, in particular the constitutional guarantees of fundamental rights. The core principles of the national Constitution are considered ‘counter-limits’ to the limitations of sovereignty. It is on the basis of this understanding that the national powers agreed to reduce their functions in order to enter the European Communities. As a result, European legislation and other European acts prevail over all national legislation, and can even derogate from the national Constitution, but only within the boundaries of these counter-limits. In order to understand this peculiar notion of counter-limits, one should consider that when the primacy of European law was established in the case law of the ECJ in the 1960s and 70s, there was no bill of rights in any European Treaty and the ECJ had declined66 any competence to rule on the protection of fundamental rights. As a result, national constitutional courts were concerned about the lack of protection afforded human rights, especially in cases of human rights violations produced by European acts. In particular, the German Federal Constitutional Court firmly expressed this concern in the Solange judgments of 1974 and 1986. Even before the

62 ITCC 20.02.2019, Sentenza 63/2019, ECLI:IT:COST:2019:63, para 4.3 Considerato in diritto. 63 ITCC 269/2017 (n 60), para 5.2. 64 ITCC 20/2019 (n 61), para 2.3. 65 ECJ 17.12.1970 11/70 (Internationale Handelsgesellschaft mbH v Einfuhr- und Vorratsstelle für Getreide und Futtermittel) ECLI:EU:C:1970:114. 66 ECJ 04.02.1959 1/58 (Stork & Cie v High Authority) ECLI:EU:C:1959:4.

372  Marta Cartabia and Barbara Guastaferro BVerfG, the Italian Constitutional Court clearly established the doctrine of counter-limits in its Frontini judgment:67 On the basis of Article 11, limitations of sovereignty have been allowed solely for the attainment of the goals indicated there; and it must therefore be ruled out that those limitations concretely delineated in the Treaty of Rome – adopted by Nations with legal systems that are inspired by the rule of law and that guarantee the essential liberties of their citizens –may in any case entail for the organs of the Community an inadmissible power to violate the fundamental principles of our constitutional order or the inalienable rights of the human person. And it is obvious that in the event that Article 189 should ever be given such an aberrant interpretation, the guarantee of the jurisdiction of this Court to decide the enduring compatibility of the Treaty with the aforesaid fundamental principles will always be assured.

In sum, European legislation must respect the core principles of the Italian Constitution from a substantive point of view; in the event of a violation of those principles, the Constitutional Court claims jurisdiction. What might be the consequences of the EU’s violation of the core principles of the national Constitution? In the Frontini judgment the Constitutional Court asserted that in case of infringements on the counter-limits the membership of Italy in the European Union would be in question. Later, however, the Court tempered its position. In the Fragd judgment, it redefined the doctrine of counter-limits as the ability to verify, through the constitutional control of the executing laws, that any norm of the Treaty, in the manner in which it is interpreted and applied by the institutions and by the Community organs, is not in conflict with the fundamental principles of our constitutional order or not mindful of the inalienable rights of the human person.68

Consequently, any act or provision of European law is theoretically submitted to the authority of the Constitutional Court to make sure there is compliance with the fundamental core of the Constitution, in which the national constitutional identity is encompassed. Departing from the approach of the Bundesverfassungsgericht, which fostered a quite open struggle with the Court of Justice of the European Union in contemplating the possible ultra vires review or identity review of EU measures,69 the approach of the Constitutional Court has been quite soft, generally avoiding a concrete application of the controlimiti doctrine.70 This is also because the Court of Justice filled the gap in the protection of fundamental rights,71 allaying the fear that 67 ITCC, Sentenza 183/1973 (n 57). 68 ITCC, Sentenza 232/1989 (n 6). 69 See in particular the Lissabon Urteil, BVerfG 30.06.2009 2 BvE 2/08 (Lissabon) ECLI:DE:BVerfG:2009:es20090630. 2bve000208. Notoriously, many other national constitutional courts reiterated the counter-limits doctrine, also grounding it on Art 4(2) TEU introduced by the Treaty of Lisbon, according to which the Union shall respect Member States’ national identities, inherent in their fundamental political and constitutional structures. On the possible implication of Art 4.2 TEU, see Barbara Guastaferro, ‘Beyond the Exceptionalism of Constitutional Conflicts. The Ordinary Functions of the Identity Clause’, (2012) 31 Yearbook of European Law 263. 70 However, the Council of State, Section V, applied this doctrine in Case 4207/2005 related to the regulation of municipal pharmacies in order to justify the decision of not referring to the Court of Justice for preliminary ruling (Consiglio di Stato 08.08.2005, Sentenza 4207/2005). In 2014 the Constitutional Court applied the counter-limits doctrine to delimit the international customary rule of state immunity in a case concerning damages claimed by victims of the Nazi regime, involving gross violation of humanitarian international law (ITCC, Sentenza 238/2014 (n 33)). 71 Starting with ECJ 12.11.1969 29/69 (Stauder v Stadt Ulm) ECLI:EU:C:1969:57; ECJ 14.05.1974 4/73 (Nold KG v Commission) ECLI:EU:C:1974:51; and ECJ 13.12.1979 44/79 (Hauer v Land Rheinland-Pfalz) ECLI:EU:C:1979:290, the CJEU affirmed that ‘[f]undamental rights form an integral part of the general principles of the law, the observance of which is ensured by the Court. In safeguarding those rights, the latter is bound to draw inspiration from constitutional traditions common to the Member States, so that measures which are incompatible with the fundamental rights recognized by the Constitutions of those States are unacceptable in the Community. International treaties for the protection of human rights on which the Member States have collaborated or of which they are signatories, can also supply guidelines which should be followed within framework of Community law.’

Italy  373 citizens might be deprived of their rights without judicial remedy. Nevertheless, the firmness in the protection of national constitutional core values from possible infringements by EU law has more recently been restated in the Taricco judgment, analysed in the concluding paragraph, where the Italian Constitutional Court clearly warned the ECJ about the possibility of applying the controlimiti doctrine to protect the principle of legality in criminal matters enshrined in the Italian Constitution.72

V.  Constitutional Rules and Practice of Implementing EU Law As far as the constitutional rules or practices on implementation of EU law are concerned, the focus must be on the Italian Parliament, which has a pivotal role in participating in the framing of European policies and in transposing EU secondary law into domestic law. Italy, indeed, was the first country to contemplate a legal instrument to ensure a timely implementation of EU law to be guaranteed by the national Parliament. The necessity to have a sound legislative instrument providing for a general framework to regulate Italian participation to the EU was particularly felt in light of the laconic provisions of our Constitution on the topic and in light of significant implementation delays.73 The very first instrument was Law No 86 of 9 March 1989 (the so-called Legge La Pergola), and the following Law No 11 of 4 February 2005 (the so-called Legge Buttiglione). Currently, the procedures through which the Italian Parliament interacts with the European Union are provided by Law No. 234 of 2012.74 The latter is very important because, on the one hand, it introduces appropriate procedures to implement the new forms of participation of national Parliaments in the EU decision-making process set out by the Treaty of Lisbon (including the subsidiarity check in the context of the early warning mechanism),75 and, on the other, it strengthens the possibility for the Parliament to control and influence the activity of the Executive within the European Union context.

A.  The Duty of Information upon the Government and the Parliamentary Scrutiny Reserve Law No 234 of 2012 obliges the Executive to inform the Parliament and to transmit specific documents related to the framing of EU policies. In particular, the Government shall forward to the Chambers: • The draft legislation of the European institutions, including the preparatory acts, indicating the date on which they will be discussed (with possible profiles of urgency or priority for their

72 ITCC 23.11.2016, Ordinanza 24/2017, ECLI:IT:COST:2017:24 (Taricco). 73 For an in-depth overview see Marta Cartabia and Joseph HH Weiler, L’Italia in Europa. Profili istituzionali e ­costituzionali (Bologna, Il Mulino, 2000) 150 ff. 74 Entitled ‘General rules on Italy’s participation in the formation and implementation of legislation and policies of the European Union’, in force since 19 January 2013. 75 For a general overview on the Lisbon innovation related to subsidiarity and parliamentary democracy see M Cartabia, N Lupo and A Simoncini (eds), Democracy and Subsidiarity in the EU (Bologna, Il Mulino, 2012); Andrea Manzella, ‘The Role of Parliaments in the Democratic Life of the Union’, in S Micossi and G L Tosato (eds), The European Union in the 21st Century. Perspectives from the Lisbon Treaty (Brussels, CEPS, 2010) 257; Pier Casalena, Cristina Fasone, Nicola Lupo, ‘Commentary on the Protocol No. 1 annexed to the Treaty of Lisbon’, in H Blanke and S Mangiameli (eds), Commentary on the Treaty of Lisbon (Berlin/Heidelberg, Springer, 2013).

374  Marta Cartabia and Barbara Guastaferro discussion) and accompanying cases of particular relevance with an information note assessing the proposal. The Government also has the obligation to prepare, within 20 days of the submission of such projects, a technical report accounting for the negotiation prospects and for the impact of EU law and policy within the domestic legal order. • Consultation documents of the European Commission and the possible comments sent by the Government itself to the EU institutions. • Reports and briefing notes prepared by the Permanent Representation to the EU in relation to meetings of the EU Council and its preparatory bodies, also in relation to the pre-litigation and litigation phases of infringement procedures initiated against Italy. • A report on financial flows between Italy and the European Union, every three months. It is also planned that the Government informs the competent parliamentary bodies a priori, on the position to be acquired at the meetings of the European Council and of the EU Council of Ministers; a posteriori, on the results of the meetings of the European Council and the Council of Ministers, within 15 days after they took place; more generally, on the initiatives or issues related to foreign policy and, most importantly, defence policy, presented to the EU Council of Ministers. Indeed, Law no 234/2012 provides that the Government should present to Parliament • by 31 December of each year, a report on the guidelines and priorities that the Government plans to pursue for the following year, with reference to deepening of the European integration process, to the institutional profiles and to the development of each policy; • and by 28 February of each year a report which shows the outcomes of the activities carried out in the previous year on the above profiles, including the response of the Government to the guidelines defined by the Chambers. These reports are reviewed by all the competent parliamentary committees and by the EU Affairs Committee which reports to the House of Parliament. The process usually culminates in the adoption of a resolution. All these duties of information of the Government are finalised to involve the Parliament in the framing of EU policy or, more specifically, in elaborating the national position to be held in the international arena. Indeed, Law no 234/2012 provides for a very useful instrument to ensure that the Parliament is not circumvented in the delineation of the national position in the European intergovernmental context: the parliamentary scrutiny reserve, which can be activated either by one of the Chambers or by the Government and can regard any EU project or act that the Government is required to transmit to the Chambers. Just to give an example, the Italian House of Deputies activated it with reference to the EU Draft Regulation on the European citizens’ initiative, aimed at implementing Article 11.4 TEU. In case the parliamentary scrutiny reserve is requested by one of the Chambers, the Government informs the Council that it can proceed with its own activities in the EU arena only once the examination of the proposed act by the national Parliament is concluded (within a time limit of 30 days from the parliamentary Chamber’s request). In case the Government requests the parliamentary scrutiny reserve on a draft act or of one or more parts of it gives notice to the Chambers. Again, once the period of 30 days has expired, the Government may proceed even in the absence of a preliminary parliamentary opinion. Besides the parliamentary scrutiny reserve, there are also other instruments enabling the Parliament to control the Government, envisaged also by the Rules of Procedures of the Chambers. Just to give an example, the EU Affairs Committee and the other Standing Committees may ask for a debate with the competent minister sitting within the European Council or Council

Italy  375 of Ministers before the Council meeting takes place; or may invite members of the European Parliament and the European Commission to provide information upon EU policies and activities. Moreover, the same Law no 234/2012 requires the Government to inform the Chambers about the appointment proposal of the Italian members of the European Commission, the CJEU and many other EU institutions and agencies, and provides that parliamentary committees may ask for a hearing of the appointed members, after their taking office. The Parliament shall also be consulted beforehand in case of agreements producing significant effects on the public finances. Moreover, the Chambers, through a joint action, might ask the Government to activate an ‘emergency brake’: thanks to this procedural device the matter is referred to the European Council, thus delaying the final deliberation when sensitive issues – particularly related to national sovereignty such as criminal law – are discussed within the European Council of Ministers. It is important to underline that the Law no 234/2012 requires the governmental national position expressed within the Council to be consistent with the guidelines of the Chambers. Where the Government has been unable to follow the guidelines of the Chambers, the President of the Council of Ministers or the responsible Minister shall report to the competent parliamentary committee and give reasons for that behaviour. The instruments provided by the Lisbon Treaty undoubtedly strengthen the accountability of Governments to their Parliaments in EU affairs. The big amount of information received by the Parliament, indeed, enables it to engage in an in-depth scrutiny of the governmental activity at the EU level. As to the Italian experience, it must be acknowledged that the Minister or the undersecretary on EU affairs and the Minister of the Economy and Finance have been often asked to appear before the competent committees and to give evidence about the results of the Council of the EU’s meetings through hearings. By way of contrast, the Parliament often fails to adopt a resolution before the relevant Council’s meeting is held at EU level.76

B.  The Role of the Italian Parliament in Implementing and Scrutinising EU Measures The draft European Union legislation and related preparatory acts are assigned for examination to the parliamentary committee responsible for the matter, and for an advisory opinion to the EU Affairs Committee of the Parliament. The competent committees – which according to the standing orders must ensure ‘prompt review’ of the documents in question – issue a final document, which is transmitted to the Government and to the European Parliament, the Council and the Commission in the political dialogue. The Lisbon Treaty provides that national Parliaments could send reasoned opinions to the European institutions regarding EU draft legislative acts’ compliance with the principle of subsidiarity in the field of shared competence of the EU (the so-called Early Warning Mechanism). Where reasoned opinions reach certain thresholds, the project will have to be reviewed and it may be confirmed, modified or withdrawn by the European legislator. It is worth mentioning that the Law no 234/2012 allows also regional assemblies to transmit to the Chambers their observations on the subsidiarity on draft EU legislation.

76 Cristina Fasone and Diane Fromage, ‘Are Parliaments (More) Committed to the Trend of Open Government After the Lisbon Treaty? The Italian and the Spanish Case’, in I Bouhadana, W Gilles, I Nguyên-Duy and R Weaver (eds), Parliaments in the Open Government Era (Paris, Les éditions IMODEV, 2016) 173.

376  Marta Cartabia and Barbara Guastaferro In the Italian Parliament, it is up to the EU Affairs Committee to verify the compliance of EU draft legislative acts with the subsidiarity principle (within 40 days from the reception of the same draft proposals). Nevertheless, upon request of the Government, one-fifth of the members of that Committee, or a tenth of the members of the Assembly (advanced within five days from the date of the issuing of the resolution) the subsidiarity check can be devolved to the full Assembly; the process must still be completed within eight weeks from the submission of the project.77 In the praxis, the Presidency of the Chamber transfers to the European institutions both a negative decision concerning subsidiarity and a favourable one, to enhance the dialogue with European institutions. Since September 2006 the European Commission has maintained a so-called political dialogue, transmitting directly to national Parliaments its legislative proposals and consultation documents, and inviting them to express comments and opinions to which the Commission itself can reply. The Commission has maintained firm political dialogue even after the entry into force of the Lisbon Treaty and Law no 234/2012, codifying this practice, provides that the Chambers may submit to the European institutions, and at the same time to the Government, all the documents needed to frame and develop European policies. In practice, the political dialogue strategy has been very successful, in the sense that many national Parliaments preferred to use this instrument to entertain a dialogue with the European Commission not necessarily linked to the subsidiarity scrutiny (and not necessarily constrained by its procedural requirements78). The Annual Report on subsidiarity and proportionality issued by the Commission shows that in 2016 there was one of the highest number of reasoned opinions together with a general increase in use of the political dialogue instrument. The Commission received 65 reasoned opinions from national Parliaments in respect of the principle of subsidiarity, an increase of 713 per cent compared with the eight opinions of the previous year.79 The Italian Parliament has presented several reasoned opinions since the entry into force of the Lisbon Treaty, with the Italian Senate being more active than the House of Deputies. It can be argued that the Italian Parliament followed the trend of many other national parliaments which endorsed a broader notion of subsidiarity then the one suggested by Protocol No 2 of the Lisbon Treaty. Such a less restricted notion includes in the subsidiarity scrutiny arguments on the substance of the proposal, on the principle of proportionality and on the choice of the legal basis. Only in few reasoned opinions issued by the Italian Senate was a breach of the subsidiarity principle found to be on grounds of the ‘necessity’ and ‘EU value added test’ required by the Treaties.80 Moreover, a very recent reform by a standing order of the Italian Senate created a sort of internal division of labour which allows all the Permanent Committee of the House to enter a political dialogue with the Commission in their area of expertise but attaches solely to the European Affairs Committee the issuance of reasoned opinions scrutinising EU legislation in light of the

77 We are referring to the procedures of the House of Deputies, which after the reform of the Standing Order of the Senate have been almost homologated. For a comparison between the role of the two chambers in the European affairs see Antonio Esposito, ‘The Definition of the parliamentary position on EU affairs in the Chamber of Deputies’; and Davide Capuano, ‘The definition of the parliamentary position on EU affairs in the Senate’, both in N Lupo and G Piccirilli (eds), The Italian Parliament in the European Union (Oxford, Hart, 2017) 99 and 119, resp. 78 Notoriously, to qualify as a reasoned opinion as defined in Protocol No 2, an opinion must state why the national Parliament considers that a legislative proposal does not comply with the principle of subsidiarity and be sent to the Commission within eight weeks of the transmission of the proposal to national Parliaments. 79 European Commission, Annual Report 2016 on proportionality and subsidiarity, COM (2017) 600 final, Brussels, 30.6.2017, 4. The yearly reports are presented in accordance with Art 9 of Protocol No 2 on the application of these principles. 80 Davide Capuano, ‘The role of national parliaments in the legislative process of the EU: a view from inside the Italian Parliament’, in M Cartabia, N Lupo, A Simoncini, (eds), Democracy and Subsidiarity in the EU (Bologna, Il Mulino, 2013) 247.

Italy  377 principle of subsidiarity.81 It should also be noted that, in some cases, the Italian Parliament used the reasoned opinions also for ‘domestic purposes’, specifically to warn its national Government not to neglect its duty to promptly inform the Parliament on EU proposals,82 and in a positive way, to endorse (rather than challenge) EU proposals emphasising its compliance with the principle of subsidiarity and proportionality. Besides providing the instruments to scrutinise EU law – especially in light of the principle of subsidiarity – Law No 234 of 2012 completely revisits the instruments and procedures devoted to the implementation of EU law. Before its entry into force, the Italian Parliament implemented EU law through the adoption of the so called ‘Annual Community Law’ (legge comunitaria), which had the form of a legislative act enacted by both the Chambers. This Community Law has been substituted by two different legislative instruments: the European Delegation Law (legge di delegazione europea) and the European Law (legge europea). The new regulatory instruments are intended to remedy the problem of the long time required for approval in Parliament of the annual Community Law. Moreover, the adoption of a single overarching legal instrument per year was not enough to implement EU secondary law, and, most importantly, to adopt measures to prevent a conviction by the CJEU within the framework of infringement procedures. By way of contrast, the two new instruments are conceived of as serving two different purposes. By way of the European delegation law, the Parliament delegates to the Government the transposition of European directives and the implementation of other acts of the European Union. By way of the ‘European law’, the Parliament directly adopts measures for the fulfilment of the obligations arising from EU law. To give one example, the Parliament may amend or repeal specific parts of a law in order to eliminate the illicit provisions that are the object of an infringement procedure, before the latter reach the final stage where the CJEU impose the payment of a penalty upon the Member State. The proposal for a European delegation law shall be presented by the Government by 28 February of each year but if necessary it can be presented twice a year. It must be acknowledged that, also thanks to the new instruments, the infringement proceedings against Italy are decreasing, although still significant, especially in the environmental and the transport sectors.

VI.  Conclusion: Resulting Relationship between EU Law and National Law The European continent has become a space of constitutional interdependence and consequently, national constitutional orders are now embedded in a constitutional fabric made of national constitutions, EU law, European treaties and conventions. This constitutional interdependence affects all the institutional actors of the legislative, executive and judicial branches. In particular national courts are on the frontline and can impact the overall institutional mood towards European integration. Although rooted in the national context, they are charged with new duties

81 See Art 144 of the Standing Orders of the Senate in its novel formulation (approved on 20 December 2017). 82 Barbara Guastaferro, ‘Coupling National Identity with Subsidiarity Concerns in National Parliaments’ Reasoned Opinions’, (2014) 21 Maastricht Journal of European and Comparative Law 320. This was for example the case of the first reasoned opinion adopted by the Italian Senate, which emphasised a lack of coordination with the Government, which in contrast declared itself in favour of the proposal on the grounds of subsidiarity (see Cristina Fasone, ‘Il Senato approva il primo parere contrario sul rispetto del principio di sussidiarietà’, (2010) 4 Quaderni Costituzionali 824).

378  Marta Cartabia and Barbara Guastaferro and to an increasing extent are called to serve as European law adjudicators. In this capacity they can maintain different approaches: • Implementation and promotion: the EU primarily relies on the cooperation of lower ordinary national courts in order to implement European legislation, which are part and parcel of the European judiciary along with the Court of Justice and other judicial bodies established by the Treaties. Even national constitutional courts are increasingly involved in the mission of implementing EU legislation: after all, national Constitutions provide European clauses so that European obligations are to some extent also constitutional obligations. As a consequence, ordinary and Constitutional Courts often decide controversial issues on European grounds and they contribute a great deal to rendering European law more effective. At times, some national courts go even further and promote super-national legal concepts well beyond the European mandate. In this regard, there are a number of cases that could be resolved on purely domestic grounds but Courts intentionally broaden the scope of their judgments to include European law. They disseminate the European legal culture throughout their r­ espective legal orders. • Reluctance: at times, national courts – especially the highest courts and some Constitutional Courts – are inclined to contain the use of European law to a minimum level in their own ­decisions and decide cases on domestic grounds rather than on European ones. Additionally, they tend to avoid formal referral to super-national law and case law, and escape all contacts with other national or super-national courts. • Defence: starting in the 1970s an increasing number of Supreme and Constitutional Courts have over time developed ‘safeguard clauses’ designed to protect the core values of the national constitutional identity from all sort of interference from European law. A similar concern brings some national courts to mark the boundaries that cannot be trespassed by European law, maintaining the control of the competences that pertain to the national constitutional order. Suffice to recall here the Solange I and II, Maastricht, Lissabon and the following line of decisions in the German Bundesverfassungsgericht. A similar position was maintained by the Italian Constitutional Court with the counter-limits doctrine, an expression that refers to the limits to the limitation of national sovereignty accepted to enter the European Communities. Moreover, at the time of the Constitutional Treaty, a number of other national Constitutional Courts have endorsed similar doctrines, such as the French, Spanish and some of the Central and Eastern European ones. • Challenge: occasionally some national courts have challenged straightforwardly a piece of European legislation or a decision of the European Court on ultra vires grounds or for contradiction with basic principles of the national constitutional order. At times the challenge has been brought before the European Court, as the Austrian Constitutional Tribunal and Irish Supreme Court have recently done in relation to the Data Retention Directive;83 at other times divergences have resulted in an unsettled and muted conflict. • Participation: generally speaking, lower national courts have promptly taken on the ­European mandate from the very beginning, whereas higher and Constitutional Courts were rather cautious of participating in the European adventure. In recent years an increasing number of Supreme and Constitutional Courts have contributed to the development of common

83 CJEU 08.04.2014 Joined Cases C-293/12 and C-594/12 (Digital Rights Ireland and Seitlinger and Others) ECLI:EU:C:2014:238.

Italy  379 legal principles, taking an active role on the European stage, interpreting and enforcing European standards and especially making use of the preliminary ruling. Suffice here to recall the Spanish preliminary ruling in the Melloni case on the right to defence in the European Arrest Warrant context;84 the French Conseil constitutionnel reference in the Jeremy case, also concerning the European Arrest Warrant and the rights to defence;85 the Italian Constitutional Court preliminary references concerning the interpretation of Directive 1999/70/EC on fixed-term employment and its applicability to public schools teachers and staff,86 the different (European and national) standard of protection of the p ­ rinciple of legality in criminal matters,87 and, more recently, the scope of the due process of law in administrative penalties issued by supervisory authorities88 and of social rights related to maternity allowances;89 and, the decision of the Austrian Constitutional Court challenging Directive 2006/24 on data retention.90 It is worth noting that each of these stances reflects and promotes different understandings of the European public sphere. Some of them insist on boundaries, limits and divides and therefore tend to advance a context of constitutional fragmentation. In many European countries this might be a by-product of the traditional dualist approach to international law that is based on a sort of ‘separate but equal’ principle. Alternatively, and sometimes in consequence of a monist approach to legal integration, other judicial doctrines foster uniformity, implying a sort of top-down relationship between European and national constitutional law and between European and national courts. Overtaking the old dichotomy between dualist and monist views of European integration, a new pluralist constitutional approach can be promoted, where harmonisation does not overlook diversity, standardisation does not disregard disparities, and generality does not ignore singularity.91 Constitutional pluralism seems to better correspond to the self-understanding of Europe itself, whose identity is defined by ‘unity in diversity’ and is founded on a ‘subsidiarity framework’ that requires a core of common legal principles surrounded by a margin of appreciation wide enough to allow national constitutional cultures to survive. Against this backdrop, in recent times, the Italian Constitutional Court is significantly contributing to a dialogue with the CJEU, fostering constitutional pluralism building on two related strategies. The first one is stressing the importance of the principle of sincere cooperation among national and European judges in order to face the increasing overlapping between judicial remedies in the multilevel protection of fundamental rights.92 In this respect, in some recent cases where domestic provisions were inconsistent with both the Italian Constitution and the CFR, the Court acknowledged the genuinely ‘constitutional’ nature of the EU legal order, especially after the entry into force of the CFR,93 respected the power of ordinary judges to 84 The case was about the interpretation of the European regulation of the arrest warrant in relation to criminal proceedings in absentia (CJEU 26.02.2013 C-399/11 (Melloni) ECLI:EU:C:2013:107). 85 FRCC 04.04.2013, 2013-314P QPC (Jeremy F) ECLI:FR:CC:2013:2013.314P.QPC. 86 ITCC 18.07.2013, Ordinanza 207/2013, ECLI:IT:COST:2013:207 (Mascolo). 87 ITCC 23.11.2016, Ordinanza 24/2017, ECLI:IT:COST:2017:24 (Taricco). 88 ITCC 06.03.2019, Ordinanza 117/2019, ECLI:IT:COST:2019:117. 89 ITCC, 08.07.2020, Ordinanza 182/2020, ECLI:IT:COST:2020:182 90 VfGH 28.11.2012, VfSlG 19702 (Vorratsdatenspeicherung) ECLI:AT:VFGH:2012:G47.2012. Eventually the case was decided in CJEU C-293/12 and C-594/12 (n 83). 91 For a deeper exploration see Marta Cartabia, ‘Europe Today: Bridges and Walls’, (2018) 16 International Journal of Constitutional Law 741. 92 ITCC 269/2017 (n 60), and ITCC 23.11.2016, Ordinanza 24/2017, ECLI:IT:COST:2017:24 (Taricco). 93 ITCC 269/2017 (n 60), para 5.2 Considerato in diritto.

380  Marta Cartabia and Barbara Guastaferro refer a preliminary ruling to the CJEU,94 and to eventually not applying national legislation inconsistent with the Charter,95 but at the same time emphasised its power to scrutinise national legislation in light of EU law. The second (related) strategy was that of using more frequently the preliminary ruling, which is one of the more powerful procedural connectors among courts serving the cause of constitutional pluralism. In many cases, indeed, the Italian Constitutional Court used this instrument to provide a ‘domestic’ view on sensitive issues going straight to the core of the constitutional principles, and to solicit the CJEU to interpret the Charter in harmony with the common constitutional traditions of the Member States, as envisaged by the very same Article 52(4) CFR.96 This is, indeed, a quite recent novelty in the Italian Constitutional Court. Notoriously, for many decades, the Court avoided taking a position on the preliminary ruling and simply ignored its possible use. In its Judgment 536/1995, however, the Court gave the theoretical reasons for its reluctance to employ the preliminary ruling procedure: as Article 267 TFEU fails to refer to Constitutional Courts within the category of the ‘national jurisdiction’, they considered their function was only to control the constitutionality of legislation.97 As a consequence, and despite a previous judgment98 leaving open the possibility of referring a question to the CJEU, from Judgment 536/1995 onwards the Italian Court excluded for many years any possibility of submitting preliminary rulings to the CJEU. The argument supporting this choice was and is very debatable. In fact, some years later the situation evolved yet again and in 2008 the Court sent its first preliminary ruling to the CJEU. Judgment 103/200899 overruled the previous statements of the Constitutional Court regarding the European preliminary ruling. However, the scope of this judgment should not be overstated. It was a minor case concerning a regional measure on tax law allegedly in conflict with the freedom of establishment and freedom to provide services and the Court limited its referral to direct proceedings, stressing the different structure that characterises the indirect form of constitutional review of legislation. The Court was also careful to prevent any extension of this reversal to indirect procedures where the main actors of the judicial dialogue remain the lower courts. In July 2013100 the Court sent a further preliminary ruling, concerning the interpretation of Directive 1999/70/EC101 on fixed-term employment and its applicability to teachers and other personnel working in public schools. Unlike the previous one, this preliminary ruling was sent by the Italian Constitutional Court in the context of an incidental procedure. The Court thus took an important step. It is worth remarking that a similar question was already pending before the European Court, sent by a lower Italian court. The Constitutional Court was fully aware of it but decided to make its own request, bringing its own arguments of a constitutional nature to the bench of judges in Luxembourg.102 In 2017, and once again in the context of an incidental procedure, the Italian Constitutional Court referred a preliminary ruling on a sensitive issue concerning the conflict between 94 ITCC 20/2019 (n 61), para 2.3 Considerato in diritto. 95 ITCC 63/2019 (n 62), para 4.3 Considerato in diritto. 96 ITCC 20/2019 (n 61), para 2.3 Considerato in diritto; and ITCC 269/2017 (n 60), para 5.2 Considerato in diritto. 97 ITCC 15.12.1995, Ordinanza 536/1995, ECLI:IT:COST:1995:536. 98 ITCC 08.04.1991, Sentenza 168/1991, ECLI:IT:COST:1991:168. 99 ITCC 13.02.2008, Ordinanza 103/2008, ECLI:IT:COST:2008:103. 100 ITCC Ordinanza 207/2013 (Mascolo) (n 86). 101 Directive 1999/70/EC of 28 June 1999 concerning the framework agreement on fixed-term work concluded by ETUC, UNICE and CEEP, [1999] OJ L 175/43. 102 On the capacity of the CJEU to accommodate the constitutional concerns of the ITCC, see Barbara Guastaferro, ‘The unexpectedly talkative “dumb son”: the Italian Constitutional Court’s dialogue with the European Court of Justice in protecting temporary workers’ rights in the public education sector’, (2017) 13 European Constitutional Law Review 493.

Italy  381 the EU Treaty obligation upon Member States to protect the financial interests of the EU and the principle of legality in criminal matters enshrined in the Italian Constitution.103 The latter being a fundamental principle belonging to the axiological core of the Constitution, the Italian Constitutional Court warned the Court of Justice of the possibility to resort to the controlimiti doctrine, thus possibly declaring non-applicable a previous ruling of the same Court of Justice on the infringement of the principle of legality. Nevertheless, in the same referral order, it offered the Luxembourg Court the possibility to revisit its decision and to promote an interpretation of EU law consistent with the principle of legality so as to avoid the constitutional conflict (which is eventually what happened). More recently, in March 2019, the Italian Constitutional Court referred a preliminary ruling concerning a national provision authorising the Supervisory Authority on stock exchange (CONSOB) to sanction an individual’s refusal to cooperate in providing information in the context of a hearing with administrative penalties having a ‘punitive nature’ according to the so-called Engel criteria. The national provision, indeed, is possibly infringing upon some pivotal principles of due process of law, including the right of defence, the right to silence, and, more specifically, the prohibition of self-incrimination (nemo tenetur se ipsum accusare), enshrined in both Italian constitutional law and in several European and international treaties, including the CFR. Nevertheless, since the national provision was introduced into domestic law to comply with an EU directive requiring the Member States to sanction the failure to cooperate in investigations, the Italian Constitutional Court, rather than ruling upon the issue on solely domestic grounds, asked the CJEU, among other things, if such an EU law requirement is consistent with Articles 47 and 48 CFR, devoted to the right to a fair trial and to the right of defence,104 thus confirming its strongly cooperative attitude. It might be interesting to highlight that in this case, the Italian Constitutional Court has referred to the CJEU a preliminary ruling both for the interpretation and the validity of the relevant European directive: as a matter of fact, the suspect violation of the prohibition of self-incrimination has been charged both to the European directive and to the Italian legislation that implemented it. Lately, also in 2020 the Italian Constitutional Court, solicited by the Court of Cassation, referred a preliminary question related to the multilevel protection of social rights (in particular birth and maternity allowances).105 The question focused on Italian legislation requiring foreigners to necessarily hold the permit for EU long-term residents in order to benefit of those allowances and checks, and on its possible inconsistency with Article 34 of the CFR (which provides for the right to social security benefits), and with Article 12, paragraph 1, letter e), of Directive 2011/98/EU (on equal treatment between third-country nationals and nationals of the Member States.) In line with its previous cooperative attitude, the Italian Constitutional Court seeks a dialogue with the Court of Justice despite the Italian legislation, in possibly fostering an unjustified discrimination against foreigners who legally reside in Italy, violated also domestic constitutional provisions (Articles 3 and 31 of the Constitution), along with the provisions of the CFR. The path to a direct dialogue has definitely been cleared between the Constitutional Court and the Court of Justice. This is to the benefit of the coherence of the legal orders, the correct interpretation of European law, constitutional pluralism and, most importantly, citizens’ rights. This evolution in the constitutional jurisprudence concerning the preliminary ruling is indicative of the predominant attitude of the Italian judiciary towards the EU. Generally speaking,



103 ITCC

104 ITCC 105 ITCC

23.11.2016, Ordinanza 24/2017, ECLI:IT:COST:2017:24 (Taricco). 06.03.2019, Ordinanza 117/2019, ECLI:IT:COST:2019:117. Ordinanza 182/2020 (n 89).

382  Marta Cartabia and Barbara Guastaferro judges are open, cooperative and participatory in the European construction. Italian judges are increasingly trained to use all the powers with which the European treaties entrust them. A side effect is that the overall role of the judiciary has evolved dramatically over the last decades and a major cause of this is precisely Italian membership of the EU. Consider that one of the basic constitutional principles of the judiciary is ‘Judges are subject only to the law’ (Article 101). This core principle reveals the understanding of the judicial function endorsed by the Italian Constitution. It echoes the idea of the ‘juge bouche de la loi’, ie the idea of the judicial function as a neutral power whose mission is to enforce the legislation adopted by the Parliament with virtually no discretionary power. In such a constitutional context, the power conferred to national courts in order to implement European legislation, according to the principles of supremacy and direct effect, brought about a veritable revolution. Judges that used to be subject to the law can now question it and even set it aside, if necessary for the implementation of European legislation. Furthermore, the new understanding of the judicial powers implied by the EU legal order did not remain within the field of EU law but spread well beyond it. In the last few decades the interpretative powers of the judicial bodies have grown in creativity thanks to a number of methods of interpretation: according to the Constitution, according to the European legislation, according to the European Convention on Human Rights, and by use of foreign law. Judge-made law is now a reality in a country like Italy that can be categorised as following the continental tradition based on written parliamentary legislation. Consequently, the judiciary has increased its relevance in the public life of the Italian society. It is not a ‘non-power’, as it used to be considered, but it has become one of the most relevant actors of the constitutional system.

References G Amato, Le istituzioni della democrazia. Un viaggio lungo cinquant’anni (Bologna, Il Mulino, 2015). A Barbera and C Fusaro, Corso di diritto pubblico, 7th edn (Bologna, Il Mulino, 2012). P Barile, ‘Il cammino comunitario della Corte’, (1973) 18 (1) Giurisprudenza costituzionale, 2406. V Barsotti, P Carozza, M Cartabia and A Simoncini (eds), Italian Constitutional Justice in Global Context (Oxford, OUP, 2015). S Bartole, Interpretazioni e trasformazioni della Costituzione repubblicana (Bologna, Il Mulino, 2004). L Besselink, M Claes, Š Imamović, JH Reestman, National Constitutional Avenues for further European Integration (Brussels, European Parliament, 2014). R Bifulco, ‘A New Senate? A First Look to the Draft Constitutional Bill’, (2014) 6 Italian Journal of Public Law 46. R Bin and G Pitruzzella, Diritto costituzionale, 14th edn (Turin, Giappichelli, 2013). P Calamandrei, Questa nostra Costituzione, Introduction by A Garrone (Milan, Giuffrè, 1955). P Calamandrei, ‘La Costituzione e le leggi per attuarla’, in P Calamandrei (ed), Opere giuridiche vol III (Naples, Morano, 1965) 553. P Calamandrei, ‘Cenni introduttivi sulla Costituente e i suoi lavori’, in P Calamandrei (ed), Scritti e discorsi politici vol II (Florence, La Nuova Italia, 1966) 421. D Capuano, ‘The role of national parliaments in the legislative process of the EU: a view from inside the Italian Parliament’, in M Cartabia, N Lupo, A Simoncini, (eds), Democracy and Subsidiarity in the EU (Bologna, Il Mulino, 2013) 247. D Capuano, ‘The definition of the parliamentary position on EU affairs in the Senate’, in N Lupo and G Piccirilli (eds), The Italian Parliament in the European Union (Oxford, Hart, 2017) 119. P Caretti and U de Siervo, Istituzioni di diritto pubblico, 10th edn (Turin, Giappichelli, 2010). M Cartabia, Principi inviolabili e integrazione europea (Milan, Giuffrè, 1995).

Italy  383 M Cartabia, ‘National Constitutional Adjudication in the European Space’, in V Barsotti, P Carozza, M Cartabia and A Simoncini (eds), Italian Constitutional Justice in Global Context (Oxford, OUP, 2015) 205. M Cartabia, ‘Europe Today: Bridges and Walls’, (2018) 16 International Journal of Constitutional Law 741. M Cartabia, N Lupo and A Simoncini (eds), Democracy and Subsidiarity in the EU (Bologna, Il Mulino, 2012). M Cartabia and JHH Weiler, L’Italia in Europa. Profili istituzionali e costituzionali (Bologna, Il Mulino, 2000). P Casalena, C Fasone, N Lupo, ‘Commentary on the Protocol No. 1 annexed to the Treaty of Lisbon’, in H Blanke and S Mangiameli (eds), Commentary on the Treaty of Lisbon (Berlin/Heidelberg, Springer, 2013). R Chieppa, ‘Nuove prospettive per il controllo di compatibilità comunitaria da parte della Corte costituzionale’, (2007) 12 Il Diritto dell’Unione Europea 493. M Claes, ‘Le “clausole europee” nelle costituzioni nazionali’, (2005) 2 Quaderni costituzionali, 283. P Costanzo, A Ruggeri, L Trucco (eds), ‘Elezioni politiche 2018: come funziona il Rosatellum bis. Il sistema elettorale spiegato dagli esperti’, Giurisprudenza italiana. Gli speciali 2018. M de Visser, Constitutional Review in Europe. A Comparative Analysis (Oxford, Hart, 2015). G della Cananea, ‘The End of (Symmetric) Bicameralism or a Novus Ordo?’, (2014) 6 Italian Journal of Public Law 1. M Degni and P De Ioanna, Il Vincolo stupido (Rome, Castelvecchi, 2012). A Esposito, ‘The Definition of the parliamentary position on EU affairs in the Chamber of Deputies’, in N Lupo and G Piccirilli (eds), The Italian Parliament in the European Union (Oxford, Hart, 2017) 99. European Commission, Annual Report 2016 on proportionality and subsidiarity, COM (2017) 600 final, Brussels, 30.6.2017. C Fasone, ‘Il Senato approva il primo parere contrario sul rispetto del principio di sussidiarietà’, (2010) 4 Quaderni Costituzionali 824. C Fasone, The reaction of Constitutional Courts towards the Euro-crisis law. The protection of social rights in Italy, Portugal and Spain, EUI Workingpaper MWP 2014/25 (Florence, EUI, 2014). C Fasone and D Fromage, ‘Are Parliaments (More) Committed to the Trend of Open Government After the Lisbon Treaty? The Italian and the Spanish Case’, in I Bouhadana, W Gilles, I Nguyên-Duy and R Weaver (eds), Parliaments in the Open Government Era (Paris, Les éditions IMODEV, 2016) 173. J Ferejohn and P Pasquino, ‘Constitutional Adjudication, Italian Style’, in T Ginsburg (ed), Comparative Constitutional Design (Cambridge, Cambridge University Press, 2012) 294. M Fioravanti, Costituzionalismo. Percorsi della storia e tendenze attuali (Rome/Bari, Laterza, 2009). S Gambino and W Nocito, ‘Crisi dello Stato, governo dell’economia e diritti fondamentali: note costituzionali alla luce della crisi finanziaria in atto’, (2012) 162 Astrid Rassegna. T Groppi and A Simoncini, Introduzione allo studio del diritto pubblico e delle sue fonti, 2nd edn (Turin, Giappichelli, 2013). G Guarino, Cittadini europei e crisi dell’Euro (Naples, Editoriale Scientifica, 2014). B Guastaferro, ‘Beyond the Exceptionalism of Constitutional Conflicts. The Ordinary Functions of the Identity Clause’, (2012) 31 Yearbook of European Law 263. B Guastaferro, Legalità sovranazionale e legalità costituzionale. Tensioni constitutive e giunture ordinamentali (Turin, Giappichelli, 2013). B Guastaferro, ‘Coupling National Identity with Subsidiarity Concerns in National Parliaments’ Reasoned Opinions’, (2014) 21 Maastricht Journal of European and Comparative Law 320. B Guastaferro, ‘The unexpectedly talkative “dumb son”: the Italian Constitutional Court’s dialogue with the European Court of Justice in protecting temporary workers’ rights in the public education sector’, (2017) 13 European Constitutional Law Review 493. R Hirschl, Towards Juristocracy. The Origins and Consequences of the New Constitutionalism (Cambridge/ MA, Harvard University Press, 2007). N Krisch, ‘The Backlash against International Courts’, Verfassungsblog 16.12.2014, www.verfassungsblog.de. E Lamarque, Corte Costituzionale e giudici nell’Italia repubblicana (Rome/Bari, Laterza, 2012).

384  Marta Cartabia and Barbara Guastaferro E Lamarque, ‘Some WH Questions About the Italian Constitutional Court’s Judgement on the Rights of the Victims of the Nazi Crimes’, (2014) 6 Italian Journal of Public Law 197. M Luciani, ‘Antifascismo e nascita della costituzione’, (1991) 22 Politica del diritto 183. M Luciani, ‘Dottrina del moto delle costituzioni e vicende della costituzione repubblicana’, (2013) 1 Rivista AIC, 1. M Luciani, ‘Costituzione, bilancio, diritti e doveri dei cittadini’, (2013) 174 Astrid Rassegna. N Lupo and G Rivosecchi, ‘Quando l’equilibrio di bilancio prevale sulle politiche sanitarie regionali’, (2012) 5/6 Le Regioni, 1062. S Mannoni, ‘The “Second Chamber”: a Historical and Comparative Sketch’, (2014) 6 Italian Journal of Public Law 8. A Manzella, ‘The Role of Parliaments in the Democratic Life of the Union’, in S Micossi and GL Tosato (eds), The European Union in the 21st Century. Perspectives from the Lisbon Treaty (Brussels, CEPS, 2010) 257. G Martinico, ‘Constitutionalism, Resistance and Openness: Comparative Law Reflections on Constitutionalism in Postnational Governance’, (2016) 35 Yearbook of European Law 318. G Martinico, B Guastaferro, and O Pollicino ‘The Constitution of Italy: Axiological Continuity between the Domestic and International Levels of Governance?’, in A Albi and S Bardutzky (eds), The Role of National Constitutions in the European and Global Governance (The Hague, TMC Asser Press, 2019). G Martinico and O Pollicino (eds), The Interaction between Europe’s Legal Systems. Judicial Dialogue and the Creation of Supranational Laws (London, Edward Elgar, 2012). FX Millet, ‘How much lenience for how much cooperation? On preliminary reference of the French Constitutional Council to the Court of Justice’, (2014) 51 CMLRev 195. B Mirkine-Guetzévitch, Les nouvelles tendances du droit constitutionnel (Paris, Marcel Giard, 1931) 48. D Mone, ‘La costituzionalizzazione del pareggio di bilancio ed il potenziale vulnus alla teoria dei controlimiti’, (2014) 3 Rivista AIC, 1. C Mortati, Lezioni sulle forme di governo (Padua, CEDAM, 1973). V Onida, ‘L’attuazione della Costituzione tra magiustratura e Corte costituzionale’, in Aspetti e tendenze del diritto costituzionale. Scritti in onore di C. Mortati, IV (Milan, Giuffrè, 1977). F Paterniti, ‘La riforma dell’art. 117, 1° co. della Costituzione e le nuove prospettive dei rapporti tra ordinamento giuridico nazionale e Unione Europea’, (2004) 49 Giurisprudenza Costituzionale 2101. A Peters, ‘Let Not Triepel Triumph – How To Make the Best Out of Sentenza No 238 of the Italian Constitutional Court for a Global Legal Order’, EJIL:Talk 22.12.2014, www.ejiltalk.org. L Pierdominici, Constitutional Change through Euro Crisis Law. Report on Italy (Florence, EUI, 2014). C Pinelli, ‘I limiti generali alla potestà legislativa statale e regionale e i rapporti con l’ordinamento comunitario’, (2001) 126 Il Foro italiano 194. E Posner and A Vermeule, ‘Tyrannophobia’, in T Ginsburg (ed), Comparative Constitutional Design (Cambridge, Cambridge University Press, 2012) 317. G Sorrenti, L’interpretazione conforme a Costituzione (Milan, Giuffrè, 2006). F Sorrentino, ‘Regioni, diritto internazionale e diritto comunitario. Nuovi profili costituzionali dei rapporti tra diritto interno e diritto internazionale e comunitario’, paper presented at the conference Regioni, diritto internazionale e diritto comunitario, Genoa 23.03.2002. S Staiano, ‘Le leggi monocamerali (o più esattamente bicamerali asimmetriche)’, (2016) 1 Rivista AIC, 1. C Tomuschat, ‘The National Constitutional Court Trumps International Law’, (2014) 6 Italian Journal of Public Law 189. L Torchia, ‘I vincoli derivanti dall’ordinamento comunitario nel nuovo Titolo V della Costituzione’, (2001) 29 Le Regioni 1203. GE Vigevani, ‘The Reform of Italian Bicameralism: the First Step’, (2014) 6 Italian Journal of Public Law 55. L Violini, ‘The Reform of Italian Bicameralism: Current Issues’, (2014) 6 Italian Journal of Public Law 33.

13 Cyprus A Territorially Divided Member State NIKOLAS KYRIAKOU AND NIKOS SKOUTARIS

I. Introduction The Republic of Cyprus (hereinafter ‘RoC’ or ‘Republic’) became independent on 16 August 1960 and acceded to the European Union (EU) on 1 May 2004 during the ‘Big-Bang Enlargement’. During those 44 years, the bi-communal RoC turned into a mono-national state in 1964, experienced a coup d’état and a military invasion in 1974 orchestrated by two of its guarantor states – Greece and Turkey respectively – and managed to become an EU Member State in 2004 without resolving the territorial segregation of the two main communities that live on the island. Because of its turbulent historical trajectory and the constitutional solutions that have been chosen to deal with the political anomalies of the island, the interrelationship between RoC’s constitutional system and the EU ‘constitutional order of States’1 is somewhat different than most of the other EU Member States. The present contribution aims exactly at analysing this interrelationship. It does so by examining the main characteristics of RoC’s constitutional system (section II); commenting on the main characteristics of its constitutional culture (section III); presenting the debate concerning its accession (section IV); analysing the constitutional limits to EU integration (section V); examining the constitutional rules on implementing EU law (section VI); and revisiting the relationship between EU law and national law (section VII).

II.  Main Characteristics of the RoC’s Constitutional System RoC gained its sovereign independence from the UK by virtue of three international treaties, namely the Treaty of Guarantee, the Treaty of Alliance and the Treaty of Establishment and a Constitution, all of which came into operation on the same day –16 August 1960.2 1 Anthony Arnull et al (eds), A Constitutional Order of States? Essays in EU Law in Honour of Alan Dashwood (Oxford, Hart Publishing, 2011). 2 See generally www.kypros.org/Constitution/English/.

386  Nikolas Kyriakou and Nikos Skoutaris The establishment of the independent RoC was seen as a political compromise between the different goals and aspirations of the two ethno-religious communities that live on the island, their motherlands and the former colonial power. In particular, Greece and the Greek Cypriots were fighting for Enosis (union with the motherland), the Turkish Cypriots and Turkey were responding by asking for a Taksim (partition), and Britain was determined to retain full sovereignty on the island. So, in order to achieve the balance between those conflicting interests, a complicated powersharing structure was designed. The Constitution was drawn up explicitly in terms of the two communities.3 Moreover, all of the principles of a consociational democracy – grand coalition, proportionality, autonomy and veto – were elaborately embodied in the 1960 Constitution. The Constitution provided for ‘an independent and sovereign Republic with a presidential regime, the President being Greek and the Vice-President4 being Turkish, elected by the Greek and the Turkish communities of Cyprus respectively’.5 The President and Vice-President exercise executive power.6 Their common powers are specifically enumerated in Article 47 of the Constitution while the two subsequent Articles provide the exclusive enumeration of their separate, almost identical, powers. According to Article 54 of the Constitution, all the executive powers not expressly reserved to the President and the Vice-President are exercised by the Council of Ministers. The cabinet had to consist of seven Greek ministers designated by the President and three Turkish ministers designated by the Vice-President. More i­ mportantly, the 1960 Constitution provided an absolute veto power to both the President and the Vice-President over decisions by the cabinet or the legislature in the fields of foreign affairs, defence and security.7 A seven-to-three ratio entailed a deliberate overrepresentation of the Turkish minority rather than strict proportionality, also affecting the composition of the legislature, which was unicameral. The House of Representatives was comprised of 35 Representatives belonging to the Greek community and 15 belonging to the Turkish one.8 Laws were passed by simple majority, but any amendment to the electoral law, the passing of laws concerning municipalities, and any law imposing taxes or duties requires a separate majority among Greek and Turkish Cypriot Representatives present and voting in accordance with Article 78(2) of the Constitution. Additionally, the amendment of any non-basic constitutional provision9 required a two-thirds majority of the representatives of each community voting separately.10 The Constitution also guaranteed a great deal of autonomy for the two ethnic segments by setting up two separately elected communal chambers with exclusive legislative powers over religious, educational, cultural, and personal status matters.11 The judicial system was to consist of a Supreme Constitutional Court,12 a High Court of Justice and lower courts.13 The Supreme Constitutional Court was comprised of a Greek Cypriot

3 According to the 1960 census, the Greek Cypriot segment comprised about 78% and the Turkish Cypriot about 18% of the population, the remaining 4% being the minorities of the Maronites, Armenians and Latins. 4 See generally Pt 3 (Arts 36–60) of the Constitution of the RoC. 5 Art 1 of the Constitution of the RoC. 6 Art 46 of the Constitution of the RoC. 7 Art 50 of the Constitution of the RoC. 8 See generally Pt 4 (Arts 61–85) of the Constitution of the RoC. 9 See section V.A. of the present chapter. 10 Art 182(3) of the Constitution of the RoC. 11 See generally Pt 5 (Arts 86–111) of the Constitution of the RoC. 12 See generally Pt 9 (Arts 133–51) of the Constitution of the RoC. 13 See generally Pt 10 (Arts 152–64) of the Constitution of the RoC.

Cyprus  387 judge and a Turkish Cypriot judge and was presided over by a neutral judge who was neither a Cypriot citizen nor a citizen of any of the guarantor states. Its jurisdiction ranged from ­constitutional issues arising from the interpretation of provisions of the Constitution14 to the settling of conflicts or disputes regarding the extent of authority of legislative and administrative bodies.15 The High Court of Justice, which consisted of two Greek Cypriot judges, one Turkish Cypriot judge and one foreign presiding judge, was the appellate court of civil and criminal jurisdiction. The composition of lower courts depended on the community of the disputants.16 In addition to that, several other constitutional provisions were designed to safeguard the bi-communal nature of the state. For example, Article 173 of the Constitution provided for the establishment of separate municipal councils in the five largest towns of the island.17 At the same time, while the public service had to be composed in accordance with the aforementioned sevento-three ratio,18 a six-to-four ratio was set for the army and the police.19 All those provisions and similar ones relied on the cooperation of the two communities but did little to encourage it. Under those circumstances, and given that the cooperation of the two communities was a prerequisite for the smooth functioning of the RoC, it was inevitable that the internal stability of the new state would soon be at stake. Indeed in December 1963, the first, low-scale, intercommunal armed conflict broke out in Nicosia. That led to the break-up and the ‘hellenisation’ of the, until then, bi-communal RoC in 1964. Since that moment the Turkish Cypriots have consistently rejected participation in the administration of the common state. Notwithstanding, RoC continued functioning mainly by invoking the ‘doctrine of necessity’. This doctrine is considered a constitutional principle, which indirectly forms part of the 1960 Constitution. Its aim is to solve problems that were not foreseen by the drafters and which threaten the existence of the RoC. The doctrine was spelled out for the first time in the emblematic Mustafa Ibrahim judgment of the Supreme Court.20 In the aftermath of the resignation of Professor Forsthoff, the President of the Supreme Constitutional Court, the House of Representatives enacted the Administration of Justice (Miscellaneous Provisions) Law, 33/1964. According to this law, a newly established Supreme Court would exercise the jurisdictions and powers both of the Supreme Constitutional Court and the High Court ‘until such time as the people of Cyprus may determine such matters’.21 The allegation was that such law, which was merging two Courts into one Supreme Court, was not enacted in accordance with the Constitution. The Court held that the doctrine of necessity should be considered to be implicit in the provisions of a strict and written constitution, and is, therefore, part of the constitutional order in Cyprus. It allows the country to safeguard its interests whenever the Constitution, due to its rigidity, one-sidedness and narrow ambit, contains no provisions giving satisfactory solutions to extraordinary situations ‘of a public necessity of the first magnitude’.22 Most importantly, the

14 Art 149 of the Constitution of the RoC. 15 Art 139 of the Constitution of the RoC. 16 Art 159 of the Constitution of the RoC. 17 Diana Markides, Cyprus 1957–63: From Colonial Conflict to Constitutional Crisis. The Role of the Municipal Issue (Minneapolis MN, University of Minnesota Press, 2001). 18 Art 123 of the Constitution of the RoC. 19 Arts 129–30 of the Constitution of the RoC. 20 Attorney General of the Republic v Mustafa Ibrahim [1964] Cyprus Law Reports [hereinafter CLR] 195. 21 Ibid, 201 and 225. 22 Ibid, 234.

388  Nikolas Kyriakou and Nikos Skoutaris Court decided that there are four prerequisites in order to determine whether the said doctrine could be applied in a particular case: 1. 2. 3. 4.

There is an imperative and inevitable necessity or exceptional circumstance. There is no other remedy. The measure taken must be proportionate to the necessity. The measure must be of a temporary character limited to the duration of the exceptional circumstances.23

The doctrine of necessity, as defined in the Mustafa Ibrahim case, has not only provided the necessary legal basis in order for the RoC to deal with the absence of the Turkish Cypriots in the Government and their subsequent substitution with Greek Cypriots,24 but also has allowed the amendment of basic and non-basic Articles of the Constitution, as we will see in section V of the present chapter.25 Most importantly, one has to note that notwithstanding the break-up and the ‘hellenisation’ of the RoC, on 4 March 1964, the UN Security Council confirmed, albeit impliedly, the RoC’s statehood by recognising the legitimacy of the Government of the RoC which was, at the time, comprised only of Greek Cypriots with the unanimous adoption of Resolution 186 (1964).

III.  Constitutional Culture Unsurprisingly, the dramatic political tensions and conflicts that have marked the history of the third largest island of the Mediterranean have also influenced its constitutional culture. In particular, we would like to highlight two important issues. First, the Constitution, as initially designed in the Cyprus Agreements, is a typical example of a consociational arrangement that treats the two main ethno-religious segments as co-rulers of the island. This bi-communalism is evident in the fact that both the Greek-Cypriot President and the Turkish-Cypriot Vice-President had vetoes, in the strict ethnic ratios for the executive and legislature, and in the composition of the judiciary and the administration. This effort, however, to build, a culture of consensual constitutional politics in a post-conflict era was rather shortlived. After the break-up of the Government in 1964 and definitely after the Turkish invasion in 1974 that consolidated the territorial segregation of the two communities, the administration of RoC has been ‘hellenised’. In that sense, since 1964, RoC has been a mono-national state. More importantly, as we mentioned in the previous section, the fragility of the post-conflict arrangement urged the Constitution-makers to provide for a very rigid amendment procedure of the non-basic constitutional provisions.26 According to Article 182(3) of the Constitution, any amendment of such provision required a two-thirds majority of the representatives of each community voting separately. However, following the break-up of the Government of the RoC in 1964 and the adoption of the doctrine of necessity, the Cypriot Parliament amended the Constitution in 15 instances without the consent of the Turkish Cypriot representatives, who

23 Ibid, 265. 24 For a more detailed account see generally Achilleas Emilianides, ‘Accession of the Republic of Cyprus to the EU, the Constitution and the Cypriot Doctrine of Necessity’, (2007) The Cyprus Yearbook of International Relations 65; Nikolas Kyriakou, ‘Report on Cyprus’ in G Martinico and O Pollicino (eds), The National Judicial Treatment of the ECHR and EU Laws: A Constitutional Comparative Perspective (Groningen, Europa Law Publishing, 2010) 151. 25 Nicolaou v Nicolaou [1992] 1 CLR 1338. 26 See section V.A. of the present chapter.

Cyprus  389 have not participated in RoC’s political and constitutional life since the mid 1960s. To the extent that the Parliament has been able to amend the Constitution via the doctrine of necessity in the post-1964 era, it has become a permanent Constitution-maker.

IV.  Constitutional Foundations of EU Membership In 1990, the internationally recognised Government of the RoC presented an application for membership to the European Community in accordance with (then) Article 237 TEEC. This ­application was not legally based on any specific constitutional provision, given that the Constitution did not explicitly envisage the Republic’s membership of the EU. Interestingly enough, however, the regime in northern Cyprus challenged the RoC’s ­application mainly on the ground that the Cypriot Government could not represent the whole of Cyprus and that the application was contrary to international and constitutional law. In a joint declaration, Turkey and the breakaway state of the Turkish Republic of Northern Cyprus (hereinafter ‘TRNC’) declared that Cyprus could not join ‘international political and economic unions to which Turkey and Greece are not members’.27 This led to an interesting legal debate concerning the legality of the Cypriot application.28 On the one hand, Turkey-hired Professor Mendelson published a legal opinion in June 1997, according to which the future EU accession of Cyprus would be illegal.29 Professors Crawford, Hafner, and Pellet, commissioned by RoC, rebutted this opinion.30 The debate concerned, inter alia, the interpretation of Article I(2) of the Treaty of Guarantee. According to it, The Republic of Cyprus undertakes not to participate, in whole or in part, in any political or economic union with any State whatsoever. It accordingly declares prohibited any activity likely to promote, directly or indirectly, either union with any other State or partition of the island.

The question was whether the aforementioned Article prevents RoC’s EU membership as it could amount to an economic and political union with 23 (now 27) other states, and in particular with Greece.31 Notwithstanding the legal merits of the debate, it suffices to note for the purposes of the current chapter, that the legal objections of Turkey and the internationally unrecognised TRNC did not stop RoC’s accession to the EU. In fact, on 1 May 2004, a week after the Greek Cypriots rejected the UN’s ‘Plan for The Comprehensive Settlement of the Cyprus Problem’, the RoC became an EU Member State. The terms of the Republic’s accession are described inter alia in Protocol No 10 on Cyprus of the Act of Accession 2003.32 According to Article 1(1) of this Protocol, the application of EU law is suspended in northern Cyprus – an area where RoC’s internationally recognised Government

27 Joint Declaration of the Republic of Turkey and the ‘Turkish Republic of Northern Cyprus’ of 28 December 1995, para 5. 28 For a comprehensive analysis of the debate see Nikos Skoutaris, The Cyprus Issue. The Four Freedoms in a Member State under Siege (Oxford, Hart Publishing, 2011) 32–38. 29 Maurice Mendelson, The Application of the ‘Republic of Cyprus’ to Join the European Union, Opinion of 6 June 1997 (reprinted in Maurice Mendelson, Why Cyprus’ Entry into the European Union Would be Illegal. Legal Opinion, by Professor M. H. Mendelson QC (London, Embassy of the Republic of Turkey, 2001) 33). 30 James Crawford, Gerhard Hafner and Alain Pellet, ‘Republic of Cyprus: Eligibility for EU Membership, Opinion of 24 September 1997’ (reprinted in A Markides (ed), Cyprus and European Union Membership. Important Legal Documents (Nicosia, 2002) 13). 31 Mendelson, Opinion of 6 June 1997 (n 28) 36. 32 For a comprehensive analysis of Protocol No 10 see Skoutaris, Cyprus Issue (n 28) 44–48.

390  Nikolas Kyriakou and Nikos Skoutaris does not exercise effective control. Such suspension can be lifted if the Council decides so unanimously, on the basis of a proposal from the Commission.33 Until such withdrawal of the suspension takes place, Article 2 of the Protocol allows the Council to decide unanimously how EU law would apply to the ‘Green Line’ ie the ‘border’ between the Government-controlled areas and northern Cyprus. Indeed, a week before RoC became officially an EU Member State, the Council passed the ‘Green Line Regulation’ which regulates how people and goods can cross this de facto border.34 More importantly, in the event of a settlement, Article 4 of the Protocol allows the Council of the EU to decide unanimously on the adaptation concerning the Turkish Cypriot community. Indeed, if the April 2004 referendum had approved the new state of affairs envisaged in the Annan Plan, the Council of the European Union, having regard to that Article, would have unanimously adopted the Draft Act of Adaptation of the Terms of Accession of the United Cyprus Republic to the European Union as a Regulation.35

V.  Constitutional Limits to EU Integration A.  Limits to EU Integration The Constitution of the RoC does not contain any express limitations to EU integration or to the delegation of competences to international organisations.36 At the time of its establishment, in 1960, participation in international organisations was associated with the classic form of participation in regional and universal international organisations. From the early years of the Republic’s life, it became a full member of the United Nations, the Council of Europe and also acceded to a number of multilateral conventions and treaties. Thus, the keystone characteristic of the Constitution is that the idea of national constitutionalism as a guarantee against the possible concentrations of power from European constitutionalism is absent.37 In 2004, at the time of accession to the EU, it was considered that no amendment to the Constitution would be necessary in order to give precedence to the application of EU law, as in the case of Ireland.38 The change to and challenge for the Cypriot legal system was obvious. Cyprus was entering ‘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’.39 On the basis of the legal understanding that no amendment to the Constitution was necessary, the House of Representatives promulgated Law 35(III)/2003, which essentially ratified the

33 Protocol No 10, Art 1(2). 34 Council Regulation (EC) No 866/2004 of 29 April 2004 on a regime under Art 2 of Protocol No 10 of the Act of Accession 2003 [2004] OJ 206/51. For a comprehensive analysis of the Green Line Regulation regime see Nikos Skoutaris, ‘The application of the acquis communautaire in the areas not under the effective control of the Republic of Cyprus: The Green Line Regulation’, (2008) 45 CML Rev 727. 35 Appendix D of The UN Plan for a Comprehensive Settlement of the Cyprus Problem. 36 Constantinos Kombos, ‘Report on Cyprus’ in J Czuczai et al (eds), Division of Competences and Regulatory Powers between the EU and the Member States. FIDE XXVII Congress Proceedings Vol. 3 (Budapest, Wolters Kluwer, 2016) 282. 37 Leonard Besselink et al, National constitutional avenues for further EU integration (Brussels, European Parliament, 2014) 207. 38 Nicos Emiliou, ‘Cyprus’, in A Kellerman et al (eds), The impact of EU Accession on the legal orders of new EU Member States and (pre-) candidate countries, Hopes and Fears (The Hague, TMC Asser Press, 2006) 303, where an interesting comparison of Cyprus and Ireland is made with regards to their constitutional capacity to receive EU law. 39 ECJ 05.02.1963 26/62 (Van Gend en Loos/Administratie der Belastingen) ECLI:EU:C:1963:1.

Cyprus  391 accession treaty of the 10 new Member States to the EU. The legal foundation of this ratifying law was Article 169 of the Constitution, which regulates the conclusion, ratification and entry into force of international treaties, conventions and agreements. Article 169(3) of the Constitution resolves the monism–dualism distinction in favour of the first: Treaties, conventions and agreements concluded in accordance with the foregoing provisions of this Article shall have, as from their publication in the official Gazette of the Republic, superior force to any municipal law on condition that such treaties, conventions and agreements are applied by the other party thereto.

It follows from this provision that, as an international convention, the Treaty of Accession prevails over national law in case of conflict between the two.40 However, Article 179 of the Constitution provided at the time that the Constitution ‘shall be the supreme law of the Republic’. Thus, any treaty, convention or agreement was ranked at a hierarchically lower position to the Constitution. A ‘black-letter law’-reading provided for a fine and precise hierarchical structure within the domestic legal order, giving precedence to the Constitution, followed by international instruments and then domestic legislation lying at its lower level. Clearly, this hierarchy of norms did not correspond to the doctrine of supremacy of EU law within the national legal systems and was an overt and general limit to EU integration, since it subjected the full application and effectiveness of EU law to the authority of the Constitution itself. To add to the already complicated legal situation, Article 4 of Law 35(III)/2004, having the side-title ‘Direct effect and supremacy’ provided as follows: ‘The rights and obligations that the Treaty [of accession] imposes, have direct effect in the Republic and prevail over any contrary legislative or regulative provision’. Article 4 was drafted to ensure the direct effect and supremacy of EU law within the national legal order. However, it was at least paradoxical to provide for the supremacy of EU law ‘over any contrary legislative or regulative provision’ in the text of a national law. As was explained above, such laws had a hierarchical inferior status to the Constitution and, accordingly, cannot override the Constitution. In essence, Article 4 pursued a legitimate aim with untenable means. One year after Cyprus’s accession to the EU, a judgment of the Supreme Court of Cyprus found that the Framework Decision for the European Arrest Warrant (EAW)41 did not prevail over the constitutional provision for the extradition of Cypriot citizens. This judgment prompted a constitutional amendment in order to provide for the supremacy of EU law within the domestic legal order. The doctrine of supremacy, as devised and applied by the Court of Justice of the European Union (CJEU) in cases such as Van Gend and Loos and Costa42 seemed to provide sufficiently safe ground for the proper reception and application of EU law within the domestic legal order. However, this did not prove to be the case because the Supreme Court of Cyprus indirectly challenged the judicially established hierarchy of norms. In Attorney General v Kostas Konstantinou43 the legislation transposing the Framework Decision for the EAW was challenged as unconstitutional.

40 Georghios Pikis, Constitutionalism – Human Rights – Separation of Powers (Leiden, Brill, 2006) 44; Emiliou, ‘Cyprus’ (n 38); Andreas Loizou, ‘Cyprus’, in R Blackburn and J Polakiewicz (eds), Fundamental rights in Europe: The ECHR and its Member States. 1950-2000 (Oxford, Oxford University Press, 2001) 220. 41 2002/584/JHA, Council Framework Decision of 13 June 2002 on the European arrest warrant and the surrender procedures between Member States [2002] OJ L 190/1. 42 ECJ 26/62 (n 39); ECJ 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66. 43 Supreme Court of Cyprus, 07.11.2005, Attorney General v Kostas Konstantinou, Civil Appeal No 294/2005.

392  Nikolas Kyriakou and Nikos Skoutaris The fifth amendment of the Constitution was an inevitable repercussion of this judgment. Law 127/06 amended four of the Constitution’s Articles in order to provide expressly for the precedence of EU and EC law in the domestic legal order.44 This amendment intended to settle in a definite manner the hierarchy of norms in Cyprus, by setting EU and EC law at the top of the hierarchy, followed by the Constitution and then ordinary legislation. Article 1A of the Constitution now reads: No provision of the Constitution is deemed to invalidate laws which are promulgated, acts effected or measures taken by the Republic which are rendered necessary due to its obligations as a member state of the European Union, nor does it prevent Regulations, Directives or other acts or binding measures of legislative character promulgated by the European Union or by the European Communities or by their institutions or competent bodies on the basis of the treaties founding the European Communities or the European Union from having legal force in the Republic.

This wording is strikingly similar to the counterpart provision of the Irish Constitution, which served as a blueprint for the amendment of the relevant Cypriot provision.45 In addition, two further constitutional amendments were introduced: (a) Article 169(4) of the Constitution was inserted, providing ‘The Republic may exercise every option and discretionary power provided for by the Treaties establishing the European Communities and the Treaty on the European Union and any treaties amending or substituting them, concluded by the Republic’ and (b) Article 179 of the Constitution, formerly providing that the Constitution is the supreme law of the Republic, was changed to: ‘Provided the dispositions of Article 1A are abided by, the Constitution is the supreme law of the Republic’. This set of amendments was considered necessary to unequivocally resolve the issue of supremacy of EU law. The only explicit discussion of constitutional limits to EU integration in the case law of the Supreme Court of Cyprus can be found in the dissenting opinion of Judge Erotokritou on a case concerning the legality of the bail-in measures adopted during the economic crisis in Cyprus. While this dissenting opinion raises interesting issues relating to the constitutional limits to EU integration and the constitutional identity of the RoC, it fails to flesh out in detail the full extent of the reasons for dissent. In addition, the very fact of it being a dissenting opinion accords it with little normative power. Despite the aforementioned caveats, there is merit in addressing some parts of the dissenting opinion the case Myrto Christodoulou v Central Bank of Cyprus.46 In this opinion, Judge Erotokritou admitted that national sovereignty undoubtedly gives way to the supremacy of European law … But it seems that the further erosion of that national sovereignty and the parallel erosion of fundamental rights, often through informal procedures, should at some stage be scrutinized by the CJEU, albeit indirectly through Article 267, as to whether it is compatible with the primary law of the European Union.

He also added that in the EU legal system, the supremacy of the rule of law and of the legal protection, which form a fundamental principle of the European Union and which are inextricably intertwined with the 44 Official Gazette of the Republic of Cyprus, no 4090, 2006. Arts 1, 140, 169 and 179 of the Constitution were amended. 45 Art 29(10) of the Irish Constitution provides: ‘No provision of this Constitution invalidates laws enacted, acts done or measures adopted by the State which are necessitated by the obligations of membership of the European Union or of the Communities, or prevents laws enacted, acts done or measures adopted by the European Union or by the Communities or by institutions thereof, or by bodies competent under the Treaties establishing the Communities, from having the force of law in the State’. 46 Supreme Court of Cyprus, 07.06.2013, Myrto Christodoulou v Central Bank of Cyprus, The Governor of the CBC and the Minister of Financial Affairs, Recourse no 551/2013.

Cyprus  393 Republic, cannot be eliminated through the creation of exemptions from judicial administrative control, each time for various reasons national governments are in trouble and take decisions which violate basic human rights that are derived from the legal order of the European Union and generally from the European acquis communataire. The legal restrictions imposed by the Constitution on the exercise of state power must be maintained even in critical and difficult conditions such as those that exist today, in order to ensure the supremacy of the rule of law and of the principle of legality.47

What is interesting in these excerpts is that Judge Erotokritou puts forward, albeit tacitly, two arguments that fit neatly within the context of the constitutional limits to EU integration. The first is that the transfer of competences may result in the erosion of national sovereignty, which in turn may affect the protection of fundamental rights. Judge Erotokritou is not opposed to the diminution of national sovereignty and the transfer of competences to the EU. He seems to accept that joining the EU has as a necessary corollary this alteration in the character of national sovereignty. He further seems to call for limits to be placed in case fundamental rights are affected by this erosion of sovereignty. The second argument appears to reach out to the notion of preservation of the constitutional identity. For Judge Erotokritou, the rule of law and legal protection are of paramount importance and, according to his understanding, inextricably linked with the Republic itself. Coupled with the primordial importance accorded to fundamental rights in the same dissenting opinion, one may conclude that these three elements constitute the core trinity of the constitutional identity of the Republic.

B.  EU Law and the Constitution: Starting on the Wrong Foot The issue of supremacy of EU law within Cyprus’s legal order arose in the context of a case involving the implementation of the Framework Decision for the EAW. In Konstantinou the defence challenged the national legislation transposing the EAW as unconstitutional. As defined in the text of this Framework Decision, an EAW is a judicial decision issued by a Member State with a view to the arrest and surrender by another Member State of a requested person, for the purposes of conducting a criminal prosecution or executing a custodial sentence or detention order.

The transposing legislation was found by the Supreme Court to be incompatible with Article 11(2)(f) of the Constitution. This Article prohibits the arrest or detention of a person to prevent him from unauthorised entry into the territory of the Republic or of an alien against whom action is being taken with a view to deportation or extradition. Deportation or extradition was constitutionally permissible only for aliens, and not for citizens of the Republic. The main issue in Konstantinou was the surrender of a person having dual nationality (Cypriot and British) under the terms of the legislation transposing EAW. The Supreme Court considered and acknowledged the CJEU case law on the supremacy of EU law, but based its own judgment on different premises. It found that the case before it was not related to a ‘European provision having direct effect’. The EAW was binding upon Cyprus as to the result to be achieved, but it was up to the state’s national authorities to choose the form and methods to achieve such result.48 47 The translation of the excerpts is taken from Kombos, ‘Report on Cyprus’ (n 36) 284 f. 48 Art 34(2) (b) TEU provides: ‘(b) [the Council may] adopt framework decisions for the purpose of approximation of the laws and regulations of the Member States. Framework decisions shall be binding upon the Member States as to the result to be achieved but shall leave to the national authorities the choice of form and methods. They shall not entail direct effect.’

394  Nikolas Kyriakou and Nikos Skoutaris Choosing to transpose the EAW Framework Decision by means of ordinary legislation was found to be incompatible with Article 11(2)(f) of the Constitution. Thus, the surrender of the Cypriot citizen on the basis of an EAW was not allowed. The Supreme Court could not find an interpretation of national law, which would conform to the requirements of EU law, as the CJEU had already indicated in Pupino.49 In the latter judgment the CJEU stated that: ‘It is for the national court to determine whether … a conforming interpretation of national law is possible’. Konstantinou stands as a missed opportunity for the Supreme Court to determine the relationship between EU and national law in favour of the former.50 With due respect to the Supreme Court, we consider that its approach to the matter at hand was overly legalistic and suffers from internal inconsistency.51 The Supreme Court’s analysis remained only at the level of identifying the conflict between the transposing law and the Constitution, without providing any convincing reasoning as to how it reached its decision. An option that was available to the Supreme Court was to give precedence of EU law over the conflicting national law, even if this was of constitutional rank. Had this been the case, no amendment of the Constitution would have been necessary and the matter could have been resolved through the Court’s interpretation of the relevant provisions. In hindsight, the amendment of the Constitution was not of genuine significance as the same result could have been achieved were the Supreme Court to adopt a different interpretative approach. Joseph Weiler suggested in one of his early writings that supremacy has a bi-dimensional connotation. On one side there is the CJEU’s well-known case law, which has not gone uncontested by Member States and national courts. On the other side, ‘its full reception … depends on its incorporation into the constitutional orders of the Member States and its affirmation by their supreme courts’.52 It is this second dimension that was the source of tension in the case of Cyprus. The Supreme Court’s stance was ambivalent. Although it explicitly recognised the supremacy of EU law in general, it did not inquire deeper into the legal significance of the notion in the area of the third pillar, and by consequence, in the case before it. This superficial approach led to the inconsistency mentioned above. The Supreme Court took cognisance of supremacy, but did not make any use of it thereafter. Rather, it found that it was not confronted with an instrument having direct effect. It is evident that the Supreme Court confused the notions of ‘supremacy’ and ‘direct effect’. The Supreme Court’s decision can also be explained by the circumspect reception of the EAW in courts of other EU Member States, such as France, Germany, Greece and Poland. Had the Supreme Court not aligned itself with the stance adopted by these other courts, it could have been the sole court to take the decisive step in advancing the interpretation of third pillar EU law. It is interesting to note that the Supreme Court recognised that, as a matter of p ­ rinciple, both Community law (as it was at the time) and EU law enjoyed precedence over the

49 ECJ 16.06.2005 C-105/03 (Pupino) ECLI:EU:C:2005:386. 50 For a critique of the implementation of the EAW in Cyprus see: Elias Stefanou and Antros Kapardis, ‘The first two years of fiddling around with the implementation of the European Arrest Warrant (EAW) in Cyprus’, in E Guild (ed), Constitutional challenges to the European arrest warrant (Nijmegen, Wolf Legal Publishers, 2006) 75. 51 Kombos is also critical of this judgment, considering that the Court failed to establish and clarify the limits of the relationship of the principle of supremacy of EU law with the national constitution. The same author considers the judgment to be an open-ended invitation to amend the Constitution without defining the limits of would be acceptable. See Constantinos Kombos, The impact of EU law on Cypriot Public law (Athens, Sakkoulas Publications, 2015) 80. 52 Joseph Weiler, ‘The Community System: the dual character of Supranationalism’, (1981) 1 Yearbook of European Law 267, 275 f.

Cyprus  395 Constitution. This was a far-reaching statement going even beyond the CJEU’s case law, which never explicitly acknowledged that acts under the third pillar had indeed precedence over national constitutions. In any case, as explained in the previous section, this judgment prompted the fifth amendment of the Constitution by which four articles were amended and a new one was introduced. These amendments were all-encompassing, in the sense that they were drafted so as to ensure Cyprus’s full integration in the EU’s legal architecture and to remove any potential doubts as to the supremacy of EU law. The amendments did not spark a debate on issues relating to Cyprus’s constitutional identity and to the preservation of core competences for the state. The exclusive focus was on the technicalities of the amendments and the safeguarding of human rights-related standards (ie, the extradition and surrender of Cypriots to other jurisdictions). But even this last concern on human rights standards was seen through the lens of ensuring an overall and unequivocal reception of EU law in the domestic legal order and establishing, in the clearest terms possible, its supremacy.

C.  Limits to EU Integration Outside the EU Legal Order The recent financial crisis has also tested RoC’s constitutional order and in particular its limits to further EU integration.53 The flexibility that characterises that constitutional order has meant that crisis management measures have been transposed without adopting any constitutional amendments thus far. All crisis management measures have been transposed into the national legal order by means of ordinary legislation. The great majority of the instruments were adopted under Article 169(2) of the Constitution on ‘treaties, conventions and international agreements’, which requires a simple majority in Parliament. In particular, the European Financial Stability Facility (hereinafter EFSF) was implemented in Cyprus through Law 13 (III) of 2010 with the title ‘The law on the participation of the Republic of Cyprus in the European Financial Stability Facility’. The amendment of Article 136 TFEU was approved by Law 13 (III) of 2012; the ‘Six-Pack’ was implemented through Law 194 (I) of 2012 on the Medium Term Budgetary Framework and the Fiscal Rules. The Treaty on the European Stability Mechanism (hereinafter ESM) was ratified through Law 14 (III) of 2012. Even the Memorandum of Understanding and the Financial Assistance Facility Agreement through which Cyprus received financial support were ratified through Law 1 (III) of 2013. The only exception so far has been the Fiscal Compact, which was adopted under Article 169(1) of the Constitution by an Act of the Council of Ministers (governmental decree) on 20 April 2012 without a vote in the Parliament. It was later ratified and published in the Official Journal of the Republic of Cyprus upon the Cypriot Council of Ministers’ decision, in accordance with Article 169(3) of the Constitution, in Greek and in English.54 The ratification was completed by the notification to the Council of the EU on 3 July 2012.

53 For a comprehensive analysis see Nikos Skoutaris, ‘Cyprus’ in S Griller and E Lentsch (eds), EMU Integration and Member States’ Constitutions (Oxford, Hart, 2021) 361. 54 Official Journal of the Republic of Cyprus 4157/ 29 June 2012.

396  Nikolas Kyriakou and Nikos Skoutaris

VI.  Constitutional Rules and/or Practice on Implementing EU Law With the exception of the EAW case, there were no further serious cases challenging the application of EU law in the Cypriot legal order. The Supreme Court recognised in several judgments the primacy of EU law and the obligation for EU regulations and directives to be applied without impediments. This application has been fairly unproblematic. Examples of this jurisprudential line are identified already in the aforementioned Konstantinou judgment, where the Court stated: We are well aware of the constant and aligned case-law of the Court of Justice of the EU, according to which EU law prevails over the law of the Member States. We appreciate this case-law and with all due respect we consider that it could not have been different, since if Member States did not abide by their obligations, as these arise out of the Treaty, that they themselves had signed, this Treaty would collapse.55

More recent instances include Karkotis, where the Supreme Court stated that ‘[t]he ­principle of supremacy of EU law over national law has already been jurisprudentially recognised (see, Attorney General v Konstantinou) and is applied according to the provisions of the Fifth Amendment and the added constitutional provision of Article 1A’.56 In Sigma Radio TV Public Ltd, the Court dealt with the application of an EU Directive in the national legal order and stated that the principle of supremacy of EU law … is expressed in the case of Directives by the obligation of n ­ ational courts to interpret national legislation in conformity with a Directive that has the same object even in the case where the national law predated the Directive as was the case in Marleasing SA v La Comercial International de Alimentacion SA (Case C-106/89) [1990] ECRI – 4135, where such an ­obligation was unequivocally recognised.57

One of the techniques employed for the transposition of secondary legislation, especially ­directives, is the designation of competent authorities for the implementation of the substantive provisions either in the main text of the transposing law or by promulgating special laws to this effect and/or to provide for sanctions in the event of non-compliance. On this latter point, a typical legal provision will usually provide that ‘Whoever infringes article X of regulation Y is liable to a fine of Z amount of euro and/or a sentence of imprisonment that does not exceed N years’.58 One illustrative example is the law on the recognition of agricultural products and foodstuffs as traditional specialities guaranteed, L 43(I)/2011, which sought to give effect to some aspects of Regulation (EC) No 1216/2007.59 In this law, the Minister of Agriculture, Environment and Natural Resources is designated as the competent authority for the implementation of the

55 Supreme Court, Konstantinou (n 43). 56 Supreme Court of Cyprus, 14.07.2009, Karkotis manufacturing and trading public limited v Republic, Recourse no 1187/2007. 57 Supreme Court of Cyprus, 03.04.2015, Sigma Radio TV Public Ltd v Cyprus Radiotelevision Authority, Appeal 56/2010. 58 Constantinos Lycourgos, ‘Cyprus Public Law as affected by accession to the European Union’ in C Kombos (ed), Studies in European Public Law: Thematic, national and Post-National Perspectives, (Athens, Sakkoulas Publications, 2010) 109. 59 Regulation (EC) No 1216/2007 of 18 October 2007 laying down detailed rules for the implementation of Council Regulation (EC) No 509/2006 on agricultural products and foodstuffs as traditional specialities guaranteed.

Cyprus  397 Regulation. In addition, Part X of that law provides for the penal law aspects of the implantation setting out the relevant offences and providing for the penalties and fines. Most importantly, an umbrella law was passed in 2007 in order to address those situations where EU law obliged Member States to ensure through penal and administrative means and penalties the effective application of EU law. The Law on the Application of Community Regulations and Community Decisions, L 78(I)/2007 covers all those situations where no other specific national legislation applies and creates a set of penal and administrative procedures and penalties relating to failure to abide by EU Regulations and Decisions. There are no specific rules in the Constitution or in primary national legislation that regulate the transposition and implementation of EU law in Cyprus’ legal order. The Ministry of Finance issued a guide in which the necessary steps for the legislative drafting and process are described. However, this is merely best practices’ guide intended to assist desk officers in the competent Ministries in their day-to-day discharge of their duties.60 In short, the competent Ministry is tasked with monitoring EU’s legislative activity and preparing the draft bills in the areas falling under its competence. The draft bills must undergo a stage of consultation with the public and they must later be submitted to the Law Office of the Republic, accompanied by two documents: (a) an impact assessment report, (b) a table of correspondence between the provisions of EU and national law. The Law Office of the Republic is responsible to check that the draft bill fully and correctly transposes the provisions of EU law. Once the Law Office clears the draft bill, the Ministry submits it to the Council of Ministers and after receiving its approval, to the House of Representatives. A positive aspect of Cyprus’s accession to the EU and the application of EU law was that new bodies, institutions and independent authorities were established pursuant to relevant pieces of EU legislation, thus changing the modus operandi of the Cypriot public administration. The main change relates to the introduction of a decentralised system of overseeing the application of EU law. This system broke away with the centralised oversight entrusted to ministries and governmental departments and introduced in practical terms transparency and multiple checks between stakeholders in the implementation of EU law. A final and peculiar characteristic of the application of EU law relates to the prevailing political circumstances on the island. While the application of the acquis communataire is suspended in the areas which are not under the effective control of the Republic, it is important to note that national laws are not. However, a growing percentage of national law is promulgated in order to implement EU law, mainly directives. This legislation continues to have effect throughout the island per national legislation. To give but one example, the right to free movement accorded to EU citizens is granted in relation to the whole of the island, but the right of residence61 and the status of third country nationals62 are applicable solely to the areas where the Republic exercises effective control. The Cypriot courts have proved extremely hesitant to submit requests for preliminary references to the CJEU.63 Although the accession took place in 2004, the first reference was

60 www.reform.gov.cy/en/better-regulation/simplification-of-legislation. 61 Directive 2004/38/EC of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States amending Regulation (EEC) No 1612/68 and repealing Directives 64/221/EEC, 68/360/EEC, 72/194/EEC, 73/148/EEC, 75/34/EEC, 75/35/EEC, 90/364/EEC, 90/365/EEC and 93/96/EEC [2004] OJ L 158/77. 62 Council Directive 2003/109/EC of 25 November 2003 concerning the status of third-country nationals who are longterm residents [2004] OJ L 16/44. 63 Lycourgos, ‘Cyprus Public Law’ (n 57) 112. This author concurs with our assessment.

398  Nikolas Kyriakou and Nikos Skoutaris submitted only in 2008. In total, the CJEU’s statistics show that the Cypriot courts submitted in total nine such requests in nearly 18 years of EU membership.64 We consider that the low number of preliminary references is a token of the lack of a judicial dialogue between the Cypriot courts and the CJEU. These statistics may be explained by the lack of familiarity on the part of the Cypriot courts both with the procedural and substantial aspects of EU law and by the additional time that will inevitably be required for the resolution of a case should a case reach the CJEU through a preliminary reference.

VII.  Resulting Relationship between EU Law and National Law The internationalisation of law and the judicialisation of international law, in the broad sense, are factors conducive to the constant transformation and adaptation of the role of national courts. In-depth knowledge of a wide array of sources of law is nowadays a sine qua non condition for the enterprise of comprehensive and just adjudication. It is in this ‘brave new world’ that the Supreme Court of Cyprus is called on to operate and to engage in a continuous judicial discourse with international adjudication bodies. This chapter has shown that the engagement of the Supreme Court of Cyprus with the fundamentals of EU law had an uneasy start. This Court indirectly challenged the hierarchy of norms within the EU when it adjudicated in Konstantinou that the EAW provisions were running counter to a constitutional provision and that, thus, the latter should prevail. The Court’s analysis is open to criticism because it conflated the notion of ‘direct effect’ with that of ‘supremacy’ of EU law and their respective operation within the Cypriot legal order. In our view, the handful of requests for a preliminary reference made by Cypriot courts to the CJEU is an indicator that allows us to conclude that Cypriot courts still remain hesitant, if not suspicious, towards active interaction with the CJEU. Cypriot courts have been increasingly resorting to and citing the case law of the CJEU. In our opinion, this trend is yet to consolidate to the same level as the extended use of the case law of the European Court of Human Rights (hereinafter ECtHR) made by the same courts.65 This difference may be explained by the fact that there is a 50-year long interaction between the ECtHR and the domestic courts, whereas there is just a 18-year congruent period for EU law. Nevertheless, since 2004, Cypriot courts are EU courts, which are obliged to apply EU law correctly.

VIII. Conclusion Cyprus’s historical itinerary since the 1960s presents a rare blend of EU, public international, and national law issues. The carefully designed consociational arrangement in the Constitution 64 Court of Justice of the European Union, ‘Annual Report 2020’, 233. 65 Constantinos Kombos and Annie Pantazi, ‘Human Rights post Lisbon – Cypriot Report’, in J Laffranque (ed), The protection of fundamental rights post-Lisbon: the interaction between the Charter of Fundamental Rights of the European Union, the European Convention on Human Rights and national constitutions: FIDE XXV Congress Proceedings Vol 1 (Tallinn, Tartu University Press, 2012) 302, 327 f: ‘the Constitution is modelled on the Convention and that structural connection has been strengthened by the remarkable willingness of the Supreme Court to rely on the jurisprudence of the ECtHR in order to construe properly the constitutional provisions on fundamental rights. However, such an approach is definitely not dynamic in nature in the sense of combing external sources of rights and deriving a common higher denominator, simply because the Supreme Court has not yet shown a willingness to be comparably extrovert, as it has been towards the convention, with the body of EU law’.

Cyprus  399 proved to be short-lived and failed to accommodate the antagonistic relations between the two communities, while a reverse reading suggests that the two communities lacked faith in these arrangements and sought revision to their exclusive benefit. Either reading points to the direction that consociational arrangements are not in themselves sufficient to resolve historic and political tensions. Instead, they require genuine political commitment by all stakeholders to produce the result for which they were initially drafted. In the first part of this chapter we also attempted to present the evolution of the initial bi-communal arrangements from the 1960 Constitution to the Annan Plan and focused on the issue of Cyprus’s accession to the EU. This accession was fiercely embattled by lawyers on both sides and eventually resulted in Cyprus becoming an EU Member State without solving the long-standing problem of division. Becoming an EU Member State brought Cyprus’s domestic legal order within the ambit of a brave new world. At the initial steps of the post-accession period, the supremacy of EU law over the Constitution yielded to the Supreme Court giving precedence to the constitutional provisions and prompted an amendment to the Constitution in order to provide explicitly for the supremacy of EU law. This initial resistance from the Supreme Court was not unusual, taking into account the reaction of senior courts of other EU Member States. However, it has to be pointed out that given the existence of the doctrine of necessity the Cypriot constitutional order has proved quite flexible in absorbing the tensions that EU membership often raises. The somewhat anomalous constitutional trajectory of the RoC can only be amended within the environment of a new comprehensive settlement of the conflict. However, for the time being, such a prospect seems distant.

References A Arnull et al (eds), A Constitutional Order of States? Essays in EU Law in Honour of Alan Dashwood (Oxford, Hart Publishing, 2011). L Besselink et al, National constitutional avenues for further EU integration (Brussels, European Parliament, 2014). J Crawford, G Hafner and A Pellet, ‘Republic of Cyprus: Eligibility for EU Membership, Opinion of 24 September 1997’ (reprinted in A Markides (ed), Cyprus and European Union Membership. Important Legal Documents (Nicosia, 2002) 13). A Emilianides, ‘Accession of the Republic of Cyprus to the EU, the Constitution and the Cypriot Doctrine of Necessity’, (2007) The Cyprus Yearbook of International Relations 65. N Emiliou, ‘Cyprus’, in A Kellerman et al (eds), The impact of EU Accession on the legal orders of new EU Member States and (pre-) candidate countries, Hopes and Fears (The Hague, TMC Asser Press, 2006) 303. C Kombos, The impact of EU law on Cypriot Public law (Athens, Sakkoulas Publications, 2015). C Kombos, ‘Report on Cyprus’ in J Czuczai et al (eds), Division of Competences and Regulatory Powers between the EU and the Member States. FIDE XXVII Congress Proceedings Vol 3 (Budapest, Wolters Kluwer, 2016) 282. C Kombos and A Pantazi, ‘Human Rights post Lisbon – Cypriot Report’, in J Laffranque (ed), The protection of fundamental rights post-Lisbon: the interaction between the Charter of Fundamental Rights of the European Union, the European Convention on Human Rights and national constitutions: FIDE XXV Congress Proceedings Vol 1 (Tallinn, Tartu University Press, 2012) 302. N Kyriakou, ‘Report on Cyprus’ in G. Martinico and O Pollicino (eds), The National Judicial Treatment of the ECHR and EU Laws: A Constitutional Comparative Perspective (Groningen, Europa Law Publishing, 2010) 151. A Loizou, ‘Cyprus’, in R Blackburn and J Polakiewicz (eds), Fundamental rights in Europe: The ECHR and its Member States. 1950-2000 (Oxford, Oxford University Press, 2001) 220.

400  Nikolas Kyriakou and Nikos Skoutaris C Lycourgos, ‘Cyprus Public Law as affected by accession to the European Union’ in C Kombos (ed), Studies in European Public Law: Thematic, national and Post-National Perspectives (Athens, Sakkoulas Publications, 2010) 109. D Markides, Cyprus 1957–63: From Colonial Conflict to Constitutional Crisis. The Role of the Municipal Issue (Minneapolis MN, University of Minnesota Press, 2001). M Mendelson, ‘The Application of the ‘Republic of Cyprus’ to Join the European Union, Opinion of 6 June 1997’ (reprinted in Maurice Mendelson, Why Cyprus’ Entry into the European Union Would be Illegal. Legal Opinion, by Professor M. H. Mendelson QC (London, Embassy of the Republic of Turkey, 2001) 33). G Pikis, Constitutionalism – Human Rights – Separation of Powers (Leiden, Brill, 2006). N Skoutaris, ‘The application of the acquis communautaire in the areas not under the effective control of the Republic of Cyprus: The Green Line Regulation’, (2008) 45 CML Rev 727. N Skoutaris, ‘Cyprus’ in S Griller and E Lentsch (eds), EMU Integration and Member States’ Constitutions (Oxford, Hart, 2021) 361. E Stefanou and A Kapardis, ‘The first two years of fiddling around with the implementation of the European Arrest Warrant (EAW) in Cyprus’, in E Guild (ed), Constitutional challenges to the European arrest warrant (Nijmegen, Wolf Legal Publishers, 2006) 75. J Weiler, ‘The Community System: the dual character of Supranationalism’, (1981) 1 Yearbook of European Law 267.

14 Latvia MARTINS MITS*

I. Introduction Latvia joined the European Union (EU) in 2004, 14 years after restoration of its independence. During a relatively short period of time the Latvian legal system has undergone tremendous changes. Latvian people have always wanted to be part of democratic Europe and EU membership was seen by many as a way to consolidate internal democratic and economic processes. The overall attitude towards the EU was positive, as was convincingly demonstrated in the public referendum on joining the EU. EU membership took place only 14 years after Latvia had escaped from another union – the Union of Soviet Socialist Republics (USSR) where it was incorporated against its will. This rendered questions of independence and sovereignty particularly sensitive, and joining the EU was viewed with suspicion by some of the population. Recent historical experience had left not only demographical marks on the composition of the population, but it had also made the constitutional vision underlying the establishment of the state of Latvia in 1919 rather remote and vague. Prompted by an internal democratic process and by worrying geopolitical developments seen by many as posing potential threats to the statehood, a significant constitutional debate took place on the question of Latvia’s constitutional identity. This debate resulted in unprecedented constitutional amendments. In 2014 the Constitution was supplemented with a Preamble clarifying the object and purpose of the state and setting out fundamental principles that allegedly cannot be amended. This development will be explored in the context of the main question of this chapter – identification of the constitutional limits in Latvia to further European integration. Particular attention will be paid to the application of EU law by courts and, in particular, to the fundamental constitutional principles or values and the principle of the highest human rights standard as developed by the Constitutional Court.

II.  Main Characteristics of the National Constitutional System The Latvian state was proclaimed on 18 November 1918. A Constitutional Assembly elected with the aim of drafting and adopting a Constitution accomplished this task on 15 February 1922 * The manuscript was completed in October 2018.

402  Martins Mits when the Satversme1 (the Constitution) was adopted. This Constitution is the basic law of the country today. The Constitution de facto applied until 1940, when Latvia was incorporated by force into the USSR. As a parliamentary democracy, though, Latvia had lasted only until 15 May 1934, when the Prime Minister Kārlis Ulmanis dissolved the Parliament and the laws from then on were passed by Ministru Kabinets (the Government). The Constitution was followed as far as it was applicable at that time. From the point of view of the Constitution, incorporation of Latvia into the USSR in 1940 took place without free expression of the will of the people and in contradiction to the core principles and procedure envisaged in the Constitution. Latvia followed a doctrine of state continuity maintaining that de jure the state of Latvia continued to exist as a subject of international law (including all legitimately adopted laws) throughout the USSR period. Therefore, when the restoration of independence was proclaimed on 4 May 1990 by Augstākā Padome, the Parliament at the time,2 the authority of the Constitution was also restored, while its application was simultaneously suspended except for four articles which were subject to amendments only through public referendum.3 Full applicability of the Constitution was restored on 6 July 1993 when the first freely elected Saeima (the Parliament) held its first session. Latvia, unlike its neighbours Estonia and Lithuania, has never adopted another constitution during its independent existence. The existing Constitution has been inspired by both the Westminster model (a strong Parliament, responsible executive body and a figurehead President) and the Weimar Constitution (strong direct democracy).4 The drafters had aimed at creating a modern Constitution, relying extensively on the best practices of other countries. During debates in the Constitutional Assembly the speakers referred to such countries as: the US (30 times), the UK (29), Switzerland (25), Germany (25), France (22), Estonia (12), Russia (9), Czechoslovakia (6), Poland (3), Finland (3), Austria (2), Hungary (2), Sweden (2), Japan (2) and 13 other countries.5 Besides, the decision not to draft a new Constitution in the 1990s was a largely pragmatic compromise in view of the variety of proposals as to how to develop the constitutional system after the restoration of independence.6 The Constitution eventually proved itself as a solid working tool even after almost a century since its creation. Latvia is a parliamentary democracy. The powers are shared among the Parliament, consisting of 100 members who are elected for a period of four years based on proportional representation, the Government, consisting of the Prime Minister and 13 ministers, as well as the judiciary. The President is elected by the Parliament for a period of four years and, apart from the representative functions, has the right to veto laws and initiate the dissolution of the Parliament. The latter right was exercised recently; at the peak of the crisis of public trust in politicians and the Government, the people for the first time dissolved the Parliament, in a public referendum held in 2011. The judiciary is comprised of the Constitutional Court and courts of general jurisdiction. The Constitutional Court is comprised of seven judges and was established in 1996.7 It exercises 1 Latvijas Republikas Satversme, 15 February 1922, Latvijas Vēstnesis (hereinafter LVLV) 43, 1 July 1993. 2 Deklarācija ‘Par Latvijas Republikas neatkarības atjaunošanu’, 4 May 1990 [‘On the Restoration of Independence of the Republic of Latvia’], Ziņotājs 20, 17 May 1990. 3 These were Arts 1 (democracy, independence, republic), 2 (sovereign power), 3 (territory) and 6 (elections of the Parliament). 4 Caroline Taube, Constitutionalism in Estonia, Latvia and Lithuania: A Study in Comparative Constitutional Law (Uppsala, Iustus Förlag, 2011) 112. 5 Jānis Pleps, ‘Par Latvijas Republikas Satversmi: vēsture un mūsdienas’ [‘On the Constitution of the Republic of Latvia: History and Present’], in Latvijas Republikas Satversme (Riga, Latvijas Vēstnesis, 2012) 47, 51 f. 6 Gunārs Kusiņš, ‘Kā pilnveidot mūsu valsts Satversmi’ [‘How to Improve the Constitution of our State’], in Satversmes reforma Latvijā: par un pret (Riga, Sociāli ekonomisko pētījumu institūts ‘Latvija’, 1995) 33–40. 7 ‘Satversmes tiesas likums’ [‘Constitutional Court Law’], 5 June 1996, LVLV 103, 14 June 1996.

Latvia  403 abstract judicial control over compliance of legal norms with higher legal norms that may be initiated by a wide range of subjects. Notably, since 2000, individuals may submit constitutional complaints about alleged violations of their fundamental rights set out in the Constitution. There is a three-tiered court system with 17 courts, including six regional courts and the Supreme Court hearing civil, criminal and administrative cases.8 Administrative courts were introduced in 2004 with the entry into force of an entirely new Administrative Procedure Law.9 The Supreme Court has a Senate with three departments (civil, criminal and administrative) acting as a court of cassation.

III.  Constitutional Culture The Latvian legal system is not uniform in terms of its doctrinal basis. It was formed in the first part of the twentieth century under the strong influence of legal positivism dominant in Europe at the time. The Soviet legal doctrine that ruled in Latvia from 1940–90 was also based on legal positivism. However, it placed a stronger emphasis on the written letter than on the spirit of the law. This meant that literal reading was the dominant method of interpretation of law during the Soviet period, whereas application of law according to its object and purpose (as far as it was not strengthening the Socialist state with the ultimate goal of achieving Communism) was almost non-existent. The role of a judge was that of passively applying legal rules passed by the Parliament. The restoration of Latvian independence at the beginning of the 1990s brought new meaning to the law. Membership of the United Nations and a desire to join the family of democratic European countries determined the need to open the country’s legal system to international law and to the best practices of other countries. This was necessary in order to bring Latvian laws and their application rapidly in line with contemporary demands of democracy, rule of law and protection of human rights. The two diametrically opposite trends – adherence to the letter of law and openness towards international law and practice – coexist and shape the legal environment in Latvia also today. The Constitution itself is of paramount importance for the legal system. It can be argued that in addition to its legal dimension it is a symbol of the interrupted Latvian statehood and independence. The peculiarity, however, is that the 1922 Constitution was brief and laconic in its wording. It did not specifically address the question of international law and it was not well-fitted to ensure a fast transition to a contemporary democracy. It lacked a catalogue of human rights, which was added only in 1998. All these factors increased the importance of legal scholarship and the work of legal practitioners. The Constitutional Court has played a particularly significant role in shaping the constitutional culture. It has consistently extended material interpretation of the Constitution to fill in gaps in its written text. Inspired by legal doctrine,10 the Constitutional Court rapidly ‘discovered’ general principles of law, deriving them from the concept of ‘democracy’ in Article 1 of the Constitution.11 Linking general principles of law with the written constitution ‘positivised’ them, made their application more understandable and more legitimate among the lawyers, despite the

8 ‘Likums “Par tiesu var”’ [‘Law “On Judicial Power”’], 15 December 1992, Ziņotājs 1, 14 January 1993. 9 ‘Administratīvā procesa likums’, 25 October 2012, LVLV 164, 14 November 2001. 10 Egils Levits, ‘Interpretation of Legal Norms and Notion of “Democracy” in Article 1 of the “Satversme”’, (1997) 1 Latvian Human Rights Quarterly 64. 11 Art 1 of the Constitution reads: ‘Latvia is an independent democratic republic’.

404  Martins Mits fact that none of the principles is expressly laid down in the text of the Constitution. Likewise, international human rights treaties and, in particular, the case law of the European Court of Human Rights (ECtHR) substantially influenced the interpretation of human rights standards contained in the Constitution and the methodology for their application. It can even be argued that the European Convention on Human Rights (ECHR) has served as a textbook in the work of the Constitutional Court and partially also in the Supreme Court.12 This is hardly surprising since the ECHR became binding on Latvia in 1997 before the Constitution was supplemented with Chapter 8 containing the national human rights catalogue. In view of the new and laconic human rights provisions in the Constitution, the Constitutional Court developed a principle of harmonious interpretation of the Constitution with international human rights provisions and applied it routinely.13 Courts of general jurisdiction also have accepted direct applicability of international treaties and the priority of the binding international human rights standards over laws passed by the Parliament. However, in contrast to the Constitutional Court, there are mixed messages as to the judicial activism that came from the courts of general jurisdiction. For example, in the Supreme Court one can find a case where the concept of a ‘living instrument’ developed by the ECtHR is applied in order to interpret provisions of the Constitution14 and a case which, with a reference to, inter alia, scholarly writings from 1939, sets strict limits to judicial activism.15 This illustrates the opposite doctrinal trends that coexist within the Supreme Court and among the lawyers at large. Overall, the Latvian legal system developed openly towards international law and international human rights law which prepared a legal environment that was friendly for reception of EU law.

IV.  Constitutional Foundations of EU Membership The Constitution was amended prior to Latvia’s accession to the EU in 2003. The amendment covered two Articles – 68 and 79.16 These amendments pursued two goals – to ensure a constitutional basis for the delegation of competencies to the EU and to regulate the procedure of accession.17 Article 68 of the Constitution was supplemented with a second, a third and a fourth part. The constitutional basis for the accession was provided in Article 68(2): Upon entering into international agreements, Latvia, with the purpose of strengthening democracy, may delegate a part of its State institution competencies to international institutions. The Parliament may ratify international agreements in which a part of State institution competencies are delegated to international institutions in sittings in which at least two-thirds of the members of the Parliament participate, and a two-thirds majority vote of the members present is necessary for ratification.

The Constitution was amended on a premise that it was ‘competencies’ and not ‘sovereign powers’ or ‘independence’ that was being delegated. The working group that drafted the amendments 12 Martins Mits, European Convention on Human Rights in Latvia: Impact on Legal Doctrine and Application of Legal Norms (Lund, Lund University Media Tryck, 2010) 278. 13 See below section V.A.2. 14 ECLI:LV:SC:2012:SKA184 para 10. 15 ECLI:LV:SC:2012:SKC4 para 10–12.2. 16 Grozījumi Latvijas Republikas Satversmē [Amendments to the Constitution of the Republic of Latvia], 8 May 2003, LVLV 76, 25 May 2003. 17 Normatīvā akta projekta ‘Grozījums Latvijas Republikas Satversmē’ anotācija, No. 158, para I(1), www.saeima.lv/bi8/ lasa?dd=LP0158_0.

Latvia  405 to the Constitution pointed to the special character of Articles 118 and 219 of the Constitution containing concepts of independence and sovereign powers respectively. Noting that these Articles preceded the text of the 1922 Constitution, that their de facto force was restored earlier than that of the absolute majority of other Articles following the restoration of independence on 4 May 1990 and that these two Articles continued to enjoy special status, the working group concluded that in view of the historical development and traditions of its composition, amendments to the text of Articles 1 and 2 of the Constitution are not acceptable.20

The annotation to the draft amendments to the Constitution submitted by the Ministry of Justice also referred to the constitutional development and traditions, pointing that a conceptual agreement had been reached during the drafting process not to amend Articles 1 and 2.21 It seems that it was a primarily procedural and not a material argument that provided the constitutional basis for the delegation of competencies to the EU since the substance of both articles was not emphasised. The working group, though, discussed concepts of ‘democracy’, ‘sovereign powers’ and ‘independence’ in the context of delegating competencies. It viewed ‘independence’ as the one most affected by becoming an EU Member State. The working group noted that a contemporary state cannot reach its goals without international cooperation and that independence implies the possibility to claim back delegated competencies. Therefore Latvia’s independence would not be infringed by becoming an EU Member State.22 The approach of the working group and later of the Parliament can be viewed as an example of treating the Constitution as a living instrument where the concepts of ‘democracy’, ‘sovereignty’, and ‘independence’ are interpreted according to present-day needs. Moreover, the Ministry of Justice suggested that by way of a broad interpretation of the Constitution it would not be necessary to amend the Constitution prior to EU membership. However, it considered that it is preferable to amend the Constitution in order to ensure its correct interpretation and ‘to create basis for interpretation of the existing Constitutional provisions in accordance with the law of the European Communities’.23 There was a substantial constitutional debate on Latvia’s accession to EU, including the question of delegation of competencies. One of the views was that the delegation of sovereignty has to be legitimised by amending Article 2 of the Constitution.24 The dominant view, though, was that the three above discussed concepts are not substantially affected by the EU accession and that, considering Article 68(2) of the Constitution, the changes can be accommodated through interpretation. Later the emphasis in the constitutional debate fully shifted towards the concept of sovereignty, but the Constitutional Court confirmed the same approach. In view of the competence of the Constitutional Court to examine compliance of legal norms (including ratified international treaties) with higher legal norms (including the Constitution), a group of individuals challenged the ratification of the Lisbon Treaty by Latvia, claiming that it violates, inter alia, the principle of sovereignty (Article 2 of the Constitution). After extensive analysis of the concept of sovereignty and the changes brought about by the Lisbon Treaty, the Constitutional Court 18 See above section III. 19 Art 2 of the Constitution reads: ‘The sovereign power of the State of Latvia is vested in the people of Latvia’. 20 Kādēļ Latvijas konstitūcijā nepieciešami labojumi in Jurista Vārds 14 (2001) ch 2. 21 Akta projekta ‘Grozījums LV Satversmē’, 158 (n 17). 22 Kādēļ Latvijas konstitūcijā (n 20) ch 3. 23 Akta projekta ‘Grozījums LV Satversmē’, 158 (n 17). 24 Ineta Ziemele, ‘Eiropas Saveinība – pozitīvs izaicinājums Satversmei’ [The European Rescue – A Positive Challenge to the Constitution], 3 Likums un Tiesības (2001) no 3, 70, 75 f.

406  Martins Mits found that the sovereignty had not been infringed and noted that the delegation of competencies to the EU is rather a manifestation of the use of sovereignty by the people, and not of its weakening.25 The first sentence of Article 68(2) of the Constitution sets four conditions: (1) only part of institutional competencies may be delegated, (2) only to international institutions, (3) by means of concluding international agreement, and (4) with a purpose of strengthening democracy. The latter requirement merits particular attention. It can be questioned whether the limited possibilities for Latvian citizens to take part in the EU’s decision-making process compared to their potential impact on the state’s level strengthens democracy in Latvia, and whether the purpose of joining the EU was to strengthen democracy at all. Despite this formulation, it is clear that the drafters sought to limit possibilities of delegating competencies to ‘undemocratic international institutions’ (or to states). This is confirmed by the annotation to the draft amendments.26 This also meant that the drafters and subsequently the Parliament viewed the EU institutions as ‘international’ (not, for example, supranational) and ‘democratic’. When viewed through the lens of history, a point of view that participation in the EU adds to the consolidation of democracy in Latvia is justified. Latvia became an EU Member State 14 years after regaining its independence. Following the principles set by the long-established democracies, and acting alongside them, provided a valuable experience and indirectly contributed to strengthening of the democratic processes at the domestic level even if it was not the aim of the EU itself. This logic was also recounted in the ruling of the Constitutional Court referring to EU law before Latvia became an EU Member State. In the context of finding unconstitutional restrictions on voting rights for persons in detention on remand, the Constitutional Court noted Latvia’s intention to become an EU Member State, that EU law is compatible only with a legal system that is based on democracy and the rule of law, and that the restrictions under review did not exist in any of the EU Member or candidate states.27 Hence, the intent to join the EU served as an argument for finding the impugned restrictions ‘undemocratic’. The ‘purpose of strengthening democracy’ is a vague concept that implies the internal dimension (the EU as a consolidating factor of the internal democratic processes) and an external factor (participation of the citizens in EU decision-making). If EU integration moved in the direction of a democracy deficit and lesser impact of EU citizens on the operation of EU institutions, the argument against the constitutionality of such integration would weigh heavier. It is relevant in the Latvian context where the principle of democracy (direct and participative) is highly valued. It is also interesting to note that the draft amendments explicitly mentioned the EU,28 but the adopted text refers to ‘international organisations’, potentially allowing delegation of a part of institutional competencies to other international (supranational) organisations as well.29 The second sentence of Article 68(2) of the Constitution requires an absolute majority for the ratification of the accession treaty – two-thirds of the members of the Parliament must be present and two-thirds of those members present must vote in favour of the accession. This is the same procedure as required for passing amendments to the Constitution (Article 76). However, the Parliament would be able to vote on the accession agreement (Accession Agreement)30 only if the citizens in a public referendum decided to join the EU. 25 ECLI:LV:CC:2009:20083501 paras 15–18.10 and 18.3. 26 Akta projekta ‘Grozījums LV Satversmē’, 158 (n 17). 27 ECLI:LV:CC:2003:20021801 paras 6 f. 28 Kādēļ Latvijas konstitūcijā (n 20), ch 2. 29 See also Egils Levits, in Jundzis (ed), Eiropas tiesības [European Law], 2nd edn, (2007) 585. 30 Treaty of Accession of the Czech Republic, Estonia, Cyprus, Latvia, Lithuania, Hungary, Malta, Poland, Slovenia and Slovakia, [2003] OJ L 236.

Latvia  407 Article 68(3) of the Constitution requires a referendum on accession to the EU: Membership of Latvia in the European Union shall be decided by a national referendum, which is proposed by the Parliament.

Although proposed by the Parliament, the wording makes it clear that the referendum is mandatory. The second sentence of Article 79 of the Constitution that regulated the procedure for passing laws through a public referendum was also amended to include EU matters.31 The procedure is the same as for adopting laws through a public referendum: at least half of the number of voters who took part in the last parliamentary elections must participate and majority of them must vote in favour of the accession. Apart from the initial decision whether to join the EU, Article 79 of the Constitution also regulates public referendums on ‘substantial changes’ in Latvia’s EU membership. It is interesting to note that, on the one hand, the substantial character of the constitutional changes prompted by the EU accession was acknowledged by introducing a mandatory referendum. On the other hand, the threshold was set lower than that required for adopting amendments to the Constitution in the public referendum – at least half of the eligible voters in Latvia have to vote in favour of the constitutional amendment (Article 79 of the Constitution). The procedure set for the public referendum was coherent with an argument that Latvia’s accession to the EU did not substantially affect the fundamental principles embodied in the Constitution. Another amendment was introduced in Article 68 of the Constitution relating to substantial changes in Latvia’s participation in the EU. Article 68(4) reads: Substantial changes in the terms regarding the membership of Latvia in the European Union shall be decided by a national referendum if such referendum is requested by at least one-half of the members of the Parliament.

Importantly, it was a deliberate choice of the drafters to leave it for the members of the Parliament to decide whether changes are ‘substantial’ and, if so, then to propose a referendum.32 The argument was to ensure that referendums are initiated only concerning highly relevant questions of EU integration and not, for example, in order to challenge directives adopted by EU institutions.33 It can be argued that leaving it to the discretion of politicians to decide whether changes in EU membership are ‘substantial’ and then whether to initiate a referendum, may pose a danger to the efficiency of the mechanism. Moreover, it may be argued that if not applied responsibly, there is a risk of conferring competencies to the EU institutions without the required degree of legitimation and to the extent which interferes with the material contents of sovereignty, independence and democracy as embodied by the Constitution. However, the Constitutional Court can prevent that by ruling on the constitutionality of delegation of competencies. In the Lisbon Treaty case the Constitutional Court confirmed the discretionary power of the members of the Parliament to decide on proposing of referendum in case of substantive changes in EU membership under Article 68(4) of the Constitution. At the same time, the Constitutional Court accepted the constitutional complaint submitted by individuals who claimed that by passing a law on the ratification of the Lisbon treaty the Parliament had violated their right to take part in the conduct of public affairs (Article 101 of the Constitution), 31 The second sentence of Art 79 reads: ‘A draft law, decision regarding membership of Latvia in the European Union or substantial changes in the terms regarding such membership submitted for national referendum shall be deemed adopted if the number of voters is at least half of the number of electors as participated in the previous Saeima [the Parliament] election and if the majority has voted in favour of the draft law, membership of Latvia in the European Union or substantial changes in the terms regarding such membership’. 32 Kādēļ Latvijas konstitūcijā (n 20) ch 4. 33 Akta projekta ‘Grozījums LV Satversmē’, 158 (n 17).

408  Martins Mits ie, that the Parliament had delegated further competencies to the EU without receiving legitimate mandate from the citizens by way of a referendum. The Constitutional Court analysed the competences that were affected by the Lisbon Treaty and concluded that its ratification by the Parliament did not infringe upon the principle of sovereignty embodied in Article 2 of the Constitution and, hence, the lack of a mandatory referendum under the examined circumstances did not lead to violation of Article 101 of the Constitution. The approach of the Constitutional Court demonstrates that individuals may challenge the legitimacy of the delegation of competencies done without citizens’ approval through a public referendum by way of a constitutional complaint and that the Constitutional Court is ready to review this matter. The referendum on Latvia’s accession to the EU was held on 20 September 2003. Of the 71.49 per cent of voters who took part in the referendum, 66.97 per cent voted in favour of the accession while 32.26 per cent voted against it.34 So far this has been the most attended referendum in the history of Latvia and the voters gave high legitimacy to Latvia’s participation in the EU. The Parliament ratified the agreement on Latvia’s accession to the EU with a law passed on 30 October 2003 (the Ratification Law)35 by 83 out of 100 votes cast in favour, four against and two abstentions.36 Latvia became an EU Member State on 1 May 2004. It must also be noted that the Parliament introduced amendments to two articles regulating fundamental rights on 23 September 2004 in the context of EU membership.37 Article 98 of the Constitution (right to movement and expulsion) was amended to provide an exception from the general prohibition to extradite Latvian citizens if this is provided for in a concluded international treaty. Article 101 of the Constitution (voting rights) was amended to allow EU citizens permanently residing in Latvia to take part in local elections. Both amendments were made exclusively to ensure compatibility of the Latvian constitutional system with EU law, namely with the Council Framework Decision 2002/584/JHA on the European Arrest Warrant in the case of Article 98 and with Article 19 of the Treaty Establishing the European Community in the case of Article 101.38

V.  Constitutional Limits to EU Integration The Latvian Constitution does not contain an eternity clause and does not pronounce any of its Articles as unamendable. Nevertheless, Article 77 of the Constitution points out limitations with respect to six constitutional articles – Article 1 (democracy, independence, republic), Article 2 (sovereign powers), Article 3 (territory), Article 4 (official language and flag), Article 6 (parliamentary elections) and Article 77 (procedure for amending the listed six articles) – to the extent that amendments of these articles by the Parliament requires approval in a public referendum. This could be seen, prima facie, as making further EU integration constitutionally smooth for Latvia, subject to the discretion of the citizens. There are, however, five aspects that require particular attention in order to assess the degree of constitutional complexity for deeper EU integration: (1) borders drawn by the fundamental principles or values embodied in the Constitution 34 Central Election Commission of Latvia, Results of National Referendum on Latvia’s Membership in EU, www.cvk.lv/ cgi-bin/wdbcgiw/base/sae8dev.aktiv03er.vis. 35 Ratification Law, 30.10.2003, LVLV 159, 12.11.2003. 36 Saeima of the Republic of Latvia, Saeimas sēžu stenogrammas, Latvijas Republikas 8. Saeimas rudens sesijas vienpadsmitā sēde, www.saeima.lv/steno/2002_8/st_031030/st3010.htm. 37 Grozījumi Latvijas Republikas Satversmē, 23 September 2004, LVLV 159, 7 October 2004. 38 Normatīvā akta projekta “Grozījums Latvijas Republikas Satversmē” anotācija, 818, para 1(2), www.saeima.lv/L_ Saeima8/lasa-dd=LP0818_0.htm.

Latvia  409 as recognised by the Constitutional Court, (2) principle of the highest human rights standard applied when interpreting the Constitution, (3) the constitutional core doctrine stemming from the constitutional scholarship, (4) its link with the adoption of the Preamble to the Constitution, and (5) the potential limits to integration beyond EU legal order such as in relation to financial instruments. Although all these aspects are interrelated, for the sake of clarity they will be addressed in turn.

A.  Constitutional Principles 1.  Fundamental Principles or Values This section examines the fundamental principles as developed exclusively by the Constitutional Court. The Constitutional Court has clearly indicated that Latvia’s participation in the EU is not unlimited. It has addressed this question in three cases. In 2008 in the Riga Harbour case the Constitutional Court, while explaining that the domestic law must be interpreted in accordance with EU law, noted that such interpretation could be applied only if it ‘does not interfere with the fundamental principles of the Constitution’.39 The Constitutional Court did not further explain what these constitutional principles are and the exact role of the fundamental principles in the case was not clear. However, this statement meant that there is something so fundamental in the Constitution that limits application of EU law and in that respect makes the Constitution superior to EU law. Recently in the Real Estate Tax case, the Constitutional Court repeated the same formula, among other relevant principles, with a reference to the Riga Harbour case.40 Furthermore, in 2009 in the Lisbon Treaty case the Constitutional Court was faced with the question of the extent of delegation of competencies to EU institutions following ratification of the Lisbon Treaty by the Parliament. The Constitutional Court made the following important statements: The Constitutional Court recognises that the state of Latvia is based on such fundamental values that, inter alia, include fundamental rights and freedoms, democracy, sovereignty of the state and of the people, separation of powers and the rule of law. The state has the duty to secure these values and they cannot be infringed by introducing amendments to the Constitution made only by law. Consequently, delegation of competencies cannot go as far that it would infringe fundaments of an independent, sovereign and democratic republic based on the rule of law and fundamental rights. Likewise, it cannot interfere with the right of citizens to decide on questions that are substantial for a democratic state … Delegation of competencies to the EU and integration of the legal acts of the European Community into our system establishes certain restrictions that are permissible only in the case if the EU law is compatible with the principles of a democratic state and sovereignty of the people that follow from Articles 1 and 2 of the [Constitution].41

The Constitutional Court indicated that the concepts of independence, sovereignty (of the state and of the people), democracy (including the right of citizens to decide on important matters), the rule of law, and fundamental rights put a limit to further delegation of competencies to the EU. However, it is important to note that the Constitutional Court made this statement within the scope of the case, ie, these concepts could not be limited by an ordinary law passed by the Parliament such as the law on ratification of the Lisbon Treaty. The Constitutional Court singled

39 ECLI:LV:CC:2008:20071103 40 ECLI:

para 25.4. LV:CC:2018:2017280306 para 10. paras 17 and 18.3.

41 ECLI:LV:CC:2009:20083501

410  Martins Mits out the principles of a democratic state and sovereignty of the people as setting limits to transfer of competencies. Furthermore, the Constitutional Court made it clear: As the EU integration develops, it is necessary to take into consideration the fact that Article 2 of the [Constitution] does not provide for an unlimited delegation of competencies, which would prohibit considering Latvia as a sovereign State.42

It can be concluded that the power of the Parliament is clearly limited with respect to delegation of competencies. Likewise, the fundamental principles of the Constitution as they are reflected in the text of the Constitution, in particular the principle of sovereignty (of the state), limit transfer of substantial competencies by the people of Latvia. The Constitutional Court even stated that it ‘has the duty to ensure supremacy of the Constitution’. However, this statement was made in the context of discussing the procedure for undertaking the international obligations but not discussing the hierarchical relationships between EU law and the Constitution.43 An important question arises – whether the people of Latvia in exercising their sovereignty would be precluded from crossing the set borders. The further reasoning might suggest that the citizens have unlimited powers. In particular, the Constitutional Court noted that the people of Latvia have ‘exclusive powers to deal with the fundamental norms of the Constitution, i.e., to revoke the Constitution or to provide for a new constitutional order.’44 Literal reading would suggest that the Latvian citizens are not limited in changing the Constitution and delegating most or all competencies to EU. Such a conclusion, however, must be approached with caution since, first, the Constitutional Court made its statements in the context of analysis of the Parliament’s powers and not of the powers of the people of Latvia per se and, second, they were made before the doctrine of the constitutional core was developed and the Preamble to the Constitution was adopted.45 The Constitutional Court has not yet been directly confronted with the question of the limits to the powers of the people of Latvia.

2.  Principle of the Highest Human Rights Standard There is a legal principle that has been long elaborated by the Constitutional Court. It is a procedural principle of the highest human rights standard applied when interpreting the Constitution. It has particular relevance in the context of the application of EU law. On the basis of Article 8946 of the Constitution the Constitutional Court has concluded that the aim of the legislator has been achievement of mutual harmony between the Constitution and international human rights provisions and that the Constitution as far as possible has to be interpreted in line with them.47 In 2005 the Constitutional Court further developed this argument into a strong statement: On its essence the Constitution cannot provide for lesser scope of ensuring or protecting human rights than is provided for by any of the international human rights acts. A different conclusion would be at variance with the idea of the rule of law contained in Article 1 of the Constitution since one of the main forms of expression of the rule of law is recognition of human rights and fundamental freedoms as the supreme value of the state.48 42 Ibid, para 18.3. 43 Ibid, para 11.1. 44 Ibid, para 14. 45 See below sections V.B. and V.D. 46 Art 89 reads: ‘The State shall recognise and protect fundamental human rights in accordance with this Constitution, laws and international agreements binding upon Latvia.’ 47 ECLI:LV:CC:2000:20000301 para 5. 48 ECLI:LV:CC:2005:2005020106 para 10.

Latvia  411 Importantly, the scope of the relevant human rights standard in the Constitution may be broader than that of the international human rights provision and then the standard of the Constitution must be applied. In the ECHR context, for example, the Constitutional Court explained that if the situation falls within the scope of the ECHR, then it usually falls also within the scope of the Constitution, but if the situation does not fall within a scope of the ECHR, the analogous conclusion cannot automatically be applied to the Constitution and it must be ascertained whether the Constitution does not contain a higher standard of protection.49 The higher standard of the Constitution may be established on the basis of domestic law or a treaty and the Constitution then accordingly must be interpreted in line with that highest standard. The Constitutional Court, for example, has found that the scope of the right to equality and prohibition of discrimination under Article 91 of the Constitution is broader than Article 14 ECHR50 or the right to education under Article 112 of the Constitution is broader than Article 2 of Protocol No 1 of the ECHR51 and consequently has applied the highest standard – the Constitution. This approach corresponds with the logic of international human rights law following which clauses are inserted into the international human rights treaties expressly allowing application of higher human rights standards contained in other international treaties or domestic law. The Constitutional Court has applied this approach also when confronted with EU law. Notably, in a judgment dealing with the rights of minority shareholders the Constitutional Court found that the challenged provisions of the Law on Credit Institutions allowing the increase of the equity capital without participation of the minority shareholders were disproportionate and violated the right to property under Article 105 of the Constitution.52 Invited experts concluded that the impugned regulation was contrary to Articles 25(1) and 29(1) of Directive 77/91/EEC.53 The Constitutional Court also established, on the basis of the case law of the Court of Justice of the European Union (CJEU), that the directive did not permit any exceptions to decision-making by a general meeting of shareholders with respect to increases of the company’s capital. Then it continued: However, it cannot be concluded in the present case that the result of litigation depends on interpretation of the abovementioned directive. International norms do not prohibit member states to establish a higher level of protection of rights in their constitutions. Consequently, the Constitutional Court does not have the duty to address the CJEU to render preliminary ruling.54

It seems that the Constitutional Court found that the regulation of the Law on Credit Institutions did not comply with the Constitution. It implied that it was also contrary to the Directive 77/91/ EEC, but since EU law was clear and since the result of the case was the same when reviewed against the Constitution, there would be no need to request a preliminary ruling. A reference to the ‘higher level of protection’ in the Constitution is a clear indication that the Constitution may provide for a higher standard than EU law. There is another case where the concept of higher protection under the Constitution was indirectly invoked in the context of EU law. The case dealt with the question of freezing financial assets in the context of money laundering and fight against terrorism. When assessing whether 49 ECLI:LV:CC:2011:20107101 para 12.1. 50 ECLI:LV:CC:2002:20010603 para 3. 51 ECLI:LV:CC:2005:2004180106 para 11. 52 ECLI:LV:CC:2011:20107101. 53 Second Council Directive 77/91/EEC of 13 December 1976 on coordination of safeguards which, for the protection of the interests of members and others, are required by Member States of companies within the meaning of the second paragraph of Article 58 of the Treaty, in respect of the formation of public limited liability companies and the maintenance and alteration of their capital, with a view to making such safeguards equivalent, [1977] OJ L 26/1. 54 ECLI:LV:CC:2011:20107101 para 15.2.

412  Martins Mits it is obliged to request a preliminary ruling from the CJEU, the Constitutional Court made the following observation: Even if the Court of Justice [of the European Union] would conclude that Directive 2005/60/EC55 requires refraining from execution of transactions for a certain period after the Person subject to the Law has reported to the Control Office, this would neither affect the outcome of the case because the Constitutional Court should verify compliance of the established procedure with the [Constitution].56

Here the Constitutional Court assessed two possible scenarios as an outcome of the preliminary ruling procedure. The argument quoted means that, irrespective of the possible outcome in the case of a preliminary ruling, the Constitutional Court has to assess compliance with the Constitution. Possibly, the Constitutional Court meant that the outcome from the preliminary ruling proceedings would not be relevant for the outcome of the case because it would lead to the same result as the analysis under the Constitution, while literal reading rather suggests that the preliminary ruling procedure would not be relevant because the case should (the term used by the Constitutional Court) be decided on the basis of the Constitution and not EU law. In the Lisbon Treaty case the Constitutional Court also had to answer the claim of the applicants that application of the Charter of Fundamental Rights of the European Union (CFR) can come into conflict with the Constitution. The Constitutional Court noted that CFR, ECHR, and the Constitution are all based on similar values. However, it reiterated that the Constitution cannot provide for a lower level of protection than that envisaged by international human rights instruments. With a reference to Article 51 CFR the Constitutional Court noted that the CFR can be applied only when the matter falls into the sphere of EU law and it cannot be applied in the areas outside EU competence. Therefore ‘the constitutional human rights mechanism is autonomous and independent from EU’.57 As to the possible conflicts involving various human rights protection systems, inter alia, those of Latvia, ECHR and the EU, the following statement was made: The Court cannot assess in abstracto the claim regarding possible collision of different systems for protection of human rights. Such collisions will have to be solved at each particular case taking into consideration circumstances of the case.58

If in the above examined cases EU law seems to have played a secondary role, then recently in the Real Estate Tax case EU law itself played the role of the ‘higher human rights standard’, the Constitution being interpreted in accordance with it. In the Real Estate Tax case the Constitutional Court was expressly requested to review the Riga Municipality regulations establishing specific tax requirements for, inter alia, citizens from other EU countries, with Article 91 of the Constitution and Articles 18(1) and 21(1) TFEU in the context of prohibition of discrimination and the right to free movement and residence. The Constitutional Court reconfirmed that, when interpreting provisions of the Constitution, international human rights documents and EU law (in this case the TFEU), a solution has to be sought that would ensure harmony among these provisions. As a matter of principle the Constitutional Court made it clear that the ‘interpretation of Article 91 of the Constitution is influenced by the provisions and the practice of their application as contained in international human rights documents and also, in certain situations, in the TFEU.’59 55 Directive 2005/60/EC of the European Parliament and of the Council of 26 October 2005 on the prevention of the use of the financial system for the purpose of money laundering and terrorist financing, [2005] OJ L 309/15. 56 ECLI:LV:CC:2009:20084701 para 15.2. 57 ECLI:LV:CC:2009:20083501 para 18.7. 58 Ibid. 59 ECLI: LV:CC:2018:2017280306 para 10.

Latvia  413 After having established that, according to EU law, which is binding upon Latvia, nationality is one of the prohibited grounds of discrimination, the Constitutional Court read nationality into Article 91 of the Constitution, which does not explicitly list prohibited grounds of discrimination, and concluded that the challenged regulations had violated Article 91 of the Constitution, without finding it necessary to continue with the assessment under Articles 18(1) and 21(1) TFEU. Overall, the principle of applying the highest human rights standard when interpreting the Constitution is firmly established in the case law of the Constitutional Court. When resorting to the principle of harmonious interpretation, the Constitutional Court has routinely referred to the ‘as far as possible’ notion, implying that its application cannot lead to diminishing the fundamental rights standard in the Constitution. This approach makes the fundamental rights provisions in the Constitution flexible and open to outside influences. The highest standard to be applied may come equally from Latvian law or from any binding human rights treaty, being read into the material contents of the constitutional provision. There are no cases where this approach would have led to express acknowledgment of a conflict between EU law and the Constitution with the latter being eventually applied. So far the Constitutional Court has avoided any conflicts through the principle of harmonious interpretation and taking a case-by-case approach. However, the application of the principle of the highest human rights standard may be problematic from the point of view of EU law if Article 53 of the CFR is interpreted, for the sake of ensuring primacy, unity and effectiveness of EU law, as prohibiting domestic courts from interpreting the Constitution in line with the domestic law and binding international treaties that provide for a higher human rights standard than EU law and ECHR in the area where they apply.60 The Constitutional Court in such a situation may be faced with a dilemma of either interpreting the Constitution in line with the highest binding human rights standard or in line with the (lower) EU standard.

B.  Constitutional Core A referendum initiated in 2011 and held on 18 February 2012, on the introduction of the Russian language as a second official language in Latvia, served as a catalyst for a broad discussion and the following constitutional amendments in relation to a constitutional core that is not subject to amendments. While the citizens convincingly dismissed the idea of introducing Russian language alongside Latvian as an official language by 74.8 per cent voting against and 24.88 per cent in favour out of 71.13 per cent of the eligible voters participating,61 the referendum initiated discussions about such existential questions as the following: are there unalienable constitutional features of the state that cannot be altered, and whether citizens are free to decide any question, including deciding to give up the state? Two political parties represented in the Parliament drafted proposals for constitutional amendments. One proposal aimed at expressly declaring Articles 1, 2, 3, 4, and 77 of the Constitution unamendable. Another proposal suggested adding to the list of questions in Article 73 that cannot be submitted for public referendum draft laws ‘which are not acceptable in a democratic society

60 CJEU 15.01.2014 C-176/12 (Association de médiation sociale v Union locale des syndicats CGT and Others) ECLI:EU:C:2014:2 188 f. 61 Central Election Commission of Latvia, Referendum on the Draft Law ‘Amendments to the Constitution of the Republic of Latvia’, www.tn2012.cvk.lv.

414  Martins Mits or threaten the fundaments of Latvia as a national state’.62 By different means, both proposals were concerned with the preservation of the constitutional basis of the state. This debate was followed up by the Constitutional Rights Committee (CRC), consisting of five independent experts. The committee was established in 2007 to serve as an advisory body to the President in matters of constitutional reform. In 2012 the CRC delivered a long opinion on the Constitutional Basis and the Unamendable Core of the Constitution.63 The opinion concluded that an unchangeable constitutional core exists in the form of an unwritten constitutional principle that is partly reflected in the written text of the Constitution and partly is not reflected. This unchangeable core reflects the constitutional identity of the Latvian state. Since this opinion has a direct link with the constitutional amendments that followed afterwards, it is relevant to provide a brief summary of its main conclusions. The constitutional identity consists of state (substantive) identity and institutional (structural) identity. The state (substantive) identity is formed by the following principles: 1. the object, purpose and basis – the core: self-realisation of the Latvian nation is the object, purpose and basis of the Latvian state; 2. the territory – the core: the territory without substantive alterations as defined in Article 3 of the Constitution; 3. the people of Latvia – the core: protection of the identity and continuity of the Latvian nation; 4. the sovereign powers – the core: the people of Latvia are limited in giving up independence and sovereignty of the state.64 The institutional (structural) identity is made up of the following principles: 1. democracy – the core: democracy as an object and purpose of the state system, but not a particular way of its realisation; 2. rule of law – the core: principles of separation of powers, legality, protection of human rights (including minority rights); 3. socially responsible state – the core: guaranteeing survival of every person (food, clothes, home, medical assistance); 4. national state – the core: not depriving the state of its Latvian identity (negative obligation) and strengthening this identity (positive obligation).65 The opinion suggested that the people and the Parliament may amend any provision of the Constitution, but only as far as these amendments do not infringe upon the above discussed constitutional core. However, the constitutional core has to be interpreted narrowly by applying a conditio sine qua non test – whether an element can exist without a part of it – and it cannot be determined in the abstract.66 The opinion essentially viewed Latvia as a nation state in the Westphalian meaning, at the same time incorporating the basic principles of a current European state. It is possible to draw 62 Both legislative proposals are analysed in opinion of the CRC, see the reference below paras 320 f. 63 ‘Par Latvijas valsts konstitucionālajiem pamatiem un neizskaramo Satversmes kodolu’, in Par Latvijas valsts konstitucionālajiem pamatiem un neizskaramo Satversmes kodolu: Konstitucionālo tiesību komisijas viedoklis un materiāli [On the constitutional foundations of the State of Latvia and the inextricable core of the Satversme in the Constitution of the Republic of Latvia and the invisible core of the Constitution: the opinion and materials of the CRC] (2012). 64 Ibid paras 291–307. 65 Ibid, paras 308–19. 66 Ibid, paras 282–89.

Latvia  415 some parallels with the ‘immutable core’ in the German constitutional doctrine based on Article 79(3) that makes the principles laid down in Articles 1 and 20 of the German Basic Law unamendable. It has been noted, though, that the principles making up constitutional identity in Germany do not include ethnic and cultural specificities of the German people.67 In the Latvian case, by contrast, constitutional identity is linked to the Latvian people. The reasons for such a difference can be explained from the historical perspective of the last century – while part of the German people sought to impose rules based on ethnic identity on other peoples, the Latvian people were oppressed of their identity to such an extent that in today’s situation threats to preserving their own identity are perceived seriously. Overall, the development related to the identification of a constitutional core and the resulting limitation of the sovereign power of the people to amend the Constitution in the name of the protection of the constitutional core, marks limits to Latvia’s possible integration into the EU.

C.  Preamble to the Constitution In its opinion the CRC did not support the above-mentioned two proposals to amend existing provisions of the Constitution, but recommended supplementing the Constitution with a Preamble. This recommendation materialised on 19 June 2014 when the Parliament with 69 votes cast in favour and 28 against added the Preamble to the 1922 Constitution which was hitherto lacking.68 The need for the Preamble was explained as follows. When the state of Latvia was established in 1922 its object and purpose was clear to everyone. Today the situation has changed, since the years of Latvia’s occupation have left an ongoing impact on the historical memory of the society. Besides, most European constitutions adopted since World War II reflect the national and constitutional identities of their respective states. Therefore, it was recommended to reflect such fundamental characteristics of a Latvian state in the Constitution, beyond merely laying down a constitutional order.69 The Preamble emphasises the establishment of the Latvian state and related important aspects, its basic principles and values that formed the basis of the Constitution when it was drafted, as well as those that have crystallised over the years. In particular, the object and purpose of the state is explained in the first paragraph of the Preamble as realisation of the unwavering will and the right of self-determination of the Latvian nation to have its own state and to guarantee the existence and development of the Latvian nation over the centuries, ensuring freedom and promoting the welfare of the people of Latvia and each individual. The Preamble does not expressly refer to unchangeable core or constitutional identity. However, its fourth paragraph resembles the doctrine of the constitutional core as developed by the CRC: Latvia as democratic, socially responsible and national state is based on the rule of law and on respect for human dignity and freedom; it recognises and protects fundamental human rights and respects ethnic minorities. The people of Latvia protect their sovereignty, national independence, territory, territorial integrity and democratic system of government of the State of Latvia. 67 Christian Tomuschat, ‘The Defence of National Identity by the German Constitutional Court’, in Saiz Arnaiz and Alcoberro Llivina (eds), National Constitutional Identity and European Integration (Mortsel, Intersentia, 2013) 212. 68 Grozījums Latvijas Republikas Satversmē, 19 June 2014, LVLV 131, 8 July 2014. 69 Likumprojekta “Grozījums Latvijas Republikas Satversmē” anotācija, 1075Lp11, paras 1–3, http://titania.saeima.lv/ LIVS11/saeimalivs11.nsf/0/64D394CCE1123A42C2257C7D00523485.

416  Martins Mits The annotation supplementing the draft law indicated that the aim of this paragraph was to reveal ‘supreme principles of the state’s operation’ and distinguished between, on the one hand, four concepts describing state’s institutional order (democracy, rule of law, social responsibility, national state) and, on the other hand, five concepts (the people of Latvia, sovereignty, independence, territory and democratic state order) describing the ‘basis of the state’.70 The former precisely correspond to the institutional (structural) principles as discussed above while the latter refer largely to the state (substantive) principles with the exception of the object and purpose of the state that was addressed separately in the Preamble as well as national independence and democratic state order added to the list in the Preamble. As already noted, the Preamble was drafted primarily as a political message, but it includes legal terms and it must be applied in accordance with the general methodology for application of legal rules.71 In essence, it is a tool for interpreting provisions of the Constitution. In view of its link with the doctrinal framework developed by the CRC, it can be strongly argued that the concepts described in the fourth paragraph of the Preamble are general principles of law having a particularly important role in the Latvian constitutional system. There seems to be a general agreement among the legal scholars that the Preamble does not create new substantive provisions, but it makes visible the Latvian constitutional identity starting from the moment of the establishment of the state in 1918 and as it has crystallised throughout the years. This explains why the Preamble was adopted by means of the same procedure as regular amendments to the Constitution, and not through a public referendum.72 Interestingly, the sixth paragraph of the Preamble defines the role of Latvia in the international arena – being an equal member and protecting its national interests, Latvia also ‘promotes sustainable and democratic development of a united Europe and the world’. Thus, the vision encapsulated in the Preamble is that of Latvia as a national (nation) state which promotes development, inter alia, of a unified Europe and, in particular, of the EU,73 towards sustainability and democracy. It could be argued that the two tracks – self-identification as a national state and promotion of a closer union of the European states do not obviously lead in one and the same direction. Europe is seen as a union of nation states with strong constitutional identities. At the same time, it could also be argued that rediscovery of the constitutional identity is in line with Article 4(2) of the TEU which requires respecting of national identities inherent in the fundamental structures of the Member States. Nevertheless, this argument did not appear either in the annotation supplementing the draft Preamble or in the opinion of the CRC and this attests to the internal focus of the debate. The constitutional identity as reflected in the Preamble may draw strict limits to further delegation of competencies to the EU. This will be the case if the existence of Latvia as a national state, based, in particular, on the principles of independence, democracy and sovereignty (both of the state and of the people), is put at risk. Sovereignty as a limit has been particularly emphasised by the Constitutional Court prior to the adoption of the Preamble.74 The Preamble now 70 Ibid, para 6. 71 Ringolds Balodis (ed), Latvijas Republikas Satversmes komentāri, I. nodaļa, Ievads, Vispārējie noteikumi [Commentary on the Constitution of the Republic of Latvia, vol I, Introduction, General Provisions] (Riga, Latvijas Vēstnesis, 2014) 124. 72 During the Parliamentary debates the Chair of the Legal Committee of the Parliament referred to the legal opinions delivered by the invited experts, but the Parliamentary Secretary of the Ministry of Justice to the opinion of the CRC, both concluding that the Preamble does not amend the substantive provisions of the Constitution (Saeima of the Republic of Latvia, Saeimas sēžu stenogrammas, Latvijas Republikas 11. Saeimas pavasara sesijas devītā sēde, www.saeima.lv/lv/ transcripts/view/248). 73 Likumprojekta ‘Grozījums Latvijas Republikas Satversmē’ anotācija, 1075Lp11, para 8, http://titania.saeima.lv/ LIVS11/saeimalivs11.nsf/0/64D394CCE1123A42C2257C7D00523485. 74 See section V.A.1. above.

Latvia  417 articulates the fundamental principles and it remains to be seen how those principles making up constitutional identity will be applied. The Constitutional Court has been reserved with respect to discussing the constitutional identity. However, recently it pronounced for the first time on the principle of the national state. After noting that the Latvian language is an inalienable part of the constitutional identity of the Latvian state, it concluded that Article 4 (official language and flag) of the Constitution embodies and expresses principles of the national state.75 It further described positive and negative obligations stemming from the above principle in the context of the Latvian language. Interestingly, the Constitutional Court made this ‘historical’ pronouncement with a reference to the opinions delivered by the CRC and not the Preamble of the Constitution.

D.  Potential Limits beyond EU Legal Order There may potentially be limits to integration in EU-related areas such as the European Stability Mechanism or the Fiscal Compact. Latvia has experience that merits attention, although it precedes the establishment of the mentioned two instruments. In 2008–09 Latvia experienced a severe economic crisis and the Latvian Government concluded agreements with the International Monetary Fund and the European Community with the aim of taking loans in the amounts of 1.7 billion and 3.1 billion EUR respectively. The Constitutional Court in a judgment dealing with an unrelated matter (state pensions), inter alia, noted that it is the Parliament which, in accordance with the principle of the separation of powers, is mandated to decide on the most important matters for the state and public life. Taking large international loans is such an important matter, but it was done without the Parliament’s participation.76 In essence this meant that the Government had acted ultra vires when taking international loans without Parliament’s approval. The matter was taken up by the CRC which, relying, inter alia, on the principle of separation of powers and the notion of the democratic legitimacy, in its opinion confirmed the need for the Parliament’s authorisation, but concluded that it can be done post factum.77 The solution indeed was provided post factum when the Parliament, first, adopted an ‘announcement’ indicating the lenders, restrictions of the loan’s amount and the main directions for its use and, second, accepted a report by the Minister of Finance who explained obligations stemming from the loans.78 One of the features of the Latvian Constitution is a strong direct democracy. This requires proper legitimation down the chain of public institutions for their actions. The Constitutional Court in this case referred to the principle of the separation of powers, but it also developed a concept of an ‘important matter’ that links with the principle of democracy and requires decision-making at the level of the Parliament which is legitimised by the people. Thus, if the actions in relation to the European financial instruments fall within the scope of an ‘important matter’, they call for a proper authorisation at the level of the Parliament. Moreover, should the participation in the financial instruments require delegation of more extensive competences, this would require legitimation by the people of Latvia. The ultimate limits, however, are set by fundamental principles as reflected in the Preamble of the Constitution, notably by the principle of sovereignty. 75 ECLI:LV:CC:2017:20170101 para 16. 76 ECLI:LV:CC:2009:20094301 para 30.1. 77 Par Saeimas apstiprinājuma nepieciešamību liela apjoma aizņēmumu saņemšanai in Valsts prezidenta Konstitucionālo tiesību komisija. Viedokļi: 2008–2011 (2011) [The need for Saeima approval of large-scale borrowing and the State Constitutional Law Commission. Opinion of the CRC]. 78 For a detailed analysis of this case see Martins Mits, ‘Legal Benchmarks for Lowering Social Protection in Times of Austerity’, (2013) 57 Revue Hellénique des Droits de L’homme 26.

418  Martins Mits

VI.  Constitutional Rules and Practices on Implementing EU Law A.  On the Level of EU Institutions There is a mechanism envisaging coordination and supervision of the national positions represented by the national representatives in the EU institutions, including in the European Council (represented by the Prime Minister) and the Council of the EU (represented by the respective ministers). According to regulations approved by the Government in 2009,79 the national positions are developed within the relevant ministries (unless a special inter-ministerial working group is established), whose work is coordinated by the Ministry of Foreign Affairs. The national positions have to be approved by the Government. However, all national positions that are to be presented at the European Council and the Council of the EU (including its working groups or committees) have to be submitted for review in the European Affairs Committee (EAC) of the Parliament. The EAC is one of the largest Parliamentary committees and, during the previous mandate of the Parliament (2014–18), it was comprised of 16 members of all factions represented in the Parliament, proportionate to their size. Thus, national positions of the Prime Minister or ministers before their presentation in the European Council and the Council of the EU have to be approved by the members of the EAC. The domestic regulation does not leave scope for discretionary actions for the Prime Minister or the ministers. There have been no notable public discussions in Latvia concerning legitimacy of the national positions expressed in Brussels. Section VII1 of the Rules of the Parliament provides that the Parliament participates in the EU decision-making through the EAC, unless it decides otherwise.80 If the EAC does not approve the national position, the relevant ministry has to adjust the position accordingly. Besides, the EAC may also co-ordinate (when necessary) compliance of draft laws processed in the Parliament with the EU law. Since the amendments brought by the Lisbon Treaty, the EAC has rejected several EU legislative initiatives. Together with 11 other EU Member State parliaments, the EAC suspended the draft regulation concerning the exercise of the worker’s collective trade union rights, resulting in its withdrawal by the Commission in 2012. According to the Chair of the EAC, while carrying out subsidiarity review, it was concluded that the draft regulation did not reach its aims as it failed to strike a proper balance between the protection of the rights of workers and providers of services from all EU countries.81 The reasons given seem to suggest that the EAC did not view the breach of the subsidiarity principle as the reason for the rejection of the draft regulation but it was rather the substantive analysis of the draft regulation. This conclusion echoes the position of the European Commission whose spokesperson indicated that the opinions of

79 Ministru kabineta noteikumi 96, 3 February 2009, LVLV 21, 6 February 2009. 80 Saeimas kārtības rullis, 28 July 1994, LVLV 96, 18 August 1994. 81 Pārskats par Saeimas komisiju darbu 2012. gada pavasara sesijā [Review of the work of the Saeima commissions during the spring 2012 session] (2012), www.saeima.lv/documents/b40b61cfb634be757b0ee00f4bf4a86c35b5122a.

Latvia  419 the national parliaments did not reveal a breach of the subsidiarity principle by the European Commission.82

B.  On the Domestic Level 1.  General Rules The Latvian legal system distinguishes between internal and external legal acts. Internal legal acts may bind only the public authority that issued these acts as well as subordinated institutions, but not individuals (instructions, rules, recommendations, etc.). External legal acts (ie, binding externally) have the following hierarchy that is reflected in Article 15(2) of the Administrative Procedure Law: • • • •

the Constitution; laws passed by the Parliament; regulations adopted by the Government; regulations adopted by municipalities.

In 2008 the Ministry of Justice issued a Guide on Implementation of EU Law (the Guide) and in 2014 an updated version followed to accommodate changes introduced by the Lisbon Treaty. The Guide provides guidance to all executive authorities in the matters of application of EU law. It explains the primary sources, EU international treaties, secondary sources, general principles and case law of the CJEU without attempting to place them in the hierarchy of legal norms in Latvia. According to the Guide, provisions of EU directives that apply to public authorities, depending on their contents, can be implemented either through internal or external legal acts. Provisions of EU directives that bind individuals have to be implemented, depending on their scope, either by regulations adopted by the Government, laws passed by the Parliament or, exceptionally, changes in the Constitution.83 Therefore, provisions of one directive can be implemented by several domestic legal acts with a different legal force, usually by regulations and laws. EU regulations become directly part of the domestic legal system and they automatically render inapplicable all conflicting provisions of domestic law. Where an EU decision has to be implemented in domestic law, the same rules as in the case of directives apply. The Guide describes two methods for implementing EU law: transposition (rewriting of EU provision in the domestic act or inserting a reference to an EU act in the domestic legal act) and reformulation (taking over the essence of an EU provision, but not its literal contents).84 Like in the case of a form of the implementing act, the choice of a method depends on the contents of the EU act. With respect to the directives, the Guide subjects the choice of the implementation method to the means that are best suited for achieving the aims of the directive. In addition, the Constitutional Court has indicated that the Parliament and the Government, when implementing directives, have to implement all obligations in a clear and precise manner so that individuals understand their rights and obligations.85 When implementing EU decisions directed towards

82 Brussels drops plans for EU law limiting right to strike (2012), www.euractiv.com/socialeurope/ec-drops-regulationright-strike-news-514793. 83 Eiropas Savienības tiesību ieviešanas rokasgrāmata (2014) 36. 84 Ibid, 22–24. 85 ECLI:LV:CC:2014:20140603 para 21.2.

420  Martins Mits Latvia that contain provisions applicable to private persons, the same rule applies with respect the choice of the implementation method as in the case of directives.

2.  Constitutional Court The Constitutional Court constantly emphasises that EU law is part of the Latvian legal system and it must be taken into account when deciding a case. In the Riga Harbour case the Constitutional Court made the following statement: [U]pon ratification of the Treaty on Accession of Latvia to the European Union, the European Union law has become integral part of the Latvian legal system. Therefore, legal acts of the European Union and interpretation provided by case-law of the European Court of Justice have to be taken into account when applying national law.86

In this case the Constitutional Court found that the land use plan of Riga city violated Article 115 (right to environment) of the Constitution due to procedural shortcomings. The Constitutional Court, inter alia, pointed out that Latvia has no legal grounds for changing the status of its territory which is part of Natura 2000 as established by the EU, besides noting what obligations in this situation stem from EU law for the state as a whole and its courts of general jurisdiction. EU acts and CJEU case law in this case served as an argument for the interpretation of the domestic law. In an earlier case the Constitutional Court had to resolve a conflict involving a law passed by the Parliament, an EU directive, and a treaty.87 Article 1142 of the Administrative Violations Code provided for sanctions against carriers of passengers without valid travel documents. Under Directive 2001/51/EC88 Latvia as an EU Member State was obliged to apply penalties to carriers who transport into its territory third country nationals without valid travel documents. However, Article 3.15 of the International Convention on Facilitation of International Maritime Traffic, to which Latvia became party before joining the EU, prohibited imposition of penalties upon shipowners who carried passengers without valid travel documents. The Constitutional Court first noted that a binding treaty has priority over conflicting law passed by the Parliament. However, after becoming an EU Member State, Latvia had to comply with its obligations stemming from Directive 2001/51/EC. Therefore, the Constitutional Court concluded that Article 1142 of the Administrative Violations Code is not applicable to ships flying under the flag of a non-EU Member State, but is applicable with respect to ships flying under the flag of an EU Member State. With reference to Article 307 TEC it was also pointed out that the state is under an obligation to take steps to eliminate the incompatibility of previously undertaken international obligations with the obligations under EU law.89 It can be concluded that the conflict between a law passed by the Parliament, a directive, and a treaty was resolved following EU law at the expense of potentially breaching the treaty. Overall, the Constitutional Court has made it clear that a principle of harmonious interpretation must be applied to avoid conflicts with EU law and has interpreted the Constitution in line with the EU law without hesitation.90 The Constitutional Court so far has not found EU legal acts contravening the Constitution or EU institutions acting ultra vires.

86 ECLI:LV:CC:2008:20071103 para 24.2. 87 ECLI:LV:CC:2004:20040106. 88 Council Directive 2001/51/EC of 28 June 2001 supplementing the provisions of Article 26 of the Convention implementing the Schengen Agreement of 14 June 1985 [2001] OJ L 187/45. 89 Ibid para 7. 90 See, in particular, discussion of the Real Estate Tax case in section V.A.2. above.

Latvia  421

3.  Courts of General Jurisdiction The Constitutional Court has indicated that the task of the courts of general jurisdiction is to ensure uniform and effective application of EU law. In fulfilling this task the courts can apply EU norms directly and they have to choose an interpretation method that ensures the most appropriate result from the point of view of EU law. If the courts cannot solve the case by following the above principles, then they can request a preliminary ruling from the CJEU.91 Overall, the Latvian courts follow these instructions and there are judgments that attest to both methods – interpretation and direct application – used with the aim of ensuring compliance with EU law. In Case SKC-410 the Supreme Court had to decide on a dispute relating to a refusal to allow an Irish company to register a land purchase in Latvia. The refusal was based on provisions of the Law on Privatisation of Agricultural Land that allowed registration only by nationals of states that have concluded bilateral agreements with Latvia. The Supreme Court ruled that the relevant provisions of the domestic law have to be interpreted in line with Article 3 (free movement of capital) of Annex VIII to the Act of Accession to the EU by Latvia,92 otherwise four EU Member States that had not concluded investment agreements with Latvia would be discriminated against.93 This case provides an example where the domestic court ensured compliance with EU law by means of interpretation. As has been pointed out, the non-compliance of the law passed by the Parliament with EU law was avoided by departing from the literal meaning of the text and following the principles established in various sources of EU law – the TFEU and Directive 88/361/EEC.94 In another case the Supreme Court was faced with the claim that provisions of Governmental Regulation No 378 dealing with issuing of passports, set the term for children’s passports at one year, while according to Directives 68/360/EEC and 2004/38/EC passports have to be issued for at least five years.95 The Supreme Court established that the Government had not adequately implemented provisions of Directive 73/148/EEC96 that was applicable in the situation. Therefore the Government was obliged directly to apply Article 2(3) (five years validity of the passport) of the said directive.97 In this case the Supreme Court indicated that incompatibility between a domestic Regulation and the directive had to be resolved by means of direct application of the directive. Another case demonstrates that the Supreme Court took a cautious attitude to establishing its competence when domestic law, EU law and the ECHR intersected. In Case SKC-1/2013 the applicant objected to recognition of a judgment in Latvia delivered by an English court ordering the applicant to pay to the defendant certain sum of money. The applicant claimed that the English court had taken the decision in absentia, without giving a possibility to appeal the 91 ECLI:LV:CC:2015:49 para 11.2. 92 Annex VIII to the Act concerning the conditions of accession of the Czech Republic, the Republic of Estonia, the Republic of Cyprus, the Republic of Latvia, the Republic of Lithuania, the Republic of Hungary, the Republic of Malta, the Republic of Poland, the Republic of Slovenia and the Slovak Republic and the adjustments to the Treaties on which the European Union is founded, [2003] OJ L 236/824. 93 LVSC 13.01.2010, Case SKC-410/2010 On right to possess land in Latvia, ECLI:LV:SC:2010:SKC410 para 7. 94 Māris Brizgo, ‘Senāts piemēro un interpretē Eiropas Savienības tiesības: Komentārs par Senāta lēmumu lietā Nr. SKC-410/2010’ [‘The Supreme Court Applies and Interprets European Union Law: Comment on the Supreme Court’s Decision in Case No. SKC-410/2010’], (2010) 9 Jurista Vārds 6. 95 Directive 68/360/EEC of 15 October 1968 on the abolition of restrictions on movement and residence within the Community for workers of Member States and their families, [1968] OJ L 257/485; Directive 2004/38/EC of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of Member States, [2004] OJ L 158/77. 96 Council Directive 73/148/EEC of 21 May 1973 on the abolition of restrictions on movement and residence within the Community for nationals of Member States with regard to establishment and the provision of services, [1973] OJ L 172/14. 97 ECLI:LV:SC:2009:SKA175 para 13.

422  Martins Mits judgment and without providing reasons for its decision. The application was based on various provisions of Regulation 44/2001 (Brussels I Regulation)98 and provisions of the Latvian Civil Procedure Law.99 The Supreme Court referred two questions for preliminary ruling to the CJEU with the aim of clarifying whether it could assess the document issued by the English court. Namely, would such an act be in compliance with the principle of mutual trust, and whether the English judgment complied with the right to fair trial under Article 47 of the CFR. After receiving the reply from the CJEU,100 the Supreme Court rejected the applicant’s objection and did not find that there had been a manifest and disproportionate breach of Article 47 of the CFR using a test developed by the CJEU.101 The Supreme Court also explained that, although the grounds for refusal to recognise judgments in the Brussels I Regulation and in Civil Procedure Law overlap, the provisions of domestic law can be applied only when EU law does not regulate the relevant matter or it explicitly requires application of the domestic law, and this was not the case.102 The Supreme Court was cautious not to exceed its competence in the area of fundamental rights. It expressly acknowledged the applicability of EU law and carried out its analysis strictly on the basis of EU law and the principles developed by the CJEU. It must be noted, though, that a narrow reading of EU law in particular circumstances may pose a problem when several legal systems have to be accommodated. In the Avotiņš v Latvia case, which dealt with a similar question of recognition of a judgment delivered in another EU Member State in the absence of the applicant, the ECtHR made it clear that it would examine whether the principle of mutual recognition is not applied automatically and mechanically, to the detriment of fundamental rights.103 In a situation when meritorious complaint has been made that the protection of fundamental rights (eg, fair trial guarantees) in the court of another EU Member State has been manifestly deficient and that this cannot be remedied through EU law (eg, by the preliminary ruling procedure), the domestic courts cannot refrain from addressing such complaint on the substance by arguing that they are applying EU law.104 Thus, the ECHR in particular circumstances requires assessment whether a court in another EU Member State has not provided for manifestly deficient protection of ECHR rights. Overall, the courts of ordinary jurisdiction and the Supreme Court in particular are mindful of the primacy and direct applicability of EU law. A study examining application of EU law by the courts of general jurisdiction during the first five years of Latvia’s membership of the EU showed that initially the courts neither resisted nor were eager to apply EU law, leaving it to the Supreme Court and the Constitutional Court to settle difficult EU law questions.105 This can be explained by the lack of specialist knowledge and is confirmed by the raising number of questions referred to the CJEU by the Supreme Court under the preliminary ruling procedure. The first request was submitted only in 2008, and by 2016 the total number of requests was 46.106 There have been 98 Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, [2001] OJ L 12/1. 99 ECLI:LV:SC:2013:SKC1; Civilprocesa likums, 14 October 1998, LVLV 326/330, 03.12.1998 100 CJEU 04.10.2012 C-30/12 (Marcinová), ECLI:EU:C:2012:619. 101 Ibid, para 62. 102 ECLI:LV:SC:2013:SKC1 para 13.5. 103 ECtHR 25.02.2014 Application No 17502/07 (Avotiņš v Latvia) ECLI:CE:ECHR:2016:0523JUD001750207. 104 Ibid, para 116. 105 Tatjana Evas, Judicial Application of EU Law in Post-Communist Countries. The Cases of Estonia and Latvia (e-book, 2014) ch 7: Jurisprudence of Latvian Courts. 106 Solvita Harbaceviča, ‘Eiropas integrācijas process un Latvijas tiesību sistēmas transformācija’ [‘The Process of European Integration and the Transformation of the Latvian Legal System’] (2016) 13 Latvijas Republikas Augstākās Tiesas Biļetens 40.

Latvia  423 no reports of cases where courts of general jurisdiction would refuse to apply EU law because it contradicted domestic (constitutional) law.

VII.  Relationship between EU Law and National Law A peculiarity of the Latvian legal system is that the Constitution does not expressly regulate the national law–international law and national law–EU law relationships. While there is an overall acceptance in Latvia that EU law must be complied with, there is no clarity with respect to the theoretical question of whether Latvian law and EU law have to be viewed as two distinct legal systems, or EU law is a part of the Latvian legal system. On the one hand, the abovementioned Guide refers to both as parallel legal systems107 and there is an opinion expressed in legal scholarship that both legal systems are separate, autonomous and have specific traditions.108 On the other hand, the Constitutional Court emphasises that EU law is an integral part of the Latvian legal system.109 Furthermore, attempts have been made to place EU norms into the hierarchy of Latvian legal acts.110 Also, Article 15(4) of the Administrative Procedure Law expressly requires application of provisions of EU law in accordance with their place in the hierarchy of the Latvian legal norms.111 There is some truth in both approaches. EU law operates in Latvia – it is a fact of life that is also confirmed by domestic courts, including its direct application. From this perspective, EU law is part of the Latvian legal system. At the same time, EU law operates according to its own principles that are not determined by domestic law. Hence, EU law preserves its autonomy. The question is how the relationship between, first, EU law and the Constitution and, second, EU law and laws and other legal acts with a lower legal force, can be defined. With respect to the second part of the question, legal scholarship has developed a procedural approach. Latvia became an EU Member State by means of a law passed in the Parliament on the ratification of the Accession Agreement. It has been argued that, since Latvia follows a monistic approach and since, as a matter of a general principle of law, provisions of international law have priority over domestic legal provisions in the event of a conflict, the Ratification Law has a higher legal force than laws passed by the Parliament, but lower than the Constitution. Moreover, it transpires from Article 2 of the Act, providing for the conditions and adjustments to Latvia’s accession to EU annexed to the Accession Agreement, that Latvia accepts as binding all EU acts that existed before Latvia’s accession to the EU, including the case law of the CJEU and the principles developed therein such as the principle of primacy of EU law.112 This approach means that EU law enters the Latvian legal system on the level of laws passed by the Parliament and takes precedence over provisions of laws and other lower legal acts. Despite of the wording of Article 15(4) of the Administrative Procedure Law attesting to the contrary, insertion of provisions of EU law into the hierarchy of the domestic legal norms is not 107 Rokasgrāmata (n 83) 15. 108 Linda Ostrovska, ‘Ārējo normatīvo tiesību aktu hierarhija pēc Latvijas pievienošanās Eiropas Savienībai (II)’ [‘The Hierarchy of External Normative Acts Following Latvia’s Accession to the European Union (II)’], 3 Likums un Tiesības (2006) 78. 109 ECLI:LV:CC:2008:20071103 para 24.2; see section VI.B.2 above. 110 Katrīne Pļaviņa, ‘Kā risināt pretrunas starp Latvijas, starptautiskajām un Eiropas Savienības normām’, [‘How to Resolve Contradictions between Latvian, International and European Norms’], 20 Jurista Vārds (2011) 667. 111 Art 15(4) reads: ‘The legal norms of the European Union (Community) shall be applied in accordance with their place in the hierarchy of legal force of external regulatory enactments. In applying the legal norms of the European Union (Community), institutions and courts shall take into account European Court of Justice case law.’ 112 Levits, Eiropas tiesības (n 29) 587 f.

424  Martins Mits relevant for their successful operation. It may also appear problematic from the point of view of the principle of the supremacy of EU law. It has been explained in the commentary to the Administrative Procedure Law that the practical significance of comprehending legal force of, for example, EU regulations, and placing them within the hierarchy of Latvian legal norms is that this could help legal practitioners to locate the level on which the conflict has arisen and to decide on the procedures for solving the conflict, for example, whether the Constitutional Court could be appropriate avenue. However, a provision of EU law will always apply in the case of a conflict, since EU law maintains its autonomous character and principles of application when they operate within the Latvian legal system.113 The Administrative Procedure Law is the only domestic law that suggests placing EU acts into the hierarchy of legal norms. In addition, the Civil Procedure Law in Article 5(3) restricts application of Latvian laws as far as it is permitted by EU law only in the areas where EU law is directly applicable. The Criminal Procedure Law in Article 2(2) also recognises applicability of EU law, but does not pronounce on the questions of the scope or priority of its application.114 Thus, the procedural laws confirm applicability of EU law, but do not clarify whether it has primacy over domestic law up to the level of laws passed by the Parliament (except in the area of civil law, and only where EU law is directly applicable). While such different approaches in the three procedural laws do not have a good explanation, the lack of express regulation of the primacy of EU law can be explained by the openness of the Latvian legal system towards international law that has been shaped by the acceptance of applicability and priority of treaties over domestic legal acts in the area of fundamental rights.115 Indeed, as can be seen from the above overview of the case law, Latvian courts ensured priority of EU law either through the harmonious interpretation or direct application of EU law, without requiring more explicit legal regulation.116 It becomes more complicated when it comes to the Constitution. There is a general acceptance among legal scholars that EU law is inferior to the fundamental principles of the Constitution. An authoritative opinion has been expressed that all six articles listed in Article 77 of the Constitution, requiring a prior approval in a public referendum of any amendments made to these articles, as well as any other article clarifying the contents of the six articles, are superior to EU law.117 This opinion has been supported either in full118 or with some modifications.119 In view of the recent adoption of the Preamble to the Constitution and the preceding opinion of the CRC, it can be argued that the fundamental principles are now listed in the fourth paragraph of the Preamble (democracy, rule of law, social responsibility, national state, the people, sovereignty, independence, territory, and democratic state order). Respect for dignity and protection of fundamental rights are part of the principle of the rule of law, although in the fourth paragraph of the Preamble they have been addressed separately. All these fundamental principles, as well as the object and purpose of the state, although reflected in the Preamble of the Constitution, transcend the scope of the written text and cannot be overruled by EU law. 113 Ineta Ziemele and Daiga Rezevska, ‘Article 15’, in J Briede (ed), Administratīvā procesa likuma komentāri: A un B daļa (Riga, Tiesu namu aģentūra, 2013) 237 f. 114 Kriminālprocesa likums, 21 April 2005, LVLV 74, 11 May 2005. 115 See above section III; compare with Evas, Judicial Application (n 105), ch 2, Substantive Coherence: Constitutional Structures and Principles. 116 See section VI.B.3 above. It has been argued that because of Latvia’s adherence to monism, the Latvian courts should be able to apply EU law without any need for amendments in the procedural laws. Arnis Buka, ‘Chapter 4, General Frameworks’ in T Kerikmae et al (eds), The Law of the Baltic States (Cham, Springer, 2017) 174. 117 Levits, Eiropas tiesības (n 29) 589. 118 Ziemele and Rezevska, ‘Article 15’ (n 113) 239. 119 Eg Ostrovska, ‘Ārējo normatīvo’ (n 108) 79 lists Articles 1, 2, 3, 4, 6, 68 and 79; Pļaviņa, ‘Kā risināt’ (n 110) lists Articles 1, 2, 3, 4, 6 and 77.

Latvia  425 The question of the relationship between EU law and ordinary provisions of the Constitution is unclear.120 One possible conclusion could be that EU law is superior to the laws passed by the Parliament (and legal acts with a lower legal force), but inferior to the Constitution. The Accession Agreement was ratified with a law passed by the Parliament. The Constitution does not expressly provide for ‘constitutional laws’. Following the principles of priority of international norms (domestic legal principle) and primacy of EU law (EU legal principle) in the case of a conflict with domestic norms, EU legal acts have priority only up to the level of laws passed by the Parliament.121 Such an outcome could be problematic from the point of view of the principle of the primacy of EU law. A different conclusion, however, can also be reached. The Ratification Law was adopted in a special procedure set by Article 68(2) of the Constitution requiring a two-thirds majority vote from two-thirds of the members of the Parliament present and voting. This is the same procedure as the one applicable to constitutional amendments. Hence, the law adopted in the special procedure that has been established for this purpose has a constitutional rank. The Constitutional Court has expressly confirmed that the Ratification Law has been adopted under a procedure that gives it constitutional rank.122 In such a case EU law enters the Latvian legal system on the level of the Constitution and, in view of the principles of EU law, it has priority over ordinary provisions of the Constitution, but not over the fundamental principles of the Constitution. This modality of the procedural approach provides for the best accommodation of EU law. It is also possible to argue that a material approach could even better explain the relationship between EU law and the Constitution. Due to the special subject-matter of the Accession Agreement (delegation of part of the competencies), the public referendum held on the EU accession (democratic legitimation for the delegation of competencies) and the special regulation in the Constitution (amendments to Articles 68 and 79), the Accession Agreement has a status equal to the Constitution and this makes EU law superior to the provisions of the Constitution as far as the fundamental principles of the Constitution are not affected. It must be noted, though, that the Latvian Constitution relies heavily on the procedural regulation and an argument based on the material approach is not easily received, protection of fundamental rights being an exception. To summarise, there are good reasons to argue that EU law enters the Latvian legal system on the level of the Constitution and has priority over the provisions of the Constitution, but not over the fundamental principles that are reflected in the Constitution.

VIII. Conclusions Becoming an EU Member State was generally perceived in Latvia as a step towards consolidating democracy. The obligations stemming from EU membership have not been notably challenged on the political level. The courts have conscientiously ensured compliance with EU law either through harmonious interpretation of the domestic law or by its direct application. Latvia is undergoing a process of rediscovering its constitutional identity. This has not been triggered by EU membership, but by internal political and external geopolitical developments. The constitutional debate resulted in supplementing the Constitution with a Preamble in 2014. The Preamble reflects fundamental constitutional principles, a constitutional core, that arguably



120 Levits,

Eiropas tiesības (n 29) 590. ibid, 587. para 56.3.

121 Compare

122 ECLI:LV:CC:2007:2007100102

426  Martins Mits may not be subject to amendments and which are superior to EU law. While the scope of and consequences flowing from these fundamental principles can be clarified only in the course of a time, they have to be seen as limiting transfer of substantive competences to the EU. Already before the adoption of the Preamble, the Constitutional Court had pointed to the fundamental principles or values which set limits to the application of EU law or transfer of the competencies. The extent of the limits in this context is not clear but, as a minimum, this is an indication that the Constitutional Court might not accept the principle of primacy of EU law if it were to come in conflict with the fundamental principles of the Constitution. Likewise, the principle of the highest domestic human rights standard as applied by the Constitutional Court poses a challenge to the supremacy of EU law in Latvia.

References R Balodis (ed), Latvijas Republikas Satversmes komentāri, I. nodaļa, Ievads, Vispārējie noteikumi [Commentary on the Constitution of the Republic of Latvia, vol I, Introduction, General Provisions] (Riga, Latvijas Vēstnesis, 2014). M Brizgo, ‘Senāts piemēro un interpretē Eiropas Savienības tiesības: Komentārs par Senāta lēmumu lietā Nr. SKC-410/2010’ [‘The Supreme Court Applies and Interprets European Union Law: Comment on the Supreme Court’s Decision in Case No. SKC-410/2010’] (2010) 9 Jurista Vārds 6. T Evas, Judicial Application of EU Law in Post-Communist Countries. The Cases of Estonia and Latvia (e-book, 2014). S Harbaceviča, ‘Eiropas integrācijas process un Latvijas tiesību sistēmas transformācija’ [‘The Process of European Integration and the Transformation of the Latvian Legal System’], (2016) 13 Latvijas Republikas Augstākās Tiesas Biļetens 40. G Kusiņš, ‘Kā pilnveidot mūsu valsts Satversmi’ [‘How to Improve the Constitution of our State’], in Satversmes reforma Latvijā: par un pret (Riga, Sociāli ekonomisko pētījumu institūts ‘Latvija’, 1995) 33. E Levits, ‘Interpretation of Legal Norms and Notion of “Democracy” in Article 1 of the “Satversme”’, (1997) 1 Latvian Human Rights Quarterly 64. M Mits, European Convention on Human Rights in Latvia: Impact on Legal Doctrine and Application of Legal Norms (Lund, Lund University Media Tryck, 2010). M Mits, ‘Legal Benchmarks for Lowering Social Protection in Times of Austerity’, (2013) 57 Revue Hellénique des Droits de L’homme 26. L Ostrovska, ‘Ārējo normatīvo tiesību aktu hierarhija pēc Latvijas pievienošanās Eiropas Savienībai (II)’ [‘The Hierarchy of External Normative Acts Following Latvia’s Accession to the European Union (II)’], (2006) 3 Likums un Tiesības 78. K Pļaviņa, ‘Kā risināt pretrunas starp Latvijas, starptautiskajām un Eiropas Savienības normām’ [‘How to Resolve Contradictions between Latvian, International and European Norms’], (2011) 20 Jurista Vārds 667. J Pleps, ‘Par Latvijas Republikas Satversmi: vēsture un mūsdienas’ [‘On the Constitution of the Republic of Latvia: History and Present’], in Latvijas Republikas Satversme (Riga, Latvijas Vēstnesis, 2012) 47. C Taube, Constitutionalism in Estonia, Latvia and Lithuania: A Study in Comparative Constitutional Law (Uppsala, Iustus Förlag, 2011). C Tomuschat, ‘The Defence of National Identity by the German Constitutional Court’, in Saiz Arnaiz and Alcoberro Llivina (eds), National Constitutional Identity and European Integration (Mortsel, Intersentia, 2013) 212. I Ziemele, Eiropas Saveinība – pozitīvs izaicinājums Satversmei’ [‘The European Rescue – A Positive Challenge to the Constitution’], (2001) 3 Likums un Tiesības, 70. I Ziemele and D Rezevska, ‘Article 15’, in J Briede (ed), Administratīvā procesa likuma komentāri: A un B daļa (Riga, Tiesu namu aģentūra, 2013).

15 Lithuania EGIDIJUS KŪRIS

I.  Main Characteristics of the Constitutional System Lithuania’s Constitution (LC) of 1992 (as amended) is a product of Central/Eastern Europe’s constitutional renaissance1 of the 1980s/90s. Like its counterparts in the region, it aims at solidifying national sovereignty, democracy, the rule of law, human rights, political pluralism, free elections, separation of powers, an independent judiciary, civil control of the military, and a socially oriented market economy. Some of its provisions have few foreign analogues.

A.  The System of State Power, with the Emphasis on the Judiciary Though not explicitly mentioned in the Constitution, separation of powers is the basis of the country’s institutional framework. The legislative power is vested in the 141-member Seimas, a unicameral Parliament, elected for four years according to a mixed electoral pattern: 71 MPs are elected in single-representative constituencies, and 70 from party lists. The dualist executive branch is vested in the Government (cabinet) and the President of the Republic – the Head of State, elected for five years by universal suffrage. The Government ‘shall manage national affairs’ (Art 94 LC).2 An average person therefore tends to see the President, who ‘shall represent the State of Lithuania’ and ‘decide the basic issues of foreign policy and, together with the Government, conduct foreign policy’ (Arts 77, 84 LC), as not a constitutive part of, but a power above the executive. His powers with regard to the Government include appointment and dismissal of the Prime Minister (PM; with the Seimas’ assent) and ministers (on the PM’s submission). Ministers are ‘responsible’ to the Seimas and the President, but ‘directly subordinate’ to the PM. The President has no power to dismiss a minister without the PM’s motion, or the Government without the Seimas’ assent, whereas the Seimas can dismiss them independently by a vote of no confidence. The Government must resign after the new Seimas is elected. After presidential elections it performs démission de courtoisie and is routinely reappointed (as it happens, with minor changes in the composition). The President is not entitled to participate in the Government’s sittings, and shall not interfere in the cabinet’s agenda. The Constitution stipulates that the Government ‘shall execute’ presidential decrees (Art 94 LC) – a dormant power so far, since no President has 1 See Imre Vörös, ‘Contextuality and Universality: Constitutional Borrowings on the Global Stage – The Hungarian View’, (1999) 1 University of Pennsylvania Journal of Constitutional Law 651. 2 References to concrete Articles of the Constitution and other legal acts are not provided, unless the issue under consideration so requires. The text of the Constitution is available in English at www.lrkt.lt/en.

428  Egidijus Kūris ever effectively made use of it.3 With regard to the Seimas, the President has bill-introducing and suspensive veto powers, but no right to challenge the constitutionality of statutes. The Seimas can remove the President from office in impeachment proceedings. These (and other) checks and balances have prompted the Constitutional Court to categorise Lithuania’s system of government as ‘parliamentary with some traits of semi-presidentialism’.4 The balance between the powers of the President and those of the Seimas and the cabinet depend on the political conjuncture. They may range from the parliamentary majority supporting the President’s agenda to the effective estrangement of the latter if s/he appears to represent a different political course. Unlike the political branches, the judiciary is formed solely on a professional basis (initiatives to introduce jury or lay judges have not brought about any result). There are three autonomous court systems. The Constitution explicitly consolidates two of them: the sui generis Constitutional Court and the courts of general jurisdiction. The latter form a four-tier system, present in Lithuania until 1940 and reinstated in 1995 (replacing a two-tier court system inherited from the Soviet era): district courts; regional courts; the Court of Appeal; and the Supreme Court. In 2018 49 district courts were amalgamated into 12, 11 of which have between two and six local chambers. The third system, the two-tier administrative courts’ system, originates from a statute and consists of two regional administrative courts (one of them having four local chambers) and the Supreme Administrative Court (SAC).5 As to the formation of the judiciary, the President enjoys more power on lower levels, while going upwards requires an accord to be reached between the President and the Seimas. The latter appoints and dismisses justices of the Supreme Court (and its President) upon the President’s submission and gives its assent to the President regarding appointment and dismissal of judges of the Court of Appeal (and its President). The Seimas plays no role in appointment or dismissal of judges of district, regional and specialised courts. When implementing these powers, the President is advised by the Council of Judges, without whose assent no judge can be appointed, promoted, transferred or dismissed. The Council consists of judges only.6 It, however, plays no role in the formation of the Constitutional Court.7 The latter, established in 1993, consists of nine justices, appointed for nine years by the Seimas upon submission by the President, the Speaker of the Seimas and the President of the Supreme Court, each nominating one candidate every three years8 – this should enable the Court’s gradual renewal by one-third.9 Its President is appointed by the Seimas upon submission by the President of the Republic.10 3 There is no shared opinion as to the subject matter of (virtually so far non-existent) presidential decrees aimed directly at the Government. 4 Lithuanian Constitutional Court ruling (hereafter LTCC) 10.01.1998, case No 19/97 (Government’s programme). LTCC’s acts are available in English at www.lrkt.lt/en. 5 Other specialised court systems may be introduced, should the legislator decide so: Art 111 LC allows for the establishment ‘according to the law’ of specialised courts ‘for consideration of administrative, labour, family, and cases of other categories’. Save the introduction of administrative courts (in 1999; reformed in 2000 and then in 2017–18), there have been no other such initiatives. 6 The inclusion, to the Council, of external members was found unconstitutional (LTCC 09.05.2006, No 13/04-21/ 04-43/04 (System of the judiciary and its self-governance)). The judiciary’s self-governance was gradually achieved owing to a series of Constitutional Court rulings. LTCC 21.12.1999, No 16/98 (Procedure for appointing judges) pruned the powers of the Minister of Justice with regard to the judiciary. The above-mentioned ruling LTTC 09.05.2006 reduced the powers of the President of the Supreme Court vis-à-vis individual judges and strengthened the role of the Council of Judges. 7 LTCC 02.06.2005, No 10/05 (Appointment of Constitutional Court Justice R. K. Urbaitis). 8 But see n 53. 9 The rotation, every three years, of Constitutional Court Justices depends on political conjuncture. In 2020 the Seimas did not appoint any of the candidates. Justices whose term had expired in March 2020 continued to serve 10 to 15 months’ overtime. Extension of the judicial mandate is provided for in a statute, not in the Constitution. This gave rise to some criticism as regards its alleged unconstitutionality, but the view has prevailed that the statutory extension of the mandate safeguards the court from becoming defunct, if the Seimas fails to appoint the successors of the sitting justices.

Lithuania  429

B.  Constitutional Review The Constitution consolidates itself as the Act of supreme legal force; all other acts must comply with it. Review of constitutionality of acts of the Seimas, the Government and the President is vested in the Constitutional Court. Its jurisdiction includes, primarily, abstract a posteriori review of constitutionality. The Court has self-assumed also the power to establish whether the Seimas’ sub-statutory legislation is in compliance with the statutes and whether statutes are not in conflict with ‘constitutional laws’. The latter, despite their confusing title, rank higher than ‘ordinary’ laws, but lower than the Constitution.11 The Court thus is called to ensure the hierarchy of the legal system from the Constitution down to the level of the central authorities’ acts. It has never accepted anything close to a ‘political question’ doctrine and examines the constitutionality/ legality of both normative and individual acts. Review of ministries’, agencies’ and municipal authorities’ acts falls on the administrative courts. A legal provision may not be applied from the day of its invalidation by a non-appealable Constitutional Court ruling. The rulings have erga omnes and ex nunc effect. Pro futuro effect has not been explicitly provided for, thus threatening that The justices that replaced those who had served overtime are expected to serve not the full nine years, but a respectively shorter term. This should allow for the orderly rotation in 2023 (again, dependent on political conjuncture). 10 The term of office of the LTCC President is explicitly defined neither in the Constitution, nor in a statute. The custom always has been that the President served in that capacity until his term of office as a justice expired. The fact that in 2020–21 the President served overtime for almost 15 months (n 9) gave rise to a not insignificant controversy as to whether the extension of the term of a justice should automatically entail also the extension of the term of office of the President, or whether the latter function should be entrusted ex lege to the most senior justice. In 2014, the statutory provision that in the event of the ‘absence’ of the President that function was to be performed by the most senior Justice was relied upon by the then outgoing President for the appointment of his ‘ad interim successor’, who at that time was not the most senior justice – the most senior justice was a ‘newcomer’ to the Constitutional Court. In 2015 the Law on the LTCC was amended so that only a justice who had held the mandate for at least three years could serve as an ad interim President. So now a ‘newcomer’ may not become an ad interim President. The question remains as to whether the outgoing President lawfully has the discretion to appoint his ad interim successor, or the practice of 2014 was a one-off instance of frivolous application of the respective statutory provision. 11 Despite their name, ‘constitutional laws’ are not part of the Constitution, but ‘supermajority’ or ‘qualified laws’, adopted in a special procedure (Art 69 § 3 LC), as opposed to the general procedure designed for ‘ordinary’ statutes (Art 69 §§ 1 and 2). The LTCC in its early case law (LTCC 08.11.1993, No 6 (Amendments to the Law on Elections to the Seimas)) construed the notion of ‘constitutional laws’ so that it encompassed, inter alia, constitutional amendments, as well as the two acts enlisted in Art 150 LC as the constituent part of the Constitution, ie the Constitutional Law ‘On the State of Lithuania’ and the Constitutional Act ‘On the Non-Alignment of the Republic of Lithuania to Post-Soviet Eastern Unions’ (see I.C.1). The Court radically abandoned that construction in its later case law (LTCC 14.03.2006, No 17/ 02-24/02-06/03-22/04 (Limitation on ownership rights in areas of particular value and in forest land); LTCC 30.07.2020, No 5/2019 (Law on Referendum)). The current doctrine, reinforced in LTCC 30.07.2020, treats constitutional amendments and the acts comprising the constituent part of the Constitution (Art 150 in its current wording enlists four acts; see I.C.1) as having the force of the Constitution. ‘Constitutional laws’ within the meaning of Article 69 § 3 LC do not have it, but they are of higher force than ‘ordinary’ laws. For the adoption of a ‘constitutional law’ an absolute majority of votes of all MPs, and for their amendment a three-fifths majority of votes of all MPs is required. Also, ‘constitutional laws’ must be entered in ‘the list of constitutional laws’ (Art 69 § 3 LC); this must be done by a ‘constitutional law’. ‘Constitutional laws’ are not an equivalent of organic laws in the French tradition, for their adoption depends on the Seimas’ discretion, except when a ‘constitutional law’ is directly provided for in the Constitution. The name ‘constitutional laws’ is most unfortunate. Not only it confused the Constitutional Court’s early case law, but in 2002 one ‘constitutional law’ was found to have been unconstitutional (LTCC 24.12.2002, No 49/2000 (Institutions of local self-government)), thus producing a contradictio in terminis, an oxymoron ‘unconstitutional constitutional law’. But there is more to that than only the terminological inconsistency. For two decades the list of ‘constitutional laws’ was non-existent. It was adopted in 2012 (supplemented in 2014). The list now encompasses the titles of ten ‘constitutional laws’. None of them has been adopted, except the one which implemented the Fiscal Compact (see section III.E). The list mentions, among others, the ‘Constitutional Law on Referendum’. Referenda always were regulated by ‘ordinary’ laws. In 2018 the Seimas adopted the new version of the ‘ordinary’ Law on Referendum, which was found unconstitutional on that basis that, once the statute was enlisted as a ‘constitutional law’, its adoption in the new wording (unlike ‘certain’, ie technical rather than substantial, amendments) had to be made not by an ‘ordinary’ law, but by a ‘constitutional law’ (LTCC 30.07.2020). The coming into force of that ruling was postponed until 01.07.2021. No ‘constitutional law’ on referendum was adopted until that day, making referenda impossible at least for some time. This has provoked major criticism.

430  Egidijus Kūris the coming into force of a ruling invalidating certain legislation as unconstitutional may bring about a ‘legal vacuum’ in a respective field. To avoid such a ‘vacuum’, the Court invented a way to circumvent its occurrence by postponing the official publication of the respective ruling;12 later, this possibility was legitimised by a statutory amendment; in the last few years such postponements have become quite frequent.13 All institutions must revoke all provisions adopted by them, which are based on provisions that are unconstitutional; decisions based on them must not be executed if they had not been executed prior to the ruling coming into effect. Ex tunc effect of the rulings is normally excluded, but the Court may allow it in exceptional cases, when the legislator attempts to overcome the Court’s earlier ruling by adopting the provision identical to that already found unconstitutional.14 It also sometimes directs the power of its rulings to the effects of the provisions found unconstitutional. An unconstitutional act, which is no longer valid law, is not automatically erased from the system of legal acts and must be formally annulled by the Seimas, the Government or the President. The Government, as a rule, loses little time in replacing such ‘legal corpses’ with new resolutions, while the legislature sometimes needs years to agree on new provisions. A ruling on constitutionality may be requested by a group of no less than one-fifth of the MPs (ie 29), the Seimas in corpore, the Government, the President (the latter with regard to the Government’s acts only), or any court. Individual constitutional complaint is available from September 2019;15 it can be lodged after the exhaustion of all legal remedies (Art 105 LC), which, as stipulated in the Law on the Constitutional Court (and repeatedly reiterated by the Court), includes ‘the right to apply to a court’, ‘all possibilities established by law for filing a complaint against the decision of the court’ and the adoption of the ‘final and non-appealable [court] decision’ (Art 65). The rudimentary case law of the European Court of Human Rights (ECtHR) relating to this the matter allows to hold that, where the potential applicant under the laws has no right to apply to a court, Lithuanian model of the individual complaint may not constitute an effective remedy within the meaning of Article 35 § 1 of the European Convention on Human Rights (ECHR).16 There have been ebbs and flows of applications by political actors, but the Constitutional Court’s main job-givers are the courts, since the Constitution stipulates that a judge may not apply any law that is in conflict with the Constitution. When there are grounds to believe that the legal act that should be applied in a case is in conflict with the Constitution, the judge should suspend consideration of the case and apply to the Constitutional Court, requesting that it decides on the compliance of that act with the Constitution (Art 110 LC). Since the introduction of the individual complaint the Court has been overwhelmed with them; but, as about 97 per cent of them are declared inadmissible, the lion’s share of cases decided on their merits still come from the courts. 12 LTCC ruling 24.12.2002 (n 11). Rulings are pronounced in the courtroom and published on the Constitutional Court’s website, but this is not considered an ‘official publication’. To be ‘officially published’, since 2013 rulings must be entered in the Register of Legal Acts or (until 2013) had to be published in the Official Gazette. The public notice of the content of the ruling yet to be ‘officially published’ gives the Seimas or the Government the time needed for the adoption of acts which would allow the threatened ‘vacuum’ to be averted. 13 See, eg, LTCC 28.08.2020, No 12/2019 (Legitimacy of Government); LTCC 19.04.2021, No 13/2019 (Direct elections of mayors). In the first of these cases the Government was found to have not properly assumed its powers after the 2019 presidential elections, when it submitted its démission de courtoisie to the newly elected President; in the second one the Constitutional Court found that the introduction, by means of an ‘ordinary’ statute, of the direct elections of mayors (as presidents of municipal councils) was unconstitutional (see nn 51, 86 and 91). 14 LTCC 19.05.2021, No 13/2020 (Bankrupcy of physical persons). Cf also LTCC 19.05.2017, No 12/2016-13/2016 (Bankrupcy of physical persons). 15 The Seimas resolution aiming at the introduction of the individual complaint was adopted in 2007. Since then, there have been several attempts to make the relevant constitutional amendment, which failed having met strong resistance. The initiative materialised only as of 01.09.2019. 16 ECtHR judgment 08.06.2021, application No 48329/19 (Ancient Baltic Religious Association ‘Romuva’ v Lithuania) ECLI:CE:ECHR:2021:0608JUD004832919 §§ 93-97 (not final at the time of the finalisation of this chapter).

Lithuania  431 In addition to abstract norm control, the Court presents conclusions on whether: (i) there were violations of election laws during the elections of the President or of MPs; (ii) the President’s state of health permits him/her to continue to hold office; (iii) Lithuania’s international treaties are not in conflict with the Constitution; (iv) concrete actions of MPs and state officials against whom an impeachment case has been instituted are in conflict with the Constitution. So far, several conclusions have been presented on alleged violations of election laws during parliamentary elections, but only one17 served as a basis for annulment of election results in one constituency. There have been no cases regarding the President’s state of health, and only one of examination, in a preliminary review procedure, of the constitutionality of several provisions of an international treaty – the ECHR;18 four conclusions were presented in impeachment proceedings against five MPs19 and one against the President.20 On the basis of conclusions a final decision ‘on the issues’ is taken by the Seimas, which is not entitled to annul the conclusion, even if it does not follow what that conclusion legally entails.21 One could speculate as to the hypothetical effects of the situation where the Seimas defies a conclusion that an international treaty contradicts the Constitution and nevertheless ratifies it.

C. Peculiarities 1.  Structure of the Constitution The Constitution defines itself as an integral act (Art 6 § 1 LC). In fact, it is an amalgam of five acts, adopted at different times, each having its title, number and due signature. The act entitled ‘the Constitution of the Republic of Lithuania’ was adopted in the referendum on 25 October 1992 (in force from 2 November 1992). Article 150 of the act, in its initial wording, provided that the constituent part of the Constitution shall be the Constitutional Law ‘On the State of Lithuania’ and the Constitutional Act ‘On the Non-Alignment of the Republic of Lithuania to Post-Soviet Eastern Unions’ (Constitutional Act on Non-Alignment), both adopted earlier.22 In 2004 two new entries were added in Art 150: the Law ‘On the Procedure for Entry into Force of the Constitution of the Republic of Lithuania’ of 25 October 1992 (Law on Entry into Force of the Constitution, adopted together with the ‘Constitution proper’) and the Constitutional Act ‘On Membership of the Republic of Lithuania in the European Union’ of 13 July 2004 (Constitutional Act on EU). The first of these new entries was prompted by the Constitutional Court’s interpretation that 17 LTCC conclusion 10.11.2012, No 15/2012-16/2012 (2012 elections to the Seimas). 18 LTCC conclusion 24.01.1995, No 22/94 (ECHR). 19 LTCC conclusion 27.10.2010, No 32/2010-33/2010 (Impeachment of MPs A Sacharukas and L Karalius); LTCC ­conclusion 03.06.2014, No 3/2014 (Impeachment of MP N Venckienė); LTCC conclusion 19.12.2017, No 7/2017 (Impeachment of MP K Pūkas); LTCC conclusion 22.12.2017, No 12/2017 (Impeachment of MP M Bastys). A request to examine the actions of one MP was declared inadmissible (LTCC decision No KT13-S6/2016 (Impeachment of MP V Gapšys)) and the impeachment case was terminated. The first impeachment case against an MP, convicted for a criminal offence, was instituted in 1999. The Constitutional Court was requested to assess not his concrete actions, but (in an abstract review procedure) the constitutionality of one element of the impeachment procedure; it failed then to interpret (which it did in its later case law) that its conclusions were needed in all impeachment cases (LTCC ruling, case No 3-99/5-99 (Impeachment proceedings for criminal offence)). As that MP’s mandate was not revoked owing to the ‘favourable’ parliamentary vote, he was serving his sentence, while remaining an MP. 20 LTCC conclusion 31.03.2004, No 14/2004 (Impeachment of President R Paksas). 21 Eg, in 2010 the Seimas did not revoke the mandate of an MP, impeached for gross violation of the Constitutution and breach of oath, but the non-application of that constitutional sanction had no bearing on the legal findings of the Court (LTCC conclusion 27.10.2010 (n 19)). 22 On 11.02.1991 (by the Supreme Council, following the plebiscite of 09.02.1991) and on 08.06.1992 (in the referendum), respectively.

432  Egidijus Kūris regardless of not being initially named the Constitution’s constituent part, that act was such owing to its subject matter and function.23 The second act legalised Lithuania’s EU membership (see section III.C). The Constitution thus is perceived not only and not so much as an ‘act’, but, first of all, as an integral legal reality, whereas the ‘act’ is, technically, not integral. That legal reality embraces the Constitutional Court’s doctrine, whereby constitutional provisions are construed. This perception of the Constitution as a legal reality rather than a document relates to the constitutional law paradigm-shift, dealt with below (see section II). A new development took place in 2020.24 The Constitutional Court produced the dicta that, in addition to the five above-mentioned acts comprising the Constitution, there are also other ‘acts of constitutional significance’. As a matter of principle, the assessment of certain historic (and at the same time legal) acts as having ‘constitutional significance’ was not new for Lithuanian legal mind. In particular, the 16 February 1918 Council of Lithuania Resolution (Act of Independence), along with the implementing 15 May 1920 Constituent Seimas Resolution on the re-established democratic State of Lithuania,25 and 11 March 1990 Supreme Council Act on the Re-establishment of the Independent State of Lithuania26 have always been treated as the foundation of Lithuanian statehood and, consequently, constitutionalism. Some authors have given similar significance to the 16 February 1949 Council of the Lithuanian Freedom Fight Movement Declaration.27 These ‘pre-constitutional acts’ (in the Court’s words) pertained to the establishment or the restoration of independence of the State of Lithuania and her identity, as well as to the foundations on which the then prospective constitutions were to be based: independence, democracy and the innate nature of human rights; these historic acts provided guiding principles, in the spirit of which the ensuing constitutions were to be interpreted; they serve as interpretive tools for contextual clarification of the constitutional provisions.28 As underlined by the Constitutional Court, these acts form the ‘constitutional foundation of the State of Lithuania’. But the Court went further. It bestowed on these historic acts the title of ‘fundamental constitutional acts of the State of Lithuania’ and ‘primary sources of Lithuanian constitutional law’, and stated that their provisions regarding independence, democracy and the innate nature of human rights and freedoms had ‘supra-constitutional force’ and ‘[might] not be denied by any constitution of the State of Lithuania’. The double-edged wording used by the Court suggests (or at least has been read so) that not only the provisions of these acts are of foundational nature, but that the acts themselves are supra-constitutional; consequently, the provisions of the actual Constitution can be tested vis-à-vis these provisions. This reading raised not a few eyebrows.29 Also, the Court held that the said historic constitutional acts ‘[might] never be altered or repealed’. That statement, 23 LTCC 29.10.2003, No 1/02 (Procedure for the publication of legal acts). 24 LTCC ruling 30.07.2020 (n 11). 25 See, eg, Mindaugas Maksimaitis, Lietuvos valstybės konstitucijų istorija (XX a. pirmoji pusė) [The history of constitutions of the State of Lithuania (The first half of XX century)] (Vilnius, Justitia, 2005) ch II and ch III.1. 26 See, eg, Kęstutis Lapinskas, ‘Lietuvos konstitucijų istorinių teisinių sąsajų beieškant’ [‘In search of historical legal links between the constitutions of Lithuania’], in Konstitucija, žmogus, teisinė valstybė [Constitution, human being, law-based state] (Vilnius, Lietuvos žmogaus teisių centras, 1998). 27 The Movement embodied Lithuania’s political and military resistance to the Soviet occupation. See, eg, Vaidotas Vaičaitis (ed), Lietuvos konstitucionalizmo istorija (Istorinė Lietuvos Konstitucija): 1387 m. – 1566 m. – 1791 m. – 1918 m. – 1990 m. [History of Lithuanian Constitutionalism (Lithuania’s Historic Constitution): 1387, 1566, 1791, 1918, 1990] (Vilnius, Vilniaus universiteto leidykla, 2016) ch III. 28 As, eg, the US Declaration of Independence, not a part of US Constitution, is a tool for its interpretation. 29 See, eg, the bewildered confession of a historian of constitutional law that he was unable to grasp the ‘logical paradox’ of how ‘constitutional acts’ could be at the same time also ‘pre-constitutional’ and ‘supra-constitutional’. Vaidotas Vaičaitis, ‘2020 m. liepos 30 dienos Konstitucinio Teismo nutarimo dictum’ [‘The dictum of the Constitutional Court ruling of 30 July 2020’], www.teise.pro/index.php/2020/08/18/v-vaicaitis-2020-m-liepos-30-dienos-konstitucinio-teismo-nutarimo-dictum.

Lithuania  433 presumably prompted by the bitter experience of countries who were pressured to revoke their acts of independence, could be regarded as a precaution, a safeguard against the threats (how realistic, would be a political assessment) stemming from Lithuania’s historically adverse geopolitical environment. But the introduction, by no one other than the Constitution’s official interpreter, of the notion ‘supra-constitutional’ and its use alongside the routine reiteration of the long-standing definition of the Constitution as the ‘supreme law’, was an unprecedented doctrinal novelty, begging for further clarification in line with canons of legal construction.

2.  International Law Under the Constitution, Lithuania shall ‘follow the universally recognised principles and norms of international law’, ‘seek to ensure national security and independence, the welfare of the citizens and their basic rights and freedoms’, ‘contribute to the creation of international order based on law and justice’ and ‘participate in international organisations provided that this is not in conflict with the interests and independence of the State’ (Articles 135, 136 LC). ‘International organisations’ is a broad term, which includes different types of alignments. The self-imposition of international law and the limits on participation in international organisations are linked to the policy goals indicated in the cited provisions. One form of ‘participation’ is excluded by the Constitutional Act on Non-Alignment: the state has pledged ‘to never join, in any form, any new political, military, economic or other unions or commonwealths of states formed on the basis of the former USSR’. Activities seeking to draw Lithuania into such alignments are regarded as ‘hostile to the independence of Lithuania’. In addition to the prohibition of ‘weapons of mass destruction and foreign military bases’ on Lithuanian territory (Article 137 LC), the Act specifically prohibits the presence not only of ‘military bases’, but also of ‘army units’ of Russia, CIS or its constituent states. This negative aspect of geopolitical orientation is one of the cornerstones of Lithuania’s foreign policy. The imperative of following the universally recognised principles and norms of international law signifies the constitutionalisation of the pacta sunt servanda principle and the state’s duty to respect her international obligations that she has freely conferred upon itself. Where a national act competes with an international treaty, the treaty must be applied.30 Lithuania adheres to a monist model where only ‘[i]nternational treaties ratified by [the Seimas] shall be a constituent part of [its] legal system’ (constituent part clause). The list includes treaties ‘on political co-­operation with foreign states’; ‘mutual assistance treaties’; ‘treaties of a defensive nature related to the defence of the State’; those ‘on the participation … in universal … and regional international organisations’; and ‘multilateral or long-term economic treaties’ (Art 138 LC).31 At one time32 the Constitutional Court construed the constituent part clause narrowly, thereby excluding direct application of criminal law norms of international treaties (in that case, ECHR): they purportedly could be applied only after being transposed into national laws.33 To the extent that this was applicable to other fields of law, the question could arise as to the direct applicability of EU public law provisions. EU law (or rather its predecessor EEC law) was perceived at that time as ‘ordinary’ international law (though the negotiations regarding the Association 30 LTCC 14.03.2006 (n 11). 31 See also LTCC 17.10.1995, No 8/95 (Law on International Treaties). 32 LTCC conclusion 24.01.1995 (n 18). 33 For the criticism of this doctrine see, eg, Vilenas Vadapalas, ‘Opinion of the Constitutional Court of the Republic of Lithuania in the case concerning the Conformity of the European Convention of Human Rights with the Constitution of Lithuania’, (1995) 55 Zeitschrift für ausländisches öffentlisches Recht und Völkerrecht 1077.

434  Egidijus Kūris Agreement had already begun).34 The post-accession doctrine repudiated the 1995 doctrine (see section III.C), and the Court’s subsequent practice developed in the direction of direct application of international treaties (including ECHR). But collisions indeed can arise between the Constitution and extra-national law. They have arisen with regard to several ECtHR judgments.35 They also allegedly have arisen with regard to bilateral treaties.36 As EU law is of an extra-national nature, its collisions with the Constitution also cannot be excluded (see section III.C).

3.  Amendment Procedure This procedure is not flexible. A motion to alter/supplement the Constitution may be submitted by not less than one-fourth of MPs (ie at least 36) or by not less than 300,000 voters.37 Amendments have to be passed by the Seimas voting twice, with a break of no less than three months; during each of them not less than two-thirds of all MPs have to vote in favour of the amendment (Arts 147, 148 LC). There are explicit exceptions to this general rule. The provision of Article 1 LC that ‘the State of Lithuania shall be an independent democratic republic’ may only be altered by referendum if not less than three-quarters of the citizens with the electoral right vote in favour thereof – a factual impossibility. The provisions of Chapter I ‘The State of Lithuania’ (on national sovereignty, independence, territorial integrity, supremacy of the Constitution, rule of law, referenda, citizenship, state language etc) may be altered only by referendum.38 This exception applies also to the rules determining the constitutional amendment procedure, laid out in Chapter XIV ‘Alteration of the Constitution’.39 Since 201440 the Constitutional Court has been developing a doctrine of limitations on the alteration of the Constitution (see also sections III.A and III.D). The 2014 doctrine (and its

34 Irmantas Jarukaitis, Europos Sąjunga ir Lietuvos Respublika: Konstituciniai narystės pagrindai [The European Union and the Republic of Lithuania: Constitutional Foundations of Membership] (Vilnius, Justitia, 2011) 485 et seq. 35 See, eg, ECtHR 06.01.2011, application No 34932/04 (Paksas v Lithuania) ECLI:CE:ECHR:2011:0106JUD003493204. Cf LTCC 25.05.2004, No 24/04 (Presidential elections); LTCC 05.09.2012, No 8/2012 (Prohibitions for person removed from office under the impeachment procedure); see also LTCC conclusion 31.03.2004 (n 20); LTTC 22.12.2016, No 7/2016 (Electoral rights of former President R Paksas). 36 Cf, eg, LTCC 04.11.2014, No KT47-S32/2014 (Treaty with Ukraine on social security), where the petition to examine a bilateral treaty was dismissed on formal grounds. 37 In 2014, a referendum was called to lower that figure to 100,000, but it failed, as less than 15% of voters took part in it. See section III.A. 38 Voter turnout and adoption thresholds for the referenda are not specified. 39 The thresholds for these referenda are also not explicitly indicated in the Constitution; they are to be set in a statute – ‘ordinary’ or ‘constitutional law’ (cf n 11). According to the Law on Referendum (not in force as of 01.07.2021; see n 11), for a referendum to be valid, more than 50% of all those eligible to vote must take part in it. In 2018 the adoption threshold for referenda on altering Art 12 LC and expanding the (currently very limited) possibilities for multiple citizenship was lowered by a statutory amendment, which was effectively vetoed by the President. New initiatives to disencumber the passing of constitutional amendments by referendum were undertaken, including the passing of the statutory amendments allowing for referenda to take place on more than one day. Those amendments aimed at stimulation of higher voter turnout in the forthcoming referendum on the constitutional amendment regarding multiple citizenship. The referendum was scheduled for two days with a two-week interval, coinciding with the two rounds in 2019 presidential elections. The LTCC found these statutory amendments and an accompanying Seimas resolution unconstitutional (LTCC 15.02.2019, No 18/2018 [Mandatory referendum on two days with a two-week break]). À propos, the 2003 referendum on Lithuania’s EU membership took place on two consecutive days (see section III.C). In line with the LTCC’s ruling, the 2019 referendum had to be conducted in one day. It failed, the proposed constitutional amendment having received the majority of votes, which still was not sufficient for its adoption. A year and a half later the LTCC found unconstitutional the whole Law on Referendum (LTCC 30.07.2020 (n 11)). 40 LTCC 24.01.2014, No 22/2013 (National Bank); and LTCC 11.07.2014, No 16/2014-29/2014 (Referenda).

Lithuania  435 follow-ups) built upon, systematised and expanded already existing fragmentary doctrinal provisions. It added new restrictions to those explicitly enshrined in the text of the Constitution and made certain formally amendable provisions effectively non-amendable. The doctrine postulates that the provisions of the Constitutional Law ‘On the State of Lithuania’ (Art 1;41 see section I.C.3) and the Constitutional Act on Non-Alignment (both containing no explicit amendment clauses) can be altered only by referendum, and that the highest threshold (three-quarters of all those entitled to vote) is required as for the alteration of Article 1 LC. Amendments ‘may not create any such new constitutional regulation under which one [provision] would deny or ­contradict … another provision … , so that it would be impossible to construe [them] as being in harmony’; also, ‘the … regulation established in the chapters and articles of the Constitution may not be opposed against [that which] is established in [its] constituent part’, which, as mentioned, includes the Constitutional Act on EU. Overall, the doctrine prohibits any such constitutional amendments that would deny ‘at least one of ’ such constitutional values as state independence, democracy, the republic,42 and the innate character of human rights and freedoms (because that ‘would amount to the denial of the essence of the Constitution itself ’). To bar any attempt to bypass the explicit limitations on constitutional amendments,43 the Court has interpreted that the Seimas was not permitted to make any such constitutional amendments that would deny the provisions of Chapters I and XIV; if the provisions of these chapters are not correspondingly amended, constitutional amendments contradicting these provisions are not permitted to be introduced also by referendum. The doctrine has severely restricted the alteration of the constitutional provisions pertaining to Lithuania’s international obligations. It is prohibited to make: (i) any such amendments to the provisions of the Constitutional Act on EU (which contain no explicit amendment provisions) which would deny Lithuania’s commitments arising from EU membership; and (ii) any such constitutional amendments that would deny Lithuania’s international obligations (such as those arising from membership in NATO, preconditioned by Lithuania’s geopolitical orientation) and at the same time the principle of pacta sunt servanda (as long as the said international obligations have not been renounced in accordance with the norms of international law). Constitutional amendments aiming at such denial should be barred, nipped in the bud.44 Articles 1 and 2 of the Constitutional Act on EU (see section III.C) may be altered only by referendum (no mention has been made of any threshold with regard to them). The above-mentioned substantive limitations on the alteration of the Constitution, set out in the 2014 doctrine (and reiterated and consolidated in later case law) are supported by the doctrine of procedural limitations, concentrating on the ‘equal applicability’ of the substantive limitations in the event of a referendum. Any such draft constitutional amendment that would disregard the substantive limitations cannot be put to a referendum. The citizens’ initiative group for a referendum can submit only such draft amendment that would observe the said limitations.

41 Art 1 LC essentially repeats Art 1 of that Constitutional Law, which establishes that the statement ‘the State of Lithuania shall be an independent democratic republic’ is ‘a constitutional norm of the Republic of Lithuania and a fundamental principle of the State’. 42 The question remains whether the republican form of government in principle could be changed. In LTCC 24.01.2014 (n 40) it is indicated as unamendable, but in LTCC ruling 11.07.2014 (n 40) the word ‘republic’ has been omitted in that context. The Constitution itself explicitly allows for the amendment of its Art 1 (Art 148 § 1). At the same time, the interpretation has been provided in LTCC 30.07.2020 (n 11) that the (historic) ‘fundamental constitutional acts of the State of Lithuania’, as ‘primary sources of Lithuanian constitutional law’, may ‘never be altered or repealed’ (see section I.C.3). 43 Such as the one regarding the 2018 statutory amendment lowering the adoption threshold for referenda on Art 12 LC (see n 39). 44 As in the case of the barring of a referendum aimed at preventing the introduction of the euro (see section III.D).

436  Egidijus Kūris If the group should disregard that duty, the Central Electoral Commission (CEC) must not register it. The Seimas, as the institution calling referenda, has a duty not to call a referendum where the proposed decision does not comply with the requirements stemming from the Constitution; its resolution on refusing to call a referendum must be adopted on the grounds provided for by law (which must stem from the Constitution). Such a resolution may be subjected to review by the Constitutional Court, should it be seized with the matter (via SAC or by means of a constitutional complaint, now available). Finally, the Constitutional Court has the exclusive competence to investigate and decide, whether an act adopted by referendum is not in conflict with the Constitution.45

II.  Constitutional Culture Until 1993 Lithuania never had an institution performing constitutional review.46 No official doctrine whereby constitutional provisions were construed was in place. The function of official interpretation of the Constitution (explicitly mentioned neither in the Constitution, nor in the Law on the Constitutional Court) was self-assumed by the Constitutional Court. Initially justices charily equated this function to a mere ‘explanation of ambiguities’ or ‘clarification of indistinctness’ of the Constitution’s text. The Court’s acts were held to be sources of law to the extent to which it acted as Kelsenian ‘negative legislator’. The reasoning underlying their operative parts was not spoken of as a source of constitutional law. This did not allow for ‘filling in the gaps’ where the Constitution’s text was silent. Building up its jurisprudence on the already accumulated case law, the Court has developed the doctrine of self-restraint, binding itself to follow its case law.47 It has affirmed that its acts in their entirety are sources of law,48 have the force of the Constitution and may not be overcome by a statute.49 Statutes whereby a regulation identical to one already found unconstitutional is reinstated, meet the fate of their predecessors.50 The Court has gradually solidified its position of not only a ‘negative legislator’, but also a provider of guidance for ‘positive’ law-making.51 It plays a broader role of a legal problem-solver, not only an ad hoc dispute-settler. 45 Strict as it was, the 2014 doctrine was made even stricter in LTCC 30.07.2020 (n 11). In addition to non-amendability of the the (historic) ‘fundamental constitutional acts of the State of Lithuania’ (‘primary sources of [its] constitutional law’) (see section I.C.1), it has been prohibited from amending most provisions of the Law on Entry into Force of the Constitution, namely those which consolidated the transitional constitutional regulation related to the entry into force and implementation of the Constitution and thus were ‘not of continuing validity’ (could no longer be applied after having been implemented). 46 Establishment of a Constitutional Court was contemplated as early as in 1938. The idea resurfaced on the eve of restoration of independence in 1990. The 1990 Provisional Fundamental Law, however, did not establish any system of judicial constitutional review. The Constitutional Court came into being in 1993, a few months after the 1992 Constitution came into force. 47 See, eg, LTCC 30.05.2003, No 21/2003 (Municipal elections); LTCC 28.03.2006, No 33/03 (Constitutional Court’s review of its own decisions); LTCC 05.09.2012 (n 35); LTCC decision 08.08.2006, No 34/03 (Dismissal of legal proceedings and reinterpretation of official constitutional doctrine). The stare decisis formula does not feature in the Court’s case law, but the idea is there. 48 Cf LTCC 30.05.2003 (n 47). 49 Not even by an ill-named ‘constitutional law’ (see n 11), and despite the wording, at the material time, of Art 72 of the Law on the Constitutional Court, which stipulated that the Court’s rulings have ‘the force of a law’. This clause was abolished in 2015. Cf n 62. 50 See, for the most recent example, LTCC 19.05.2021 (n 14). 51 Cf Allan Brewer-Carías, Constitutional Courts as Positive Legislators: A Comparative Study (Cambridge, Cambridge University Press, 2011). Recently, however, the Court pronounced itself as to what the constitutional amendment on the direct elections of mayors (see ns 13, 86 and 91) should contain (LTCC 19.04.2021 [n 13]). Self-restraint was put aside. This raised a controversy, especially in view of the Court’s earlier reserved stance, where it did not go further than

Lithuania  437 It was owing to the Constitutional Court that Lithuania underwent a fundamental constitutional law paradigm-shift, bearing on the perception of its relation to all other law, national and extra-national. In the new paradigm52 the Constitution is not reduced to and not identified with its text, but is seen as a normative reality. Its text is not its ideal appearance.53 Lithuania’s legal system is conceived as Constitution-centred. Constitutional law is no longer perceived as a ‘branch’ of law with a distinct ‘subject matter’ – it is distinguished from ordinary law by its supreme force, delineated from it not horizontally, but vertically. It is regarded as having no lacunae. Where the Constitution’s text is silent or limited to a blanket norm allowing for a certain area to be statute-regulated, the legislator’s discretion is restrained by the constitutional principles ensuring coherence of the entirety of constitutional regulation. To hold that the Constitution has gaps would amount to allowing for some fields of human activity to be regulated by ordinary law not subject to constitutional review, whereby its compliance with these highest principles could be tested. The Constitution is perceived as non-contradictory, even where its text is not. Being contradistinguished from all other law, its provisions are not stratified into ‘superior’ and ‘inferior’:54 they all have the same force and must be construed so that harmony of the entire constitutional regulation is ensured55 and provisions are not opposed against each other, which would distort the essence of some of them.56 Where the wording seems mutually exclusive, every effort should be put for construing the provisions so that the resulting doctrine is coherent. In the new paradigm the system of constitutional law’s sources has been reconstructed. All sub-constitutional sources (including ‘ordinary’ and ‘constitutional’ statutes), erstwhile held to belong to that system, have been excluded from it, and the Constitutional Court’s acts have been embraced as its part. The system has been reduced to the Constitution (the five acts comprising it57) to state that certain regulations could be introduced only by a constitutional amendment (cf LTCC 13.11.2006; case No 45/03-36/04 (Lithuanian citizenship); LTCC 05.09.2012 (n 35); LTCC 22.12.2016 (n 35); LTCC decision 20.10.2017, No KT14-S7/2017 (Lithuanian citizenship)). Another type of the Court’s forgetfulness of the virtue of self-restraint, increasingly more present in its case law, is the examination extra petita and the annulment of not challenged provisions, justified by the ratiocination that ‘if [the Court] finds the unconstitutionality of provisions whose compliance with the Constitution is not impugned by a petitioner but which are consolidated in the same legal act the constitutionality of whose other provisions is impugned … it must state that [those] provisions … are unconstitutional’, because ‘the implementation of constitutional justice implies that a legal act (part thereof) that is in conflict with the Constitution must be removed from the legal system’ (LTCC 03.06.2020, No 2/2019 (State pensions of scientists)). That applies also to ‘provisions that have not been impugned … but regulate essentially identical relationships and are consolidated in another legal act’ (LRCC 09.03.2020, case No 16/2019 (Judges’ immunity)). While, as a rule, the provisions examined extra petita have deserved, on their merits and in the light of the Court’s findings in the case under consideration, to be declared unconstitutional, the cited unqualified justification allows for unfettered judicial activism. 52 I dealt with it at length elsewhere. See, eg, Egidijus Kūris, ‘Konstitucinės justicijos proceso teisės klausimu’ [‘On the issue of the law in constitutional justice procedure’], (2011) 78 Teisė 7; Egidijus Kūris, ‘On Perception of Constitutional Law: A Theoretical Approach to Constitutional Justice’, in Gagik Harutyunyan (ed), New Millenium Constitutionalism: Paradigms of Reality and Challenges (Yerevan, NJHAR, 2013); Egidijus Kūris, ‘On Jurisprudential Constitution and European Integration’, in Iris Nguyen Duy et al. (eds), Uttensammenligning: Festskrift til Eivind Smith 70 år (Bergen, Fagbokforlaget, 2020). 53 One example: ‘The Constitutional Court shall consist of nine Justices, each appointed for a single nine-year term of office. Every three years, one-third of the Constitutional Court shall be reconstituted. The Seimas shall appoint three candidates for Justices of the Constitutional Court from the candidates submitted by the President of the Republic, the Speaker of the Seimas, and the President of the Supreme Court, and [the Seimas] shall appoint them as Justices.’ (Art 103 § 1 LC; the official English translation has omitted the words in square brackets). The readership may legitimately wonder: how many candidates must the three mentioned officials nominate, and does the Seimas appoint ‘candidates’ or ‘justices’? The rational and pragmatic construction of this clumsy wording, however, has been not difficult, as well as the ensuing practice (cf section I.A). 54 But note the most puzzling doctrinal novelty of ‘supra-constitutional force’ of ‘fundamental constitutional acts of the State of Lithuania’/‘primary sources of Lithuanian constitutional law’ (see section I.C.1). 55 LTCC 14.01.2002, No 25/01 (State and municipal budgets). 56 LTCC 24.12.2002 (n 11). 57 See section I.C.1. Again, note the doctrinal novelty of ‘supra-constitutional force’ of ‘fundamental constitutional acts of the State of Lithuania’/‘primary sources of Lithuanian constitutional law’.

438  Egidijus Kūris and the Court’s case law. In theory, it could include also constitutional custom or international treaties. But so far no constitutional custom has been unambiguously discerned, and no treaty has been ratified by a constitutional provision. The treaties ratified by a statute have the force of a statute and, thus, are not sources of constitutional law. The treaty by which Lithuania acceded to EU was ratified by a statute. The perception of constitutional law as the law of the Constitution, as construed by the Constitutional Court, bore on the perception of democracy, which is no longer reduced to its majoritarian component, but is seen as an ambivalent alliance of majoritarianism and antimajoritarianism. Their rivalrous conjuncture is given sense by the notion of constitutional democracy. Individual rights’ protection is an antimajoritarian doing; so is polity’s protection against irrational populism or elected autocracy. Yet, it would be wishful thinking to claim that constitutional democracy is adequately accepted at the grassroots level58 or that all of the political class has ungrudgingly accepted the Constitutional Court’s increased role in law-making (both ‘negative legislation’ and guidance of ‘positive’ legislation) and restriction of politics by the supreme law of the land.59 There have been occasional attempts to bypass the Court’s rulings by repeatedly adopting the same regulation or to thwart the enhancement of its role by blocking the introduction of individual complaint,60 publicly attacking the Court and its justices. The appointment procedure for justices has seen some attention shift from the career or just deserts of nominees to their views on the force of the Court’s role vis-à-vis political branches of government,61 which reveals the political class’s eagerness to impel the change of the official constitutional doctrine by occluding the path to the bench to aspirants professing Constitution-centred approach.62 The Court’s predictability (even if some unpredictability has also been there, and increasingly so) is lauded by the legal community, but decried by part of the political class, if what can be predicted mismatches their agenda. At times the hostility takes grotesque forms, as when the Court was requested to curb its authority by denouncing its status of a court.63 The new paradigm, with its emphasis on augmentation and advancement of constitutional law in the Court’s case law through constitutional adjudication, presupposes a living Constitution, jurisprudential law – akin to the characterisation of ECHR as a ‘living instrument’ in ECtHR case law. 58 Especially given the almost overwhelming distrust in almost all public institutions in Lithuania. Public confidence in the Seimas, the Government, the courts and the political parties almost never exceeds 25%. Confidence in the President is usually higher, as well as in the Constitutional Court where it fluctuates around 40%, sometimes reaching as high as 60%. 59 Egidijus Kūris, ‘Standards of the Rule of Law’, in Egidijus Kūris (ed), Crisis, the Rule of Law and Human Rights in Lithuania (Šiauliai, Titnagas, 2015) 23, 51–55. 60 For over a decade – see n 15. 61 This is not to say that the personal qualities of the candidates are not a factor in the process of appointment of justices. That process in 2020–21 involved orchestrated personal attacks (some indeed gratuitous) on the candidates; so did the procedure of appointment of the new President of the Court in 2021. 62 In 2008 one of the nominees to the Court was asked at the Seimas hearing of his opinion of the legal force of the Court’s acts. The Seimas seemed to be satisfied with the frugal answer that, under Art 72 of the Law on the Constitutional Court, they have ‘the force of a law’ (the clause was abolished in 2015). See n 49. 63 A group of MPs requested the Court to examine the constitutionality of the statutory definition of the Court as a ‘judicial institution’ and a ‘free and independent court … implement[ing] judicial power’. They gave prominence to the fact that the LTCC and other courts are dealt with in separate chapters of the Constitution. Accordingly, the LTCC, being not a ‘court’ and being not mentioned in Art 5 LC (which provides that ‘State power shall be executed by the Seimas, the President of the Republic and the Government, and the Judiciary’, the latter word (Teismas) in Lithuanian being the same as ‘court’), does not execute state power. Its construction of the Constitution is therefore not binding. The claim was rebutted. One fragment of the Court’s reasoning reasoning was that ‘[t]he presumption … that the Constitutional Court is not a court and does not implement State power is utterly irrational [and] strikes the raison d’être of the petition, [since if it] is not a court and does not implement State power, it is not comprehensible why the petitioner has applied namely to this court, requesting to investigate whether a legal act, passed by [the Seimas] – one of the institutions implementing State power (in this case – legislative power) is not in conflict with the Constitution’. See LTCC 06.06.2006, No 12/06 (Status of the Constitutional Court).

Lithuania  439 The constitutional order is perceived as constantly developing and self-generating due to its case-by-case reinterpretation. Living Constitution does not allow for subdual of constitutional principles to political/economic conjuncture or societal pressure. Rather these principles, when they confront with developing reality, mutate, without losing their core, and are able to meet new challenges, to which the society is exposed. Politics remains Constitution-restricted. The Constitution neither shirtfronts politics, nor succumbs to it, but reacts to what cannot be prevented or changed by law alone, and regenerates accordingly. Where that is not possible, there remains the last resort of a constitutional amendment. The paradigm-shift dealt with here coincided in time with Lithuania’s aspirations to join the EU. The Constitution, an irreplaceable centre of the domestic legal system, would have done a disservice to Lithuania’s polity, had it appeared unable to self-adjust to these aspirations. But the popular support for EU accession had its own strange bedfellow – the negation of the Constitution’s ability to conform to EU membership. This contrariness merits a closer look.

III.  Constitutional Foundations of EU Membership and Closely Related Instruments Lithuania’s EU accession on 1 May 200464 was not fluent, from a constitutionalist perspective; nor was her first decade of EU membership. In-built constitutional roadblocks had to be removed by constitutional amendments and doctrinal polishing. One amendment was introduced post factum. The EU could not care less for that domestic law-making blunder, but the constitutionalist community had many grounds to be disconcerted. Later, new boulders were rolled in the way to aggravate Lithuania’s EU integration, and had to be dislodged by invoking the Constitutional Court’s problem-solving capacity. They (in chronological order) pertained to land ownership, local self-government, the status of EU law, and the national bank.65

A.  Land Ownership Article 47 LC initially did not provide for foreign ownership of land, internal waters, forests and parks. These could be owned only by the State of Lithuania or her citizens66 – an impediment to EU accession. The state also had exclusive ownership rights to subsurface, roads, historical, archaeological and cultural objects of state importance.67 64 Here is the concise timeline: On 27.08.1991 the European Community recognised Lithuania’s independence. After a series of steps of economic cooperation and political dialogue, Lithuania, as a state aspiring to EU membership, took upon itself an obligation to accept the acquis communautaire. On 18.07.1994 the EU-Lithuania Free Trade Agreement was signed. On 16.12.1994 negotiations regarding the Association Agreement began, to be completed in April 1995; on 12.06.1995 that Agreement was signed. On 08.12.1995 Lithuania submitted an EU membership request. Lithuania was invited to start accession negotiations in 1999; these were completed in December 2002. The Accession Treaty was signed on 16.04.2003. The majority of citizens approved Lithuania’s EU membership in the referendum of 10–11.05.2003. Lithuania (and nine other states) became an EU Member State on 01.05.2004. In 2004 the Seimas was the first parliament to ratify the abortive TECE, a few days after it was signed. It ratified the Treaty of Lisbon in 2008. 65 I dealt with these issues in Kūris, On ‘Jurisprudential Constitution’ (n 52), from which fragments are borrowed. 66 This barrier applied also to enterprises of Lithuanian origin. One exception regarding foreign institutions was allowed: plots of land could belong by right of ownership to a foreign state for the establishment of diplomatic missions and consular posts. It also was stipulated that Lithuania shall have the exclusive rights to the airspace over its territory, its continental shelf and the economic zone in the Baltic Sea. 67 Art 47 LC (and several other provisions) could be altered in an easier procedure within a year after the adoption of the Constitution – a possibility never made use of.

440  Egidijus Kūris In 199668 the Seimas eased these restrictions by allowing ‘foreign subjects’ (enterprises etc69), ‘engaged in economic activity in Lithuania’, to acquire ‘non-agricultural land, necessary for construction and operation of the buildings and facilities designated for the purpose of their direct activities’.70 There was a caveat: ‘foreign subjects’ had to meet ‘the criteria of European and transatlantic integration’, as defined in the ‘constitutional law’.71 The ‘constitutional law’, defining the said geopolitical criteria and establishing the order of land acquisition, was adopted on the same day. It was not entered in the ‘the list of constitutional laws’ (non-existent at that time72), despite the doctrine requiring that ‘constitutional laws’ must be entered into that list before their adoption.73 In 2001 the Constitutional Court saved the day by expounding that that requirement was not applicable to the ‘constitutional laws’ explicitly indicated in the Constitution.74 The Seimas has no discretion to adopt them either as ‘constitutional’ or as ‘ordinary’ laws, or not to adopt them at all: it must adopt them. The 1996 amendment did not suffice for fully fledged EU membership. Later that year, a referendum was held on, inter alia, alteration of Article 47 LC aimed at permitting less restrictive land acquisition by foreigners. It failed, as the voter turnout threshold was not met (over 17 per cent of all eligible voters voted ‘yes’, less than 16 per cent ‘no’). At that time public support for EU membership did not exceed 50 per cent. Article 47 LC was amended anew by the Seimas in 2003,75 after EU accession negotiations had been completed and pending the signing of the Accession Treaty. The state reserved for itself the right to exclusive ownership of subsurface, internal waters, forests, parks, roads, historical, archaeological and cultural objects of state importance,76 allowing ‘foreign entities’ to ‘acquire ownership of land, internal waters and forests according to a constitutional law’.77 The new ‘constitutional law’ established a seven-year transition period, during which the restriction on agricultural land acquisition was applied to citizens of other EU Member States. That period had to end on 1 May 2011. The Ministry of Agriculture initiated its extension, alleging that foreigners, once given a chance, would ‘buy plots of land for half-price’, still high for Lithuanians. The Government began negotiating the extension of the transition period with the Commission, which on 14 April 2011 agreed to its extension for three more years.78 Fuddled by this bargaining victory, the Government did not present any amendments to the ‘constitutional law’ in force.79 Thus from 1 May 2011 foreign EU citizens were ex lege entitled to acquisition of agricultural land, notwithstanding the agreement with the Commission regarding the extension of the transition period till 1 May 2014. 68 Amendment of 20.06.1996, in effect as of 21.07.1996. 69 In later translations ‘entities’. The awkward term ‘subjects’ (subjektai) includes both physical and legal persons. 70 Land ownership was also allowed to municipalities and ‘other national subjects’. 71 Thus, Russian or CIS ‘subjects’ and those from non-EU or non-NATO states were excluded. The explicit mentioning of ‘European and transatlantic integration’ signified the positive dimension of the constitutional principle of geopolitical orientation, in addition to the negative one (see section I.C.2). 72 Adopted in 2012. The enlisted ‘constitutional laws’ still await their adoption. See n 11. 73 LTCC 08.11.1993 (n 11). 74 LTCC 02.04.2001, No 18/99 (Restoration of land ownership rights). 75 Amendment of 23.01.2003, in effect as of 24.02.2003. 76 The provisos remain regarding Lithuania’s exclusive rights to airspace, continental shelf and sea economic zone and regarding foreign land ownership for the establishment of diplomatic missions and consular posts (see n 66). 77 Thus, the ‘positive’ aspect of geopolitical orientation (see n 71) has disappeared from this wording. It was restored, in another form (by mentioning only the EU), in 2004, in the Constitutional Act on EU (see section III.C). In 2011, the Constitutional Court restored, doctrinally, the ‘transatlantic’ element of the said principle (see LTCC 07.07.2011, No 22/2008-31/2008-9/2010-35/2010 (State and official secrets)). 78 2011/240/EU: Commission Decision of 14 April 2011 extending the transitional period concerning the acquisition of agricultural land in Lithuania, [2011] OJ L 101/122 f. 79 A conspiracy theory circulated that the failure to present the amendment was a deliberate choice of the liberal Minister of Justice, who strongly opposed the extension of the transition period.

Lithuania  441 There was more to that.80 On 5 May 2011 the Chamber of Notaries81 requested the legal situation to be clarified by the National Land Service (under the Ministry of Agriculture), the Ministry of Justice and the European Law Department (ETD; under the latter Ministry). Where you stand depends on where you sit: the latter two bodies opined that the restriction set forth in the ‘constitutional law’ no longer applied, for the extension of the transition period was not incorporated in that statute; the National Land Service, on its part, argued for the direct application of the Commission’s decision, which allegedly rendered that restriction no longer applicable. The Chamber of Notaries forwarded the puzzle to the Seimas Committee on Legal Affairs, which limited itself to observing that the ‘the opinions of … institutions had diverged’ and forwarded the correspondence to the PM. In the subsequent correspondence the Ministry of Agriculture asserted that the Commission’s decision needed no transposition to national law, which amounted to holding that domestic statutes’ validity could be discontinued by the Commission, as if it was the national legislature, and – no less paradoxically – that although domestic law allowed for agricultural land acquisition by foreign EU citizens upon expiration of the seven-year period, it was prohibited for three more years by the Commission, which thus restricted the application of an underlying common market principle in one Member State. Then why on earth was that ‘constitutional law’ needed at all, if the seven-year transition period was allowed by the Commission prior to Lithuania’s EU accession? To confuse things more, on 30 June 2011 the Seimas adopted a resolution stating that the transitional period had been extended by the Commission and forbidding notaries to authorise transactions involving ‘foreign subjects’ regarding agricultural land and (sic!) forest land. But nothing was ever agreed with the Commission regarding extension of the transition period for acquisition of forest land!82 Such ambiguity could not last. On 3 November 2011 the Seimas amended the ‘constitutional law’ by extending, for three more years, the transition period for the acquisition of agricultural land, but made no mention of forest land. As the end of the extended transition period was approaching, a group of citizens, concerned that foreigners would ‘cheaply engross our land’, accomplished the impossible – to collect over 300,000 signatures (in a country of 3 million) to call a referendum on de novo amending Article 47 LC, reinstating its original (1992) wording.83 A powerful campaign was mounted to dissuade citizens to take part in it or, alternatively, to vote ‘no’. The referendum was held on 29 June 2014; 70 per cent voted ‘yes’, 26 per cent ‘no’. As less than 15 per cent of voters took part, the referendum failed. Had it succeeded, a confrontation between Lithuania’s Constitution and her EU membership commitments would have been forced by a popular vote. It was in this context that the Constitutional Court reinforced the doctrine of substantive limitations on the alteration of the Constitution, which stemmed ‘from the entirety of the constitutional regulation’ and did not allow for a constitutional amendment, if it ‘[destroyed] the harmony of constitutional regulation as integrity’84 (see section I.C.3). The ruling of 24 January 2014 was adopted while the collection of signatures for the land ownership referendum was ongoing, but was not directly related to it (it concerned the national bank; see section III.D).

80 See Egidijus Kūris, ‘Apie nepagarbą teisei’ [‘On the disrespect for law’], (2012) 13 Notariatas. 81 Notaries are charged with authorisation of land transactions. 82 In 2009 the Seimas suggested that the Government should negotiate with the Commission and ‘persuade’ it to extend the transition period regarding restriction on acquisition of agricultural and (sic) forest land until 2013. That non-binding resolution was never implemented. 83 In order to attract more supporters it was proposed also to amend the Constitution by lowering the threshold of 300,000 signatures required for holding a referendum down to 100,000. 84 LTCC 24.01.2014 (n 40).

442  Egidijus Kūris

B.  Local Self-Government Compared to the constitutional and para-constitutional battles over land ownership, the 2002 constitutional amendment of Article 119 LC85 regarding local self-government86 was not controversial. Until then municipal councils were elected by Lithuanian citizens, permanently residing in respective municipalities; those eligible to stand for election had to be Lithuanian citizens too, even if not necessarily residents of respective administrative units. The amendment, adopted in anticipation of-EU membership, expanded the electorate – it dropped down the citizenship requirement both for those electing and those standing for elections,87 provided that they permanently resided in respective municipalities. Electoral right on the local level was thus granted not only to non-Lithuanian EU citizens permanently residing in respective municipalities, but also to citizens of non-EU states and even stateless persons. The coming into force of the amendment was flawed,88 but this was immaterial to Lithuania’s EU accession. It was applied not to the immediate 2002 municipal elections, but to those of 2007. Non-Lithuanian EU citizens’ right to stand for elections was nominal, as the competition was allowed between party lists, while foreigners (even permanent residents) had no right to be party members. The provisions of the statute applicable at the 2007 elections were quashed (on several grounds) by the Constitutional Court.89 After a series of legislative ups-and-downs,90 in June 2021 the restriction for citizens of other EU Member States residing in Lithuania to be founders/ members of political parties was lifted, so now they can compete for seats in municipal councils, although not in the elections for mayor’s office.91

C.  The Status of EU Law As the date of Lithuania’s accession to the EU approached, the debate as to whether the Constitution allowed that accession became heated. It concentrated on the topic of Member States’ giving up part of their sovereignty. EU sceptics (but also those who supported EU accession, but thought it to be not in line with the Constitution unless the latter was not amended) pointed to such provisions as ‘[t]he State of Lithuania shall be an independent democratic republic’; ‘[s]overeignty shall belong to the Nation’; ‘[n]o one may restrict or limit the sovereignty of the Nation or make

85 Amendment of 20.06.2002, in effect (with exception) as of 21.07.2002. 86 Communities of administrative units exercise their self-government through municipal councils (Art 119 LC). In 1997 their term of office was prolonged from two to three years, and in 2002 to four years. In 2015 direct elections of mayors (as presidents of municipal councils) were introduced by statute, which in 2021 was found to have been adopted by bypassing the requisite constitutional amendment (see n 13, n 51 and n 91). 87 Citizenship is still mentioned, but that mention serves no practical purpose: ‘The members of municipal councils shall be elected … from among citizens of the Republic of Lithuania and other permanent residents of the administrative unit by the citizens of the Republic of Lithuania and other permanent residents of the administrative unit’. 88 The ‘constitutional law’ which established the order of application of the amendment was recognised as unconstitutional by LTCC 24.12.2002 (n 11). 89 LTCC 09.02.2007, No 6/07 (Elections to municipal councils). It was adopted a few days before the elections, which, following the urging of the Court, were not postponed. 90 The Seimas amended the electoral law before the 2011 elections, but its provisions pertaining to non-party list participation were again quashed by the LTCC 11.05.2011, No 136/2010-17/2011-18/2011-19/2011 (Elections to municipal councils). Relevant statutory amendments were introduced in 2014, however, several provisions of the law on municipal elections were found unconstitutional by LTCC ruling 19.04.2021 (n 13); see also nn 51, 86 and 91. 91 Major uncertainties have arisen owing to the invalidation, by the LTCC, of the legislation which introduced direct elections of mayors (see nn 13, 51 and 86). At the time of writing it is not clear whether, following the prospective constitutional amendment, mayors or heads of municipal councils will be the highest local executives.

Lithuania  443 claims to the sovereign powers belonging to the entire Nation’; ‘[t]he Nation shall execute her supreme sovereign power either directly or through her democratically elected representatives’ (Arts 1–4 LC), which were claimed to bar the accession. They can be altered only by referendum, any change of Article 1 LC being unthinkable in practice (see section I.C.3). Other articles, too, were invoked, such as those entrenching Lithuanian as state language, foundations of foreign policy, the legislative process, the budgetary process, and, most importantly, the supremacy of the Constitution: ‘[a]ny law or other act that contradicts the Constitution shall be invalid’ (Art 7 LC). In response, academic publications (in law, economics, political science, history, and philosophy) put forward the pragmatic arguments as to the advantages of EU membership, but also that nowadays the notion of sovereignty was no longer confined in its Bodinian sense, but was more nuanced, in view of globalisation and states’ interdependence. The view prevailed that Lithuania could not accede to the EU without amending her Constitution. Indeed, ‘[a]ny law or other act that contradicts the Constitution shall be invalid’ is not a provision that can be legitimately rebutted by referring to states’ factual interdependence. From 1998 on, several constitutional amendments were drafted and put aside. Intervening in Chapter I, alterable only by referendum, was forgone because of trepidation that the referendum might not succeed and because it was not agreed what provisions had to be put to it, which could outweigh the political and emotional weight of the notions of ‘sovereignty’ and ‘independence’, employed in the most ‘resisting’ Articles 1 to 4 LC. The idea of a new Chapter of the Constitution on EU membership would have required renumbering of Chapter XIV, also amendable only by referendum; this was rejected too. In the end, the idea of a referendum on constitutional amendment(s) was abandoned. Instead, a referendum on approval of EU membership was held on 10–11 May 2003. Fearing a low turnout, it was extended for two days.92 Almost 90 per cent of those who took part, over 63 per cent of all eligible voters, voted in favour of EU accession, over 8 per cent against.93 The nascent EU membership was legitimised and the path for the constitutional amendment to be passed by the Seimas was cleared, whatever that amendment was to be. However, there was no time left for a proper amendment procedure. On 1 May 2004, accession day, when EU law became applicable in Lithuania, no EU-related constitutional amendment was yet in place. The Seimas passed it on 13 July 2004 by adopting a separate Constitutional Act on the EU,94 which came into force on 14 August 2004, three and a half months after the EU membership had become fait accompli.95 For more than three months the state was in the EU, having ‘smuggled’ itself in without being permitted to do so by her own Constitution. For if a

92 This was based on a special amendment of the Law on Referendum, and it worked. In 2019 the legislative trick of tailoring the number of days to a particular referendum did not work so well (see n 39). Another amendment, of the Law on the Central Electoral Commission, allowed for the printing of ballots for referenda ‘pertaining to Lithuania’s membership in international organisations’ not only in Lithuanian, but, in areas where ethnic minorities constituted ‘a significant part of the population’, also in Russian and Polish. That amendment, undisguisedly tailored to attracting ethnic minorities’ votes, later was quashed as unconstitutional, but the Constitutional Court stated that the results of 2003 referendum could not be challenged on that ground (LTCC 10.05.2006, No 25/03 (Referendum ballot-papers)). 93 The referendum was marred not only by printed ballots’ deviation from state language requirements (see n 92), but also by the last-day agitation on national TV (otherwise forbidden) and business incitement to participation by reducing prices for certain goods for those who presented a sticker proving that s/he did vote. 94 Also entered in Art 150 LC as the ‘constituent part thereof ’ (see section I.C.1). 95 On the accession day Lithuania was ‘the only Baltic State where the necessary constitutional changes in terms of EU accession have … not yet been enacted’ (Jenő Czuczai, ‘Constitutional Preparation for EU Accession in the New Central and Eastern European Member States: Is the Rule of Law Better than the Rule of Politics?’, in Jaap W. de Zwaan et al (eds), The European Union: An Ongoing Process of Integration. Liber Amicorum Alfred E. Kellermann (The Hague, TMC Asser Press, 2004) 276). The delay cannot be justified, but can be explained by the fact that the country was gripped by the President’s impeachment, which ended on 06.04.2004.

444  Egidijus Kūris constitutional amendment was not required for the accession, why then was it later nevertheless adopted? That period presented also constitutional litigation-related risks. For example, had a case regarding the status of an EU legal act substituting for a national statute been brought to the Constitutional Court, the Constitution would not necessarily have been interpreted as granting (during these three and a half months) EU law the formal status which it had under the Accession Treaty. In 2006 the Constitutional Court acknowledged that by the Constitutional Act on EU, Lithuania’s EU membership was constitutionally ‘confirmed’,96 thus indirectly barring the raising of the issue of its constitutionality. The Constitutional Act refers to the Lithuanian citizens’ will expressed in the referendum and to the Accession Treaty. It declares the Seimas’ aspiration to ensure Lithuania’s fully fledged participation in European integration, as well as her security and her citizens’ welfare. Two provisos are meant to allay the misgivings of part of the society regarding EU membership, expressing the Seimas’ conviction that EU respects: (i) human rights and fundamental freedoms;97 and (ii) Member States’ national identity and constitutional traditions. As long as this is so, EU membership is not seen as undermining national sovereignty. The normative part of the Constitutional Act on EU consists of four sets of provisions. Firstly, Lithuania as an EU Member State shall share with or confer on the EU the competences of her state institutions in the areas provided for in the EU founding Treaties and to the extent she, together with other Member States, would jointly meet her membership commitments in those areas as well as enjoy membership rights (Article 1). Secondly, the norms of EU law shall be a constituent part of the legal system of Lithuania. Where the EU founding Treaties are concerned, they shall be applied directly. In the event of collision of legal norms they shall have supremacy over the laws and other legal acts of Lithuania (Article 2). Thirdly, the Government shall inform the Seimas about EU law legislative proposals. Where such proposals seek to regulate areas which, under Lithuania’s Constitution, are related to the competences of the Seimas, the Government shall consult the latter. The Seimas may recommend to the Government a position of Lithuania in respect of these proposals. The Seimas Committees on European Affairs and on Foreign Affairs may, according to the procedure established by the Seimas Statute (ie Rules), submit to the Government the opinion of the Seimas concerning the legislative proposals. The Government shall assess the recommendations or opinions submitted by the Seimas or its Committees and shall inform the Seimas about their execution following the procedure established by legal acts (Article 3). Fourthly, the Government shall consider EU law legislative proposals following the procedure established by domestic legal acts; it may adopt decisions or resolutions for the adoption of which ‘the provisions of Article 95 of the Constitution are not applicable’ (Article 4; see section V). Article 2 LC is central to understanding the Constitution’s approach to the place of EU law in Lithuania’s legal system. EU law, though ‘extra-national’ owing to originating beyond Lithuanian institutions, is a constituent part of Lithuania’s legal system – but not termlessly. The word ‘supremacy’ used in Article 2 does not lend itself to a forthright recognition that any EU law prevails over any national legal act, especially in the context of the proviso that the EU respects its Member States’ national identity and constitutional traditions and the provision of Article 7 LC that any law or other act that contradicts the Constitution shall be invalid. As a constituent part of the Constitution, the Act is subject to judicial interpretation. That is in place since 2006:98 96 LTCC 14.03.2006 (n 11). Prepositions matter: the membership was ‘confirmed’ or ‘approved’ (patvirtinta), not ‘endorsed’ (įtvirtinta). 97 And that Lithuania’s EU membership will contribute to a more efficient securing of human rights and freedoms. 98 LTCC 14.03.2006 (n 11).

Lithuania  445 the Constitution establishes expressis verbis the ‘collision rule’, consolidating priority of application of EU legal acts, where the provisions of EU law arising from EU founding Treaties compete with the regulation established in domestic legal acts (regardless of their legal force), save the Constitution itself. This doctrine escapes the head-on confrontation between Lithuanian and EU law in terms of the supremacy of either of them. This model has already been established in the Law on International Treaties: if a treaty and a domestic legal act compete, the treaty shall be applied. With regard to EU law, instead of the supremacy of legal force criterion that of priority of application is invoked: priority is granted to the EU legal act, provided that it ‘concerns’ EU founding Treaties.99 This doctrine has abandoned that of 1995, which restricted direct application of extra-national public law (see section I.C.2). The Constitutional Court has not yet given a clear pronouncement as to what extent the old doctrine has been abandoned, but the fact that no exceptions to the collision rule have been formulated portends that with regard to EU law there are none100 (they may still be maintained to exist with regard to non-EU-originating criminal law). Overall, the interpretation of national law is EU law-friendly. The Court has postulated that the jurisprudence of the Court of Justice of the European Communities/Union (CJEU) and the General Court is a source of interpretation of Lithuanian constitutional law – though not part of that law itself.101 It is the law of Lithuania, even if not the ‘national law’. The Constitutional Court has affirmed, that ‘the constitutional imperative of full participation by … Lithuania in the EU … implies [Lithuania’s] constitutional obligation … to properly implement the requirements of EU law’.102 Whether the doctrine of primacy of application (and not superiority) of EU law is satisfactory from an EU law perspective is another matter.

D.  National Bank Lithuania’s national bank is the Bank of Lithuania (BL). A constitutional amendment aimed at Lithuania’s entering into the EU monetary union was not expected to be controversial. It was – years later having become a matter of constitutional litigation. Article 125 LC in its original wording provided, inter alia, that BL shall have the sole right to issue currency103 (litas, from 1993). That article was amended in 2006 by deleting the sole right clause.104 The amendment was actuated in anticipation of the introduction of the euro in 2007, so the formal obstacle had to be removed. The euro was eventually introduced in 2015. More than seven years later the Seimas in corpore requested the Constitutional Court to examine the constitutionality of the procedure of the 2006 amendment on the ground that the amendment presented for voting in the Seimas had been essentially transformed, as compared 99 Cf Matej Avbelj, ‘Supremacy or Primacy of EU Law – (Why) Does It Matter?’, (2011) 17 European Law Journal 744. 100 Having been on the bench at the time when the collision clause was interpreted, the author can testify that this was the intention. But in constitutional interpretation the ‘I remember’-type of evidence is not valid, and even its very invocation is quite limited owing to the secrecy of deliberations. 101 LTCC 21.12.2006, No 30/03 (National broadcaster). This was postulated long before that with regard to ECtHR case law. 102 See, among many authorities, LTCC 24.01.2014 (n 40); LTCC decision 16.05.2016, No KT16-S9/2016 (Other ­remuneration of judges); and especially LTTC 11.01.2019, No 16/2016 (Family reunification). 103 It also stipulated that the procedure for the organisation and activities of BL as well as its powers shall be established by law. In 2006 this provision was amended by providing that also the legal status of the Chair of its Board, as well as the grounds of his/her dismissal shall be established by law. This addition was meant to consolidate the position of that official. This was done together with the deletion of the sole right clause. 104 Amendment of 25.04.2006, in effect as of 26.05.2006.

446  Egidijus Kūris to that which had been introduced by the bill-initiating MPs, and was presented to the Seimas not by its authors but by a parliamentary committee (not entitled to that motion105). Behind that petition an expectation lurked that the Court, in view of the introduction of the euro scheduled for 1 January 2015, would not quash the amendment. It did.106 The text of its ruling is loaded with considerations as to the status of BL as an integral part of the European System of Central Banks. But, even having found that the impugned amendment procedure had been unconstitutional, the Court did not restore the validity of the earlier wording of Article 125 LC (which would have run against Lithuania’s EU commitments, constitutionalised by the Constitutional Act on EU), thus enabling the euro to be introduced in the scheduled time without a new amendment, which would have risked being the subject of a mercato delle vacche. It surely would have become the subject of trade-offs, if not payoffs. In parallel to the land ownership referendum (see section III.A), another referendum was initiated to amend Article 125 LC by reinstating the LB with the sole right to issue currency, retaining the litas and forbidding the introduction of another currency without a referendum. Having interpreted the applicable laws in the light of the Constitutional Court’s ruling in the Article 125 LC amendment case, the CEC refused to register the citizen’s initiative group for that referendum, barring it from legally meaningful signature-collection. The dispute regarding the powers of CEC and the Seimas to preclude referenda on constitutional amendment initiatives which allegedly disregarded constitutional requirements was brought to the Constitutional Court.107 The Court found it unconstitutional that, inter alia, the Seimas was obliged, upon collection of 300,000 signatures, to call a referendum in which what was proposed to be put to it did not meet the constitutional requirements.108 Its reasoning expanded on both substantive and (for the first time) procedural limitations on the alteration of the Constitution. Having precluded the euro referendum, the Court’s ruling facilitated the implementation of Lithuania’s commitment to enter the European Monetary Union.

E.  Further Integration Outside the EU Legal Order The euro was introduced on 1 January 2015 after the end of the legislative and jurisprudential zigzags around the LB and the Lithuanian national currency versus the euro (see section III.D). The question whether new constitutional amendments were necessary for further Lithuanian EU integration was raised in the context of the preparation and ratification of the Fiscal Compact, signed on 2 March 2012 and ratified by Lithuania on 28 June 2012. After being ratified by 12 eurozone Member States it came into force on 1 January 2013. Lithuania was not bound by the Compact before entering the eurozone, but supported the affirmation both in EU law and national law of the duty to conduct a responsible and sustainable fiscal policy and even contemplated an inclusion in her Constitution of a balanced and/or surplus budget clause. At that time, the Compact’s constitutionality was not debated; its goal was seen as part of the strict austerity policy during the economic crisis of these years,109 ie not as a ‘union’ but as financial stability, 105 The request was prompted by the controversy regarding the procedure of unrelated constitutional amendment (pertaining to the execution of ECtHR judgment in Paksas (n 34)), wherein the bill-initiating MPs were later substituted by a parliamentary committee, who had transformed the text of the initial motion. 106 LTCC 24.01.2014 (n 40). See also sections I.C.III and III.A. 107 The case was initiated by the SAC, to which the signature-collection initiative group applied, and by the Seimas in corpore. 108 LTCC 11.07.2014 (n 40). 109 Cf Mindaugas Lukas and Aistė Medelienė, ‘Public Finances’, in Egidijus Kūris (ed), Crisis, the Rule of Law and Human Rights in Lithuania (Šiauliai, Titnagas, 2015) 107 et seq.

Lithuania  447 not encroaching per se on sovereignty.110 Besides that, certain financial stability clauses have been long ago included in domestic legislation. The issue of constitutionality of the Compact was raised by Eurosceptic politicians shortly before it was to be signed. A protocol decision was adopted, linking implementation of the Compact with unrelated conditions.111 Some analysts assessed its legal significance as ‘next to nothing’.112 What attracts constitutionalists’ eyes is the legal force of the statutes by which the Compact was ratified and implemented. From the outset, there was pressure to constitutionalise the Compact, so that the issue of its constitutionality could not arise. For that purpose, the Ministry of Foreign Affairs suggested to ratify the Compact not by an ‘ordinary’ but by a ‘constitutional law’. The forewarnings by law experts113 that ‘constitutional laws’ were subject to constitutional review (see section I.B) were not taken into account;114 neither were the proposals to seize the Constitutional Court, in a preliminary review procedure, to examine the Compact’s constitutionality, in order to identify constitutional provisions possibly in need of amendment before ratification. The Compact was ratified in an expedited procedure by an ‘ordinary’ law and implemented by a ‘constitutional law’, in which the balanced budget rule was reproduced.115 Until the Constitutional Court examines these statutes, doubts as to the constitutionality of the Compact may be raised, but no such disputes have been voiced so far. On 18 January 2014, not long before the introduction of the euro, Lithuania ratified the Treaty establishing the European Stability Mechanism (the amendments to the related Article 136 TFEU were already ratified in 2012). This measure has not prompted any legal debate.116

IV.  Constitutional Limits to EU Integration As has been shown, the constitutional foundations of Lithuania’s EU membership were laid by sequential removal of constitutional barriers that were present in the Constitution, or that attempts had been made to be imported by anti-integration initiatives. Whatever Constitutionrelated roadblocks were piled on that path, they were pulled out by (i) application of ordinary law (as CEC’s barring the referendum aimed at preventing the introduction of the euro); (ii) constitutional amendments (1996–2004); or (iii) the Constitutional Court’s interpretation (also ex propriu

110 Loreta Šaltinytė, ‘Lithuania’, in Constitutional Change through Euro Crisis Law: A Multi-Level Legal Analysis of Economic and Monetary Union, https://eurocrisislaw.eui.eu/wp-content/uploads/sites/55/2019/05/Lithuania.pdf. 111 That Lithuania receives the same structural funding for agriculture as that received by EU Member States which acceded to the EU earlier; that Lithuania’s funding from the Cohesion Fund is not reduced; and that the EU ensures the funding necessary for the closure of the Ignalina nuclear plant. 112 Šaltinytė (n 110). 113 Including the author. 114 Šaltinytė (n 110). 115 It prescribes that ‘[e]ach budget attributable to the general government sector, with the exception of the budget of the State Social Insurance Fund of [Lithuania], the State budget and the budget the planned appropriations which do not exceed 0.3% of GDP in the preceding year at current prices, must be planned, approved, amended and implemented to be in surplus or balanced when judged by its structural balance indicator calculated on accrual basis’ (stepped into force on 01.01.2018), and that ‘[e]ach budget attributable to the general government sector the planned appropriations of which do not exceed 0.3% of GDP in the preceding year at current prices shall be planned, approved, amended and implemented in such a way that the appropriations of the budget would not exceed its revenue, with the exception of the year when a negative output gap is projected according to the economic development scenario which is made public by the Government or its authorized institution and in regard to which the monitoring authority published its conclusion. In the latter case, the appropriations may not exceed revenue by more than 1.5%’ (stepped into force on 01.01.2016) (Art 4). 116 Save the far-from-constitutional considerations as to whether Lithuania’s contribution was not too burdensome for the country’s economy. Lithuania’s five-year contribution was 327.2 m euros.

448  Egidijus Kūris moto) of EU-related constitutional provisions. The role of the Court in removing obstacles on the integration’s path was not limited to the settling of disputes at hand. It was also an active problemsolver both in a posteriori rectification of Constitution-making inadequacies and in preventing initiatives aimed at hampering Lithuania’s EU integration. But there are no guarantees that all roadblocks had been espied and no new ones can emerge. After all, the Court has never been requested to rule on the constitutionality of the EU founding Treaties, and its review of domestic laws transposing EU law has been but episodical. There still can be founded speculations as to whether there are any constitutional limits to further EU integration.

A.  Procedural Aspects Of the procedural limits to EU integration three can be mentioned. Firstly, an amendment to an EU Treaty must be ratified by the Seimas (if not by a referendum). The Constitution and the Law on International Treaties117 set no requirements as to the majority of votes needed for ratification;118 they are provided for in the Seimas Statute: treaties are ratified by a majority of MPs taking part in the Seimas sitting, but not less than two-fifths of all MPs, and are denounced by not less than three-fifths of all MPs (Art 181). The Seimas Statute can be changed by the Seimas unilaterally, depending on political conjuncture. The second limit relates to referenda. Referenda are called by the Seimas or when ‘not less than 300,000 citizens with the electoral right so request’ (Art 9 LC). The Law on Referendum119 required a mandatory referendum regarding Lithuania’s participation in international organisations, if it involved conferring parts of the competences of state institutions to institutions or jurisdiction of international organisations. This applied to EU founding Treaties, provided that they met the said criterion (eg, the Treaty of Lisbon was not put to referendum). As to citizen-initiated referenda, the high signature-collection threshold is not unachievable (as the land ownership referendum proved). As to what issues could be put to referenda, the Constitution merely states that they must be ‘[t]he most significant issues concerning the life of the State and the Nation’ (Art 9 LC). Legal scholarship considered that any issue, regarding which not less than 300,000 signatures had been collected, [could] be deemed as the ‘most significant’.120 That approach now has been rejected by the doctrine of limitations on the alteration of the Constitution (see section I.C.3). In particular, prohibited are any such amendments to the provisions of the Constitutional Act on EU which would deny Lithuania’s commitments arising from her EU membership or any such constitutional amendments that would deny her international obligations and at the same time the principle of pacta sunt servanda. Coupled with the prohibition to make any such constitutional amendments that would deny, eg, the innate character of human rights and freedoms, presumably including also those enshrined in the EU Charter of Fundamental Rights, this prohibition has handled the tool to CEC and to the Seimas to effectively bar amendments contradicting Lithuania’s EU commitments. Whether the Seimas can ratify a treaty before the signature-collection for calling a referendum aiming at blocking the ratification is an open question. Of no lesser sensitivity is the question as to by which portion of votes the EU-related issue put to referenda can be approved. As mentioned, the Constitution is not explicit on that. The Law on Referendum contained no 117 In it reference is made to TFEU amendments made under Art 48(6) TEU. 118 Save the treaties by which the State boundaries are altered: they must be ratified by four-fifths of all MPs. 119 Not in force as of 01.07.2021 (see n 11). 120 Petras Ragauskas, Įstatymų leidyba Lietuvoje: Samprata ir institucinis modelis [Legislation in Lithuania: Concept and institutional model] (Vilnius, Teisės institutas, 2005) 55.

Lithuania  449 explicit provisions pertaining specifically to ratification of treaties; the general rule applied that an issue put to a mandatory referendum was approved by more than half of the votes of those who participated, but no less than one-third of all citizens, having the electoral right and included into voters’ lists (except Article 1 LC and other constitutional provisions requiring a higher threshold; see section I.C.3). In the past, amendments to the Law on Referenda had been proposed aiming at lowering the thresholds, even where the issue put to referendum concerned alteration of the Constitution. Now, a new ‘constitutional law’ on referendum must be adopted to enable referenda in principle.121 As things now stand, it is unclear how far the Seimas can go in tailoring the conditions for a particular referendum,122 but given its experience with the Constitutional Court123 and the fact that the latter’s 2014 doctrine (as consolidated in its later case law) is aimed at restricting populist referenda, it must be mindful of not going too far. The third procedural aspect pertains to the challenging of EU Treaties (or their amendments) in the Constitutional Court in the preliminary review procedure. This may delay ratification. There was only one case of this kind (regarding ECHR; see section I.B). A conclusion may be requested also by the Seimas. In the event of finding of unconstitutionality of impugned Treaty (or its provisions), a constitutional amendment should be pursued; if that amendment would pertain to the alteration of Articles 1 and 2 of the Constitutional Act on EU (a logical scenario), a referendum would be needed (see section I.C.3). This would cause another delay and, depending on political conjuncture, may be conducive to trade-offs of sorts. An option of ratifying the treaty in defiance of the Court’s conclusion (formally not impossible; see section I.B) would raise high legitimacy concerns. Ex post petitions regarding constitutionality of statutes by which EU Treaties have been ratified also are not inconceivable: they can be challenged both from the perspective of the procedure of their adoption and from substantive perspective.

B. Sovereignty EU membership is not seen as compromising per se Lithuania’s sovereignty. Various aspects of sovereignty are enshrined in the Constitution, not only where sovereignty/independence is explicitly fixated, but also in articles pertaining to the state’s geopolitical orientation, foreign policy, democratic government etc. Sovereignty’s external aspect, the state’s independence, is perceived as not having been affected by EU membership more than by other forms of states’ interdependence and the need for international cooperation, which involved a considerable element of limitation of sovereign discretion even prior to EU accession. Limitations on the exercise of state’s competences were constitutionalised years ago by providing that Lithuania was bound by international law and participated in international organisations, if that did not conflict with her interests and independence (see section I.C.2). It is held that the Constitutional Act on EU, by emphasising the EU’s respect for human rights and fundamental freedoms and Member States’ national identity and constitutional traditions, repeats in a modified wording the policy goals enshrined in Articles 135 and 136 LC,124 which serve as an international law-related constitutional basis for EU membership. EU integration is perceived as an enhanced interstate cooperation allowing for further pursuance of such goals as Lithuania’s security and her citizens’ welfare, and, thus, for



121 Cf

n 11. nn 39 and 93. 123 Cf nn 11 and 39. 124 Jarukaitis, Europos Sąjunga (n 34) 265. 122 Cf

450  Egidijus Kūris better security of Lithuania’s national interest. The Act has conferred national competences on the EU not termlessly but ‘in the areas provided for in [EU] founding Treaties’ and ‘to the extent it would, together with other [EU Member States], jointly meet her membership commitments in those areas as well as enjoy the membership rights’. What this means in practice will always remain at least partially an open question, especially if EU competences are further expanded. Be that as it may, such expansion cannot take place outside the domain of EU Treaties, subject to ratification. Also, no constitutional provision can be construed as prohibiting Lithuania from leaving the EU, should it deem her interests to be better ensured outside it. As to sovereignty’s internal aspect, the Nation’s suprema potestas, EU membership has not affected the state’s government by the Nation through her democratically elected representatives or (in rare cases of referenda) directly. It can be asserted with a high degree of certainty that the Constitution, as construed today, sets no (more) limits to Lithuania’s further EU integration, but this pertains to the status quo. The question remains what is meant by ‘further’ integration. If it is kept within certain limits, the Constitution in principle is not an obstacle to it. But if the ambitions of the integrationist project pour overboard (in contrast to its limits defined in the Constitutional Act on EU), then the Constitution sets (at least) one limit for ‘further’ integration: sovereignty. Attempts to ‘modernise’ the concept of sovereignty in the context of EU or such theoretical models as ‘supranational organisation sui generis’, ‘intergovernmentalism’ or ‘multilevel governance’, when employed in academic writings or EU policy documents, do not provoke opposition on the part of the Constitution – they more or less fairly reflect what the EU represents, highlight its specificity, but do not equate it to a ‘state’ or ‘federation’ (which, too, is a ‘state’). Sovereignty can live with such ‘modernisation’. But such inventions as ‘supranational federation’, ‘integrative federalism’, ‘multilevel constitutionalism’ or ‘mixed commonwealth’ penetrate into what has been the exclusive domain not of international organisations, even sui generis, but of statehood. The concepts of statehood and sovereignty are resistant to the heart and the lifeblood of their meaning being leached out. Until such intellectual exercises in political and legal linguistics have not attained the status of operating projects and law-making initiatives and have not entered the EU legal vocabulary,125 any consideration as to how Lithuania’s Constitution may react to such hypothetical transformation would be speculative.

C.  National Identity, Constitutional Traditions, Human Rights National identity is a multi-faceted concept; the popular support to EU integration may vary depending on the particular ‘profile’ of national identity vis-à-vis EU integration and is not necessarily in opposition to it.126 Article 4(2) TEU obliges the EU to respect Member States’ ‘national identities, inherent in their fundamental structures, political and constitutional, inclusive of regional and local self-government’. The Constitutional Act on EU is based on the conviction in the EU’s respect for Member States’ national identity and constitutional traditions. Both acts point essentially at the same values. The Preamble of the Constitution enshrines the fundamental values underlying the Constitution: the tradition of statehood, the ‘staunch’ defence of the Nation’s freedom and independence, the ‘innate right of the human being and the Nation to live and create 125 With the exception of the ill-fated ‘Constitution for Europe’, whose miscarriage attests, in its own right, to unprofitableness of volitional remaking of the meaning of fundamental concepts. 126 Cf Julian Aicholzer, Sylvia Kritzinger and Carolina Plescia, ‘National Identity Profiles and Support for the European Union’, (2021) 22 European Union Politics 293.

Lithuania  451 freely in the land of their fathers and forefathers’, the ‘preserved’ spirit, native language, writing and customs, the strivings for national concord, open, just, and harmonious civil society and a state under the rule of law.127 Beyond this starting point, Lithuanian perception of national identity is revealed in greater detail by looking into other constitutional provisions, which enshrine rights and duties of individuals, social structures, democratic government, or political system. Some Constitutional Court rulings deal with aspects of Lithuania’s national identity, such as the protection of Lithuanian language or culture,128 or the continuity of the state.129 But there is no strictly defined, uncontested ‘Lithuanian definition’ of national identity. As this concept is at an early stage of its judicial construction (also in CJEU case law130), it would be premature to speculate as to whether it is narrower or broader in Lithuanian or in the CJEU’s interpretation, and in what respect. Overall, it could not be claimed that the CJEU does not respect Member States’ national identity. On the contrary, with regard to Lithuania one of its aspects – state language – was upheld along the lines drawn in the Constitutional Court’s case law.131 National identity could be invoked as a substantive check against further EU integration in the event of (unlikely) unilateral and/or unconditional EU requirement to submit to such ‘supremacy’ of EU law, where the latter would enter into what always has been an authentic part of constitutional traditions, such as structures of government. The same applies to democracy and human rights. The EU is based on human rights. Despite the (not unfounded) criticism of the EU democracy deficit, there hardly is any case law whereby these values had to be defended not under EU law (or national law interpreted in the light of it), but against it. EU law inevitably affects domestic system of human rights protection.132 The preliminary ruling mechanism presents a venue for avoiding the confrontation and has a harmonising potential. So far, only a few ‘Lithuanian’ cases decided by the CJEU pertain to individuals’ rights proper, and even less to their non-commerce-related rights.133 The domestic courts’ stance is that Lithuanian law can provide for a wider scope of rights than EU law, but it cannot provide a lesser scope; this applies to constitutional law, too.134 References to EU law, in particular to CJEU 127 On the fundamental constitutional values, as interpreted in the Constitutional Court’s case law, see section I.C.1. 128 Eg, LTCC 21.10.1999, No 14/98 (Spelling of names and family names in passports), whereby the writing of the names in Lithuanian passports in the state language is affirmed; LTCC 13.11.2006 (n 51), whereby the State of Lithuania is described as a national state which ‘came into being on the basis of the ethnic nation – Lithuanian nation’; LTCC 21.12.2006 (n 11), whereby it is postulated that ‘[n]o consumerist interest of any social group, inter alia, that denying the cultural identity of Lithuania, may have any impact on the broadcasts of the public broadcaster’; LTCC 05.05.2007, No 18/06 (Minimum qualification requirements to scientists), whereby the ‘identity of the nation’ is construed as commanding that the use of Lithuanian language has to be ensured while developing the humanitarian and social sciences. 129 Eg, LTCC 22.02.2013, No 102/2010 (Period of service required to qualify for a state pension of officials and servicemen); and LTCC 18.03.2014, No 31/2011-40/2011-42/2011-46/2011-9/2012-25/2012 (Criminal responsibility for genocide). 130 Egidijus Jarašiūnas, ‘Pagarba nacionaliniam tapatumui pagal Europos Sąjungos teisę: Aiškinimo potencialas ir Europos Sąjungos Teisingumo Teismo jurisprudencija’ [‘Respect for national identity under European Union law: The potential for interpretation and jurisprudence of the Court of Justice of the European Union’], (2014) 93 Teisė 7. There have been relevant developments in CJEU case law, subsequent to that publication, but they apparently have not changed this general assessment. 131 See CJEU 12.05.2011, C-391/09 (Runevič-Vardyn and Wardyn) ECLI:EU:C:2011:291; regarding the writing of personal names in documents. 132 Which, by extension, applies to legal persons’ rights. See, eg, ECJ 19.04.2007 C-63/06 (Profisa) ECLI:EU:C:2007:233 regarding customs; CJEU 21.10.2010 C-385/09 (Nidera Handelscompagnie) ECLI:EU:C:2010:627 regarding VAT. 133 See, eg, ECJ 11.07.2008 C-207/08 (Babanov) ECLI:EU:C:2008:407 regarding the growth of fibrous hemp; CJEU 12.06.2014 C-314/13 (Peftiev and Others) ECLI:EU:C:2014:1645 regarding the freezing of assets. 134 See LTTC 11.01.2019 (n 102). The LTCC upheld the family reunification of a homosexual couple, having relied on CJEU 05.05.2018 C-673/16 (Coman and Others) ECLI:EU:C:2018:385, despite the overwhelming oposition of the critical segment of Lithuanian political class and the majority of population to homosexual partnership, let alone marriage (even partnership between two heterosexual persons is not enshrined in domestic law, the attempt having failed in the Seimas in May 2021). As to marriage, Art 38 LC enshrines that it is possible between a man and a woman; as to partnership, the Court’s stance is that it is not only not prohibited, but even stems from the Constitution, as construed in the light of EU

452  Egidijus Kūris case law, are increasingly present in domestic courts’ case law. True, regulation discrepancies cannot be excluded with regard to rights’ various segments,135 but the constitutional collision rule (see section III.C) renders this possibility minimal.

D.  Transfer of Competences and Scrutiny of EU Secondary Legislation The competence-transfer clause of Art 1 of the Constitutional Act on EU is rooted in the preference for general positive declarations over the method of exceptions. There is no list or criteria of competences which have been or have not been (or may or must not be) conferred to the EU. Thus, non-transferable competences are not an autonomous safeguard against EU integration. This should not be dissociated from the proviso that the EU respects human rights, fundamental freedoms and Member States’ national identity and constitutional traditions, as well as that EU accession ensures Lithuania’s security and her citizens’ welfare. The said clause refers to the ‘areas provided for in [the EU] founding Treaties’ and to the ‘extent [Lithuania] would, together with the other [Member States] jointly meet her membership commitments in those areas, as well as enjoy membership rights’. There is no official interpretation as to whether these ‘areas’ and ‘extent’ can be expanded on the present constitutional basis. Expansion is possible in the negotiation process, which Lithuania would be part of. Under Lithuanian constitutional law certain competences are non-transferable, but they pertain to areas which have little to do with the EU competences, such as a single market, free movement of people, capital or goods etc. They relate to non-EU agenda issues, such as the formation of state administration institutions, and are secured by other safeguards: national identity and sovereignty. There have been no legal Lithuania-EU competence disputes so far. Likewise, there has been no litigation in Lithuanian courts regarding ex post scrutiny of secondary EU legislation. No constitutional doctrine has been coined regarding such legislation allegedly being passed ultra vires from the standpoint of the national Constitution. But there are preconditions for it, first laid down by the Constitutional Court in 2006,136 when it held that EU legal acts enjoy priority in application against domestic legal acts, save the Constitution itself. Here, the Court’s first application to the ECJ for a preliminary ruling137 deserves mentioning. The request concerned interpretation of a directive provision transposed in a statute, the constitutionality of which was challenged before the Court. In the application, the collision rule was cited verbatim. In its preliminary ruling the ECJ reproduced only part of it, omitting the ‘save the Constitution itself ’ clause, thus representing the Court’s principled position contrary to what it was.138 The preliminary ruling, however, allowed for a wide leeway for the Court, which decided the case on the merits based on the Constitution.139 This peculiar ‘judicial dialogue’ suggests that law. LTCC 11.01.2019 has been heavily relied upon in ECtHR judgment 14.01.2020, application No 41288/15 (Beizaras and Levickas v Lithuania) ECLI:CE:ECHR:2020:0114JUD004128815, which expanded, inter alia, on the prevailing (hostile) societal attitudes regarding homosexuality, as opposed to European standards. 135 One such segment may be social security (and some other) rights, because constitutional provisions granting respective rights are formulated in such a way that, verbatim, rights are granted to ‘citizens’ (see, eg, Anneli Albi, EU Enlargement and the Constitutions of Central and Eastern Europe (Cambridge, Cambridge University Press, 2005) 102 f). Legislation tends to interpret that notion expansively. 136 LTCC 14.03.2006 (n 11). 137 LTCC decision 08.05.2007, No 47/04. The request concerned interpretation of Art 20 of Directive 2003/54/EC of the European Parliament and of the Council of 26 June 2003 concerning common rules for the internal market in electricity and repealing Directive 96/92/EC, [2003] OJ L 176/37. 138 ECJ 09.10.2008 C-239/07 (Sabatauskas and Others) ECLI:EU:C:2008:551. 139 LTCC 04.12.2008, No 47/04 (Connecting to electricity network).

Lithuania  453 an ultra vires doctrine, if it ever materialises into anything more than a general postulate, will not necessarily be welcomed in Luxembourg.140 The ‘reverse’ ultra vires approach is applied by national courts in cases where Lithuanian legislation is tested against the yardstick of EU law. Lithuanian courts are obliged to interpret national law in an EU law-friendly manner (as mentioned, the CJEU and the General Court’s case law is treated as a source of interpretation even of constitutional law). References to EU law may serve as weighty arguments for finding impugned statutory provisions or omissions unconstitutional.141

V.  Constitutional and Sub-Constitutional Rules on Implementing EU Law Apart from the collision rule the Constitution contains no other explicit provision consolidating the priority of application of EU law over national statutory and sub-statutory legislation in the event of their collision. The Law on Legislative Framework postulates the principle that legal provisions must not negate the rights, freedoms and legitimate interests enshrined in EU law.142 Time limits for coming into force of the adopted acts are not applied to domestic legal acts amended in implementation of obligations stemming from EU legal acts. One of the requirements of the law-making procedure is to obtain conclusions from the ETD on the conformity of the draft with EU law, including CJEU case law.143 If the draft is modified so that it may deviate from EU law, a new conclusion may be requested. Drafts must contain references to provisions of the EU legal acts they are implementing; in fact, every statute has an annex with such a list, and some (eg, Civil Code, Code of Criminal Procedure) contain a general provision that their provisions are in conformity with the enlisted EU legal acts, while others reiterate the collision rule. Procedural laws (eg, Code of Civil Procedure; Law on Administrative Proceedings) oblige the courts to apply EU law and, in the event of collision, to request a preliminary ruling from the CJEU.144 The Law on Administrative Proceedings obliges the SAC to analyse and provide surveys on the application of EU law by administrative courts and to give recommendations as to their cooperation with EU courts regarding the uniform interpretation and application of EU law in Lithuania. The preparation, coordination, and presentation of Lithuania’s position on EU draft legal acts, transposition of EU law to Lithuanian national law, and coordination of its implementation (often referred to by the cumulative term ‘coordination’) are regulated by different legal acts, steadily reviewed and updated, responding to the changes in EU legislation. For the legislative branch, that act is the Seimas Statute. It contains a special chapter, designed to regulate in detail the procedures pertaining to the ‘consideration and solution’ of EU affairs. The Seimas is in constant contact with the Government on the matters of coordination of EU affairs. The central role in 140 There has been another application, requesting for a preliminary ruling providing interpretation of Art 148(4) of the Regulation No 1308/2013. See LTCC decision 20.12.2017, No 11/2016 (Trade in milk); CJEU 13.11.2019 C-2/18 (Lietuvos Respublikos Seimo narių grupė) ECLI:EU:C:2019:962; LTCC 06.02.2020, No 11/2016 (Trade in milk). No omission similar to the above-discussed one is to be found in the latter CJEU judgment, maybe owing to the fact that the LTCC did not mention, in its application to the CJEU, the ‘save the Constitution itself ’ clause. 141 See, eg, LTCC 09.02.2007 (n 89); LTCC 06.02.2015, No 18/2012 (Right-hand-drive vehicles on public roads). See also, and especially, n 134. 142 The provisions cited here come from the Law on Legislative Framework and/or the Seimas Statute. 143 And also with treaties, ECHR, and ECtHR case law. 144 The failure to do that may entail the finding of a violation of Art 6 § 1 ECHR. See, eg, ECtHR judgment 16.04.2019, application No 55092/16 (Baltic Master Ltd v Lithuania) ECLI:CE:ECHR:2019:0416JUD005509216.

454  Egidijus Kūris everyday coordination activities falls on the Committees on European Affairs and on Foreign Affairs, explicitly mentioned in the Constitutional Act on EU – a rare case of constitutionalisation of parliamentary committees. As to the executive branch, the variety of acts is wider. The Law on the President contains no provisions directly pertaining to the coordination of EU affairs (but some are to be found in other statutes). However, the President signs the laws (thus, also those by which EU law is implemented) and has a suspensive veto power over adopted laws and a bill-introducing power. The coordination of EU affairs is concentrated in the Government, whose functions, powers, and procedural obligations in this field are regulated by the Law on the Government and by two acts, approved by the Government itself: the Rules of Work of the Government and the Order of Coordination of European Union Affairs. Lower in the hierarchy of executive acts there are (typical) orders of coordination of EU affairs in ministries and agencies, attributed to their sphere of regulation. A series of technicality prescribing acts have been approved by the ETD Director-General. The dominant model for interinstitutional cooperation in dealing with EU affairs is one of coordination, not subordination. Though Lithuanian system of government is predominantly parliamentary, in matters pertaining to EU affairs the Seimas, as a rule, does not issue unilateral binding resolutions for the Government or its ministries/agencies, which they should implement without expressing their opinion. In the Seimas the leading role in these matters falls on the Committees on European Affairs and on Foreign Affairs, who usually sit separately, but may (in certain cases must) hold a joint sitting. The competence of the former Committee is more general, whereas the latter is vested with the examination of issues pertaining to EU common foreign and security policy, common security and defence policy, EU enlargement, and certain aspects of EU external relations (trade etc); it also must hear the opinion, on the matter under consideration, of the President or his representative. The Government shall promptly inform the Seimas (and the President) about adopted EU legal acts and other documents. It can address the Seimas in order to obtain proposals or conclusions on Lithuania’s position in preparation, ask the Seimas’ consultation on other EU documents, and must consult with the Seimas on the draft legislative acts related to its competence. Having found that for the implementation of EU legal act a statute needs to be adopted, it shall inform the Seimas about it as soon as possible. The issues to be put on the agenda are prioritised by specialised committees, whose proposals are summarised by the Committees on European Affairs and on Foreign Affairs in joint sitting and forwarded to the Government. These Committees are entitled to submit to the Government, on behalf of the Seimas, opinions concerning the proposals to adopt EU legal acts. The Government shall assess the recommendations or opinions submitted by the Seimas or its Committees and shall inform the Seimas about their execution. The Committees also can submit the Seimas’ opinion to the PM or ministers taking part in the Council’s meetings. Specialised parliamentary committees may also hold hearings on draft legal acts. Lithuania’s position on the proposal to adopt EU legal acts (prepared by a Government-appointed body) must contain a reasoned opinion on the draft’s conformity with the principle of subsidiarity. The Government must present its opinion on this matter to the Seimas. The control whether the principle of subsidiarity has been heeded is vested in specialised committees and, in case of necessity, in the Committee on European Affairs. In the event that the Committees on European Affairs or on Foreign Affairs find that the draft may contradict the principle of subsidiarity, their conclusion, alongside with the draft Seimas resolution, is presented to the latter to be considered at a plenary sitting which has to take place within a week; they are deliberated in an expedited procedure; the resolution is then forwarded to the Government. The Committee on European Affairs is responsible for the prompt communication of the resolution to Members of the European Parliament, the President of the Council and the Commission, and Member States’ Parliaments. The Seimas can apply to CJEU for an action disputing a legal act on the grounds of its non-conformity to the principle of subsidiarity.

Lithuania  455 The Government shall consider the EU legislative proposals. Regarding them, it may adopt decisions or resolutions, for adoption of which the provisions of Article 95 LC are not applicable. In this speedy, but far from perfect,145 procedure, provided for in the Constitutional Act on EU, issues are deliberated and decisions are adopted not at the Government’s sittings but at (other) conferences, the majority vote of all Government members is not required, and the decision is signed not by the PM and the minister of the respective area, but by that minister alone. The position of Lithuania is presented by the PM or the respective minister who may, depending on the circumstances, modify it (this may raise ultra vires or even constitutionality issues). The Government may – on its own initiative or upon a motion of the Committees on European Affairs or on Foreign Affairs – resort to the parliamentary reservation, by declaring that it will not present its position on the draft legislative act until it is coordinated with the Seimas. The Government can also request in advance the opinion of any of the said Committees and to oblige them to present a parliamentary reservation. The Government is accountable to the Seimas and reports to it on its own performance at the meetings of the Council of EU and the European Council. Reports are presented in writing and orally at sittings of the Committees on European Affairs and on Foreign Affairs, or (on especially important matters) at the Seimas sittings. The Government is responsible for the transposition of EU law to Lithuanian law. This activity is regulated mostly by orders and recommendations of the ETD Director-General, including those pertaining to the evaluation of the translation of legal acts and its authenticity, references to EU legal acts in Lithuanian legal acts, coordination of national legal acts with EU law, monitoring of Lithuanian legal acts (in order to abolish transposed provisions of EU regulations and to ensure direct application thereof), etc. The ETD presents conclusions as to the compliance of the Seimas’ draft legal acts with EU law. The judiciary, a non-political branch of power, has no function in the co-ordination of EU affairs. Under the Constitution, as interpreted by the Constitutional Court in an early ruling,146 a judge ‘may not be qualified as a functionary’ or ‘be requested to cooperate in the implementation of a … political programme’. At the same time, a judge is called ‘to secure human rights … by way of administration of justice founded on laws and other legal acts’. Thus, the courts implement EU law autonomously by applying it, including the EU Charter of Fundamental Rights. The Constitutional Court also has no role in the coordination of EU affairs. It contributes to that coordination indirectly – by interpreting national law in an EU-friendly manner. The frequency and intensity of references to EU law is visibly increasing in all three court systems, as well as the number of applications to EU judicial institutions.147

VI.  Resulting Relationship between EU Law and National Law Lithuanian law operates in an extra-national legal environment, of which EU law is an important part. EU law as a constituent part of Lithuania’s legal system (not ‘national’),148 enjoying

145 Cf LTCC 29.11.2001, 12/2000 (Concentration of market structures), wherein disregard to procedural requirements for consideration of legislative proposals served as a basis for the finding of the impugned Government decision unconstitutional. 146 LTCC 06.12.1995, 3/95 (Law enforcement employees’ work pay). See also LTCC 21.12.1999 (n 6) and LTCC 12.07.2001, No 13/2000-14/2000-20/2000-21/2000 (Judges’ salaries). 147 Eg, over 40 applications for preliminary ruling were submitted to ECJ/CJEU from 2006 to mid-2018. 148 But not its only extra-national part; the other two are international law (applicable in Lithuania) and foreign law (to the extent it is applied by Lithuanian courts/institutions).

456  Egidijus Kūris exceptional constitutional status. The collision clause grants EU legal acts priority of application over Lithuanian legal acts, save the Constitution itself. EU law is based on the idea of its supremacy over national law, including – in the event of collision – constitutional law. In its turn, Lithuanian constitutional law perceives its relationship with EU law not in terms of supremacy, but of priority of application. As long as the ‘competition of supremacies’ endures and common algorithm is not found, there will always remain a space for tension. On the other hand, the potential conflicts do not seem to pose a great threat for several reasons. Firstly, Lithuania’s EU commitments stem from her Constitution, and the new inclusive paradigm of constitutional law is able to accommodate extra-national elements, such as EU law, as sources of interpretation of national constitutional law. Secondly, Lithuanian courts (alongside the legislature and the executive) profess an EU law-friendly approach, manifested, inter alia, in their testing of Lithuanian legislation against the yardstick of EU law. Thirdly, as long as EU integration does not infringe on the basic values underlying Lithuania’s constitutional system, her Constitution seems to have no in-built barriers against further EU integration, provided the latter does not penetrate into what has long been the exclusive domain of statehood. Fourthly, EU itself is mindful of the obligation to respect Member States’ national identity and constitutional traditions. All this makes the confrontation between Lithuania’s constitutional order and EU law unlikely in practice. If it takes place, we perhaps should witness further judicial and political will allowing for tackling the problem.

References J Aicholzer, S Kritzinger and C Plescia, ‘National Identity Profiles and Support for the European Union’, (2021) 22 European Union Politics 293. A Albi, EU Enlargement and the Constitutions of Central and Eastern Europe (Cambridge, Cambridge University Press, 2005). M Avbelj, ‘Supremacy or Primacy of EU Law – (Why) Does It Matter?’, (2011) 17 European Law Journal 744. A Brewer-Carías, Constitutional Courts as Positive Legislators: A Comparative Study (Cambridge, Cambridge University Press, 2011). J Czuczai, ‘Constitutional Preparation for EU Accession in the New Central and Eastern European Member States: Is the Rule of Law Better than the Rule of Politics?’, in Jaap W de Zwaan et al (eds), The European Union: An Ongoing Process of Integration. Liber Amicorum Alfred E. Kellermann (The Hague, TMC Asser Press, 2004) 276. E Jarašiūnas, ‘Pagarba nacionaliniam tapatumui pagal Europos Sąjungos teisę: Aiškinimo potencialas ir Europos Sąjungos Teisingumo Teismo jurisprudencija’ [‘Respect for national identity under European Union law: The potential for interpretation and jurisprudence of the Court of Justice of the European Union’], (2014) 93 Teisė 7. I Jarukaitis, Europos Sąjunga ir Lietuvos Respublika: Konstituciniai narystės pagrindai [The European Union and the Republic of Lithuania: Constitutional Foundations of Membership] (Vilnius, Justitia, 2011). E Kūris, ‘Konstitucinės justicijos proceso teisės klausimu’ [‘On the issue of the law in constitutional justice procedure’], (2011) 78 Teisė 7. E Kūris, ‘Apie nepagarbą teisei’ [‘On the disrespect for law’], (2012) 13 Notariatas. E Kūris, ‘On Perception of Constitutional Law: A Theoretical Approach to Constitutional Justice’, in G Harutyunyan (ed), New Millenium Constitutionalism: Paradigms of Reality and Challenges (Yerevan, NJHAR, 2013). E Kūris, ‘Standards of the Rule of Law’, in Egidijus Kūris (ed), Crisis, the Rule of Law and Human Rights in Lithuania (Šiauliai, Titnagas, 2015). E Kūris, ‘On Jurisprudential Constitution and European Integration’, in I Nguyen Duy et al (eds), Uttensammenligning: Festskrift til Eivind Smith 70 år (Bergen, Fagbokforlaget, 2020).

Lithuania  457 K Lapinskas, ‘Lietuvos konstitucijų istorinių teisinių sąsajų beieškant’ [‘In search of historical legal links between the constitutions of Lithuania’], in Konstitucija, žmogus, teisinė valstybė [Constitution, human being, law-based state] (Vilnius, Lietuvos žmogaus teisių centras, 1998). M Lukas and A Medelienė, ‘Public Finances’, in Egidijus Kūris (ed), Crisis, the Rule of Law and Human Rights in Lithuania (Šiauliai, Titnagas, 2015). M Maksimaitis, Lietuvos valstybės konstitucijų istorija (XX a. pirmoji pusė) [The history of constitutions of the State of Lithuania (The first half of XX century)] (Vilnius, Justitia, 2005). P Ragauskas, Įstatymų leidyba Lietuvoje: Samprata ir institucinis modelis [Legislation in Lithuania: Concept and Institutional Model] (Vilnius, Teisės institutas, 2005). L Šaltinytė, ‘Lithuania’, in Constitutional Change through Euro Crisis Law: A Multi-Level Legal Analysis of Economic and Monetary Union, https://eurocrisislaw.eui.eu/wp-content/uploads/sites/55/2019/05/ Lithuania.pdf. V Vadapalas, ‘Opinion of the Constitutional Court of the Republic of Lithuania in the case concerning the Conformity of the European Convention of Human Rights with the Constitution of Lithuania’, (1995) 55 Zeitschrift für ausländisches öffentlisches Recht und Völkerrecht 1077. V Vaičaitis, 2020 m. liepos 30 dienos Konstitucinio Teismo nutarimo dictum [The dictum of the Constitutional Court ruling of 30 July 2020], www.teise.pro/index.php/2020/08/18/v-vaicaitis-2020-mliepos-30-dienos-konstitucinio-teismo-nutarimo-dictum. V Vaičaitis (ed), Lietuvos konstitucionalizmo istorija (Istorinė Lietuvos Konstitucija): 1387 m. – 1566 m. – 1791 m. – 1918 m. – 1990 m. [History of Lithuanian Constitutionalism (Lithuania’s Historic Constitution): 1387, 1566, 1791, 1918, 1990] (Vilnius, Vilniaus universiteto leidykla, 2016). I Vörös, ‘Contextuality and Universality: Constitutional Borrowings on the Global Stage – The Hungarian View’, (1999) 1 University of Pennsylvania Journal of Constitutional Law.

458

16 Luxembourg JÖRG GERKRATH

The Grand Duchy of Luxembourg is a strongly integrated EU Member State whose membership relies, however, on poorly developed constitutional foundations. This is yet to be changed by a major constitutional overhaul that started in 2009 and is still not yet adopted at the end of 2021. The following lines will, thus, refer to the current Constitution, dating back to 1868,1 as well as to the draft of the new Constitution, available as part of the final report of the parliamentary committee in charge of constitutional amendments from 6 June 2018.2

I.  Main Characteristics of the National Constitutional System Luxembourg’s constitutional system may be succinctly presented referring to its five predominant qualities: a system relying on a nineteenth-century document (section A), a small state longing for integration (section B), a monarchy (section C), a political system based on parliamentary democracy (section D), and a judiciary in need of reform (section E).

A.  A Constitutional System Based on a Charter Dating Back to 1868 Luxembourg’s constitutional history as an independent state begins in the first half of the nineteenth century. Although the political existence of the Grand Duchy is usually considered to be a result of the Treaty of Vienna of 1815, its first ‘home-grown’ Constitution was only granted on 12 October 1841. First, there was a phase of instability from 1815 to 1868 during which the country was consecutively governed by five different constitutional documents, only the three most recent (1841, 1848, 1856) being genuinely Luxembourgish. A phase of constitutional stability started with the Constitution of 1868, which, after some 38 amendments, still is in place today. Despite these various amendments the current Constitution still very much resembles the 1868 text, which borrowed many provisions from the liberal Constitution of 1848, which, in turn, was widely inspired by the Belgian Constitution of 1831. The amendments to the Constitution have not been as significant as one might expect. Altogether 70 out of 121 articles of the original 1 The Constitution can be accessed at www.legilux.public.lu/leg/textescoordonnes/recueils/Constitution/. An up-todate English version is available at www.constituteproject.org/constitution/Luxembourg_2009.pdf. All Luxembourgish legal documents are accessible in French: www.legilux.public.lu/. 2 See the ‘Rapport de la Commission des institutions et de la révision constitutionnelle’, as of 6 June 2018 on the webpage of the Chambre des députés, www.chd.lu/wps/portal/public.

460  Jörg Gerkrath text of the Constitution have never been revised. Forty-seven articles have been revised at least once. Articles 11 (social and economic rights of the individual person and the family) and 51 (Chambre des Députés) for instance have been modified five times each. Articles 10, 63, 73 and 121 have been abolished, and eight new articles have been inserted. Substantively, the main revisions sought to strengthen parliamentary democracy, to adapt the Constitution to the needs of European integration, to complete the catalogue of fundamental rights, and to modernise the Constitution according to the ‘rule of law’ principle. However, none of these four objectives has been fully achieved. As a result, the Constitution of 1868 appears somewhat outdated. While several initiatives for a general revision of the Constitution have been undertaken since the 1970s, none of them has been successful. Only fractional revisions were adopted for a century and a half. Occurring at different times and touching on a variety of topics, they have to some extent undermined the coherence of the initial text. In April 2009 the Chambre formally initiated an amendment procedure designed to deliver a complete overhaul of the Constitution, to modernise its out-dated terminology, to erase a number of old formulations and incoherencies resulting from past puncturing modifications, to adapt the legal text to the political reality, and to bring it to reflect the ‘living constitution’ as operationalised by the institutions. The current state of this amendment procedure will be analysed in section VII.

B.  A Small State Seeking Integration The Grand Duchy is the second-smallest Member State within the Union. It covers an area of 2,586 square kilometres and has a resident population of about 600,000 inhabitants, of whom 48 per cent are foreigners. More than 80 per cent of these foreigners are EU citizens.3 The statistical growth of past years suggests it is very likely that Luxembourgers will become a minority in their country in the near future. Luxembourg’s successive incorporation in the German Confederation (1815–66), the ‘Zollverein’ (1842–1918), the Belgo-Luxembourg Economic Union (from 1921) and Benelux (from 1944) helped it acquire the experience indispensable for being prepared for the legal implications of membership of the European Communities (EC) and the European Union (EU). All the state organs and the major political parties as well as the vast majority of the population share the opinion that the economic and political interests of the Grand Duchy favour an evercloser integration within the Union Benelux, the Council of Europe and the EU. These interests also foster the development of the ‘Greater Region’, which includes Lorraine in France, Saarland and Rhineland-Palatinate in Germany, the Walloon Region and the French and German-speaking communities of Belgium. The Grand Duchy of Luxembourg is the only sovereign state member of the Greater Region and its institutional framework. Though Luxembourg’s population accounts for less than 5 per cent of the total population of this cross-border space, it boasts the most dynamic growth. Cross-border or ‘frontier workers’ represent 45 per cent of all employed persons on its domestic labour market. All in all some 180,000 cross-border workers are employed in Luxembourg. Half of them commute from France, a quarter from Belgium and another quarter from Germany. The Grand Duchy is, thus, highly interested in all efforts to imagine new forms of governance for the internationalised labour market.



3 Figures

published by the statistical agency for 2018, www.statec.lu.

Luxembourg  461

C.  A Constitutional Monarchy Given that Luxembourg is a constitutional monarchy, the Grand Duke, being the head of state, the symbol of its unity and the guarantor of national independence (Art 33), only holds the powers that are explicitly conferred on him by the Constitution and the laws (Art 32). By virtue of Article 3 of the Constitution, the crown of the Grand Duchy is hereditary in the house of Nassau in accordance with the Family Compact of the House of Nassau of 30 June 1783, Article 71 of the Treaty of Vienna of 9 June 1815 and Article 1 of the Treaty of London of 11 May 1867. The currently pending constitutional revision aims to clarify the position of the Grand Duke as head of state, ‘symbol of its unity and guarantor of national independence’ and to adapt the provisions of the Constitution that might create confusion as to his responsibility for political decisions, which have to be endorsed by other constitutional bodies. He will, therefore, lose some of his remaining formal legislative and executive competences. Furthermore, all provisions relating to succession, regency and lieutenancy, which the Family Compact of the House of Nassau currently provides for, will be integrated into the Constitution in an adapted form. The attachment of the Luxembourgish people to its monarchy has been proven by the very first referendum that was held in the Grand Duchy on 28 September 1919 on the dynastic question and the form of the state. This so-called ‘political’ referendum allowed the voters to make a real choice between several options. These did not only consist of a single question to be answered by ‘yes’ or ‘no’, but the possibility to choose between four options: to keep the Grand Duchess Charlotte, who had taken over the throne a few months before; to retain the dynasty, but replace Charlotte; to retain the monarchy, but replace the dynasty; to change the form of state to a republic. Though the vote was not compulsory at that time, there was an electoral turnout of over 72 per cent. The vast majority of almost 78 per cent of valid votes chose the first option, to retain the Grand Duchess in function. Almost 20 per cent were in favour of establishing a republic.4 The monarchical form of government has never been challenged since. The Grand Duke, although being the head of state, may not represent Luxembourg in the European Council and never did so. He does indeed not meet the condition, made explicit by (post-Lisbon) Article 10 TEU, of being ‘democratically accountable’ either to the Chambre des Députés or to the citizens of Luxembourg.5 The hereditary nature of his function is even difficult to reconcile with one of the principles of the Union’s constitutional system, which makes the equality of all citizens a central rule.6 In any event, Union membership of a constitutional monarchy does not come without creating frictions, which could adversely affect its constitutional law.7

D.  A Political System Based on Parliamentary Democracy According to Article 51(1) of the Constitution, Luxembourg is ‘ruled by a system of parliamentary democracy’. Consequently it is the Chambre that ‘represents the country’ (Art 50). Thus the political system is clearly of a representative nature. The use of referenda has remained exceptional. 4 Ben Fayot, Les quatre référendums du Grand-Duché de Luxembourg. Essai (Luxembourg, Éditions de la Petite Amérique, 2006) 35. 5 Jan-Herman Reestman, ‘The state of the European Union’s monarchies. An introduction to the series’, (2011) 7 European Constitutional Law Journal 270. 6 Wim Roobol, ‘Twilight of the European Monarchy’, (2011) 7 European Constitutional Law Journal 286; Luc Heuschling, Le citoyen monarque. Réflexions sur le Grand-Duc, la famille grand-ducale et le droit de vote (Windhof, Promoculture Larcier, 2013). 7 See Hans Ulrich Jessurun d’Oliveira, ‘The EU and Its Monarchies: Influences and Frictions’, (2012) 8 European Constitutional Law Journal 63.

462  Jörg Gerkrath Legislative power is vested in the Chambre, the members of which are elected directly for fiveyear terms. The Chambre includes 60 deputies elected from lists submitted by different political parties. Since 14 October 2018, following the last legislative elections, the Chambre has been composed of 21 seats for the CSV (Social Christian Party), 10 seats for the LSAP (Luxembourg Socialist Worker Party), 12 seats for the DP (Democratic Party), nine seats for Déi Gréng (the Green Party), four for Alternativ Demokratesch Reformpartei (ADR), and two for Déi Lénk (left-wing). Constitutional amendments require at least 40 votes within the Chambre, and need to be accepted by MPs from at least three different groups. In practice, constitutional amendments are adopted by majorities exceeding 80 per cent. A recent constitutional procedure, which lasted from December 2008 to March 2009, is an interesting illustration of how the political system influences constitutional amendments. After the Grand Duke announced that he would refuse to approve a bill on euthanasia that the Chambre was about to adopt, the Government and the Chambre decided unanimously to abolish the requirement for royal assent to laws which had existed in the Constitution since 1848.8 That meant that the second vote on the law on euthanasia was postponed until the Constitution was amended. Then the law was finally adopted by the Chambre and simply promulgated by the Grand Duke. Under Luxembourg’s unicameral system, all laws are subject to a second vote by the Chambre, an interval of at least three months being observed between the two votes. However, the Chambre may decide in accordance with the Conseil d’État, sitting in public, that there is no need for this second ‘constitutional vote’. The Conseil d’État has a veto power, and may refuse to grant the waiver of the second constitutional vote. This body advises the Chambre in the drafting of legislation. The Conseil’s opinions have no binding effect but are widely followed. In practice its role is that of a second chamber. The Conseil comprises 21 members, including at least 11 holders of a doctoral degree in law or of a foreign higher education degree in law. They are nominated by the parties represented in the Chambre, and appointed by the Grand Duke. The responsibilities of its members are in addition to their normal professional duties. To be appointed, candidates must be Luxembourger, enjoy civil rights, reside in the Grand Duchy, and be at least 30 years of age. In addition, the Grand Duke’s heir may be appointed at the time this title is conferred. The Grand Duke and the Government exercise executive power in conjunction. The latter includes the Prime Minister, who serves as head of Government. He is the leader of the political party or coalition of parties having the majority of seats in the Chambre.

E.  A Judiciary in Need of Reform Currently, the judiciary is divided into two main branches: ordinary courts and administrative courts. In addition there are also specialised courts in the field of social security. The Superior Court of Justice, comprising the Court of Appeal and a Cour de cassation, heads the branch of ordinary courts as well as the social security courts. Reorganising this judicial system is an additional goal introduced into the current revision process by the Government under the responsibility of its Ministry of Justice. In addition it is also intended to establish a National Justice Council (Conseil National de la Justice) which would oversee the independence of judges and ensure good judicial conduct. In the course of the 8 Luc Frieden, ‘Luxembourg: Parliament abolishes royal confirmation of laws’, (2009) 7 International Journal of Constitutional Law 539.

Luxembourg  463 debates that took place during the meetings of the parliamentary committee, the Government and the committee members discussed extensively the possibility of transforming the present highest court, the Superior Court of Justice, into a Supreme Court for both the ordinary and the administrative courts. This would have implied abolishing the Constitutional Court and transferring power to the ordinary judges to review the constitutionality of laws. This idea was finally abandoned in spring 2018. The Constitutional Court was created only recently by a constitutional amendment of 12 July 1996 and an Act of 27 July 19979 to exercise ex-post constitutional review of laws. Its function is to examine the constitutionality of ordinary statutes, apart from laws approving treaties, which are explicitly excluded from judicial review. If a party questions the constitutionality of a law before an ordinary or administrative court, and if the issue of constitutionality is deemed vital to the solution of a dispute, the Court must refer the matter – explicitly citing the legal norm in ­question – to the Constitutional Court. The public has no direct recourse to the Constitutional Court. The review technique applied by the Constitutional Court leads to a rather narrow and abstract way of reviewing the constitutionality of enacted laws. The Court confines itself to reviewing the legal disposition referred to it against the articles of the Constitution specified in the preliminary question. It does not review the constitutionality of the law in question as a whole, nor does it extend its review to other articles or principles of the Constitution that have not been explicitly cited. Furthermore, the Court adheres to a rather literal interpretation of the wording of the Constitution. From 1998 until September 2018 the Constitutional Court delivered 138 judgments. Most of them concern the principle of equal treatment enshrined in Article 10bis of the Constitution.10 The effects of judgments of the Constitutional Court are in principle limited to the litigation, which raised the preliminary question. The Chambre therefore remains free in its response to adverse judgments of the Court. In fact several legal provisions declared contrary to the Constitution remain in place.

II.  Constitutional Culture The Constitution of Luxembourg falls within the category of constitutions tending to be more ‘evolutionary’ in nature. Although it is clearly part of the positive law in force, it is at the same time considered rather a historic and political document than a truly normative one. Consequently there is quite a difference between the written document and the ‘living constitution’, for instance regarding the relations between the main political organs of the state. To understand the country’s constitutional culture, three patterns must be taken into account: the Constitution had been somewhat forgotten, at least until 1997 (section A); its political system functions according to the idea of a ‘consensus democracy’ (section B); its leading political and legal principle is pragmatism (section C).11 9 Loi du 27.7.1997 portant organisation de la Cour Constitutionnelle. 10 Jörg Gerkrath (ed), La jurisprudence de la Cour Constitutionnelle du Luxembourg 1997–2007 (Luxembourg, Pasicrisie luxembourgeoise, 2008); and Jörg Gerkrath (ed), Les 20 ans de la Cour Constitutionnelle: trop jeune pour mourir? (Luxembourg, Pasicrisie luxembourgeoise, 2018). 11 The following considerations are widely inspired by a paper written in French by my colleague Luc Heuschling, who summarised in a very convincing way the main characteristics of Luxembourg’s political culture. See Luc Heuschling and Philippe Poirier, ‘L’opposition politique au Grand-duché de Luxembourg. Regards croisés, de science juridique et de science politique, sur une démocratie consociative’, in J-P Derosier (ed), L’opposition politique (Paris, Lexis Nexis, 2016) 225.

464  Jörg Gerkrath

A.  The Somewhat ‘Forgotten Constitution’ A first distinguishing element of Luxembourg’s constitutional culture is that political and legal actors lack an interest in constitutional law. Though many efforts were made to modernise branches of law such as banking or media law, the Constitution had been to some extent forgotten.12 Several reasons may explain this. First, the monistic tradition of the Grand Duchy’s legal order and the principle of primacy of international law led judges and lawyers to invoke and apply human rights based on international treaties rather than stemming from the Constitution. Thus the main source of human rights in Luxembourg is the European Convention on Human Rights (ECHR). Second, many practices and usages have developed praeter legem besides the old Constitution. The political actors, thus, often prefer to refer to such practices than to the formal rules of the Constitution that proved furthermore to be difficult to revise because of a very strict amendment procedure in place until 2003. The great difference between the written and the living constitution stresses the importance of the current revision procedure aiming to modernise the outdated and, on some aspects, fragmentary constitution. Subject to many amendments in the past, the Constitution needs a ‘restatement’ in order to reconcile the written with the living constitution. Third, before the amendment of 1996, any constitutional review of parliamentary acts resided solely with the political organs, namely the Chambre des Députés and the Conseil d’État. The latter still exercises a kind of a priori review of the conformity of any draft bill with rules of higher law, including general principles of law, international treaties and the provisions of the Constitution. A constitutional court with limited powers was only established in 1997. It does not have the power to invalidate legislative acts, which can only be submitted to it by ordinary courts through a preliminary ruling procedure.

B.  A ‘Consensus Democracy’ As a small monarchy with a parliamentary form of government, the Grand Duchy of Luxembourg developed a particular model of consensus democracy, sometimes referred to as the ‘Luxembourgish model’. This model of a Konsensdemokratie functions very well in a small country like Luxembourg where, in addition to the political parties and trade unions, the different interest groups are well organised through a system of professional chambers. The political system is strongly characterised by stability, proximity to the citizens and a common desire to take decisions based on consensus. This is particularly true for decisions on constitutional revision, which are mostly supported by much stronger parliamentary majorities than necessary (40/60), and are very often taken unanimously. The political culture favours consensus, and parties share a broad agreement on key issues, including the value of European integration. Consequently, Luxembourg experienced until recently a remarkable governmental stability. Seven political parties exist for the moment. The Christian Social People’s Party (CSV) resembles Christian Democratic parties in other West European countries and enjoys broad popular support. Its former leader, Prime Minister JeanClaude Juncker, in power from 1995 to 2013, was the longest-serving head of government in the EU. The Socialist Party (LSAP) is a centre-left party similar to most Social Democratic parties in Europe. The LSAP defends state intervention in the economy and the sustainability of the welfare

12 Alex

Bonn, La Constitution oubliée (Luxembourg, Imprimerie Centrale, 1968).

Luxembourg  465 system. Part of the government from 1984–99, it currently participates in the government coalition with the Democratic Party and the Greens. The centre-right Democratic Party (DP) draws much of its support from civil servants, the liberal professions, and urban middle class. Like other West European liberal (ie, libertarian) parties, it advocates both social legislation and minimum government involvement in the economy. In opposition from 1984–99, the DP overtook the LSAP to claim the role of junior partner in the government from 1999–2004. It is currently heading the coalition government with Prime Minister Xavier Bettel. The Green Party has received growing support since it was officially formed in 1983. Other notable parties are the ADR (Alternative Democratic Reform Party), a right-wing party defending national identity, and the Left (former communist). Currently, they occupy four and two seats in the Chambre, respectively. The functioning of the Luxembourgish political system in the way of a consensus democracy is to some extent facilitated by the role of the so-called ‘professional chambers’, which represent corporative interests. Five commercial and professional chambers were established by a Law of 4 April 1924, namely one for agriculture, one for the crafts, one for trade, one for private employees and one for labour (chambre d’agriculture, des métiers, de commerce, des employés privés, and de travail respectively).13 A Law of 12 February 1964 added a sixth chamber for public servants (chambre des fonctionnaires et employés publics).14 The number was reduced to five again when the Chamber of Employees took office on 5 January 2009, following a merger of the Chamber of Private Employees and the Chamber of Labour. The chambers are public bodies with civil law legal personality. They consist of five ordinary members and the same number of substitutes, elected every five years. Their primary task is to protect the professional interests of their members. The Chambre des Députés and the Government must request the opinion of the chambers on all bills and legislative proposals and on all Grand Ducal and ministerial regulations which are primarily concerned with the trades and professions represented in the chambers. The Government can dissolve the chambers if there are serious reasons for doing so. In that case, new elections must be held within three months after their dissolution. All political actors also share a strong attachment to the ‘Luxembourgish social model’. This model is based on an ‘index’ mechanism of automatic salary rises according to the inflation rate, and powerful trade unions. It also benefits from the tradition to negotiate important reforms – touching eg the labour market, the pensions system or the social security policy – within the so-called ‘Tripartite’, an encounter of representatives of the government, the trade unions and the employers.

C.  Pragmatism as a Leading Principle of Policy and Legal Philosophy In the Grand Duchy’s legal order, law is considered as a toolbox, an instrument to regulate the social sphere. Written mainly in French at present, the law is a result of a multitude of legal transfers. Constitutional law is rooted largely in nineteenth-century Belgian constitutional law, while featuring elements derived from German constitutional monarchy. If there is something specific in Luxembourg’s legal culture, it is possibly this uninhibited attitude, in case of problems, to look beyond the immediate boundaries to seek the appropriate legal solution. Without falling into blind adulation of solutions of a country in particular the usual approach is: to compare, 13 Loi du 4 avril 1924, portant création de chambres professionnelles à base élective. 14 Loi du 12 février 1964 ayant pour objet de compléter la loi du 4 avril 1924, portant création de chambres professionnelles à base élective par la création d’une chambre des fonctionnaires et employés publics.

466  Jörg Gerkrath to evaluate, to choose, and to combine foreign legal solutions in order to make them work in Luxembourg. In addition there is a certain distrust vis-à-vis abstract legal constructs that appear too intellectual. Until 2003, Luxembourg’s legal system had to do without the presence of a university and law professors. It relied exclusively on practitioners trained either in Belgium or in France. Legal rules are supposed to solve real problems. Thus, some trends of constitutionalism in Europe and in the world (rationalisation of parliamentarism from the 1920s, the development of constitutional courts since 1945, the tendency, in the last 20 years, to admit material limits to the power of constitutional amendment) produced no (or only a reduced) echo in Luxembourg. Some problems simply did not occur (the threat of government instability, the risk of a shift towards a dictatorship) or were not perceived at fair measure (the risk of violation of rights and freedoms by the legislator).

III.  Constitutional Foundations of EU Membership and Closely Related Instruments The current Constitution does not contain any explicit reference to EU membership. Since a 1956 amendment, Article 49bis allows simply for ‘the exercise of powers reserved by the Constitution to the legislative, executive, and judiciary branches to be temporarily vested, by treaty, in institutions of international law’. According to Articles 37 and 114, such treaties are to be approved by a law meeting the qualified (two-thirds) majority requirements established for constitutional amendments. The Chambre approved the ratification of the Rome, Maastricht, Nice, Amsterdam and Lisbon treaties by adopting each time such a law.

A.  No Specific Provision on EU Membership With regard to EU membership, Luxembourg has pursued a strategy of minimal constitutional adjustments. The Grand Duchy ratified the European Coal and Steel Community (ECSC) and European Defence Community (EDC) treaties in absence of any constitutional provision on the transfer or delegation of powers to international organisations. Contrary to the opinion of the Government, which thought that the Constitution implicitly allowed such a transfer, the Conseil d’État took the view that the Constitution did not. For reasons of expediency, the Conseil d’État, however, did not oppose the legislature’s approval of both treaties, but urged an immediate constitutional amendment to correct the perceived lacuna. Thus, on 25 October 1956, the Constitution was amended to add Article 49bis. Simultaneously, Article 37 was modified to require such treaties to be approved by a law meeting the voting requirements established for a constitutional amendment in Article 114. The regrettable wording of Article 49bis, allowing only ‘temporary’ transfer of competences, is to be modified in the course of the current revision procedure. In the past this provision has always been construed very widely and has not hampered any ratification of subsequent EC/EU founding, revision or enlargement treaties. A proposal made in 2009 to introduce an entire new chapter on the EU was ultimately withdrawn without debate. Luxembourg’s constitutional culture also explains to some extent the sentiment that EU related amendments are superfluous. In addition, the level of public support for the EU has always been one of the highest throughout the Union and is shared by all political parties in Parliament.

Luxembourg  467 European integration has never been conceptualised in Luxembourg as a threat to constitutional rules, principles or values. Thus, only constitutional rules that have become explicitly contrary to requirements of EU law have been amended. Constitutionalising Luxembourg’s EU membership has been put on the agenda by the opinion of the Conseil d’État of June 2012 on the pending constitutional amendment proposal from April 2009. Although the Grand Duchy is one of the founding states of the EU, its present Constitution does not contain any reference to this membership, or to its constitutional foundations and implications. In its opinion the Conseil d’État recommended constitutionalising Luxembourg’s participation in the process of European integration via a new Article 5. Furthermore, the Conseil suggested further insertions: a reference to the voting rights of European citizens, an adoption of the Court of Justice of the EU (CJEU) formula with respect to access to public employment, and codification of the Grand Duke’s power to adopt regulations in order to ensure compliance with the legal instruments adopted by the EU, rather than proceeding by parliamentary statute. As previously mentioned, the 2009 Revision Proposal has yet to come to fruition, but a version thereof is expected to be approved through referendum in 2019. The draft as it stands in June 2018 contains several proposals for EU-related amendments. A new Article 5 reads: ‘The Grand-Duchy of Luxembourg participates in European integration. The exercise of powers of the state may be transferred to the European Union and to international institutions by an Act of Parliament adopted by qualified majority.’ Article 11 contains a reference to the CJEU’s well-established case law regarding access to public employment. It provides that: ‘The law determines access to public employment. It may reserve for Luxembourgers public employment including direct or indirect participation in the exercise of public authority and in the functions having as their object the safeguard of the general interests of the state.’ Article 47(3) allows the Grand Duke to ‘adopt the necessary regulations for application of the legal acts of the European Union’. Finally, the draft provides for the ex-ante control (through consultative opinions of the Conseil d’État, Article 88) and ex-post review (by binding decisions of the courts, Article 96) of Acts of Parliament and regulations with regard to ‘higher law’, namely the Constitution, international treaties, EU legislation and the general principles of law. According to Article 96, ‘[t]he courts shall apply Acts and regulations only in so far as they conform to the norms of higher law’. With regard to judicial control, it must be recalled that the jurisdiction of the Constitutional Court is strictly limited. Acts of the Chambre approving international treaties are explicitly excluded from the competence of the Constitutional Court (Article 95ter[2]). In Luxembourg, this is considered as a consequence of the primacy of international treaties. In effect this means that cases such as Lisbon and ESM in Germany cannot arise in Luxembourg. Acts of the Chambre transposing or executing obligations deriving from secondary EU legislation are not explicitly excluded from the Constitutional Court’s competence. Until now, however, no such Act has been submitted to the Constitutional Court as a preliminary question by an ordinary court. The opinion seems to prevail that the ‘immunity’ of Acts approving treaties also covers Acts implementing secondary EU legislation.

B.  Continued Relevance of General Provisions on International Treaties In absence of any specific ‘EU clause’ in the Constitution, Luxembourg’s membership in the EU as well as its participation in any further steps of EU integration are currently governed by the general constitutional provisions on international treaties.

468  Jörg Gerkrath Regarding the ratification of international treaties, Article 37 of the Constitution provides that treaties are operative after having been approved by law and published in the form specified for the publication of laws. Secret treaties are abolished. International treaties that transfer powers to international institutions need to be approved by Parliament, according to a special two-thirds majority requirement, similarly to the constitutional amending procedure as provided by Article 114. Reservations to international treaties by Luxembourg require legislative approval as well.15 Article 95ter [2] explicitly prohibits the Constitutional Court from reviewing Acts of Parliament that approve international treaties. There is no reference to international customary law in the current Constitution nor in the draft of the future Constitution, which is under consideration. Luxembourg’s experience as one of the EU founding members shows that, if necessary, it is also feasible to participate successfully in the process of European integration for more than 60 years without any amendment of the Constitution referring explicitly to the EU or EU law and on the basis of a constitutional provision that explicitly allows only temporary delegation of powers to international institutions. The absence of a Europe clause has never been considered as a lacuna endangering constitutional values or fundamental rights standards. The intended amendments within the current revision procedure are rather seen as a commitment to European integration and an overdue adaptation of the Constitution to the legal reality.

C.  Practice and Doctrinal Debate on these Provisions When Article 49bis was introduced in 1956, it essentially raised two controversial questions: the position of the Article in the Constitution’s structure and its content. Regarding the first question, the Conseil d’État opined that this provision needs a distinct position in the Constitution because it introduces derogation to the exercise of the sovereign powers (executive, legislative, and judicial) that are also regulated in Articles 33-49. Moreover, this provision deals with the exercise of a new type of power, which is also linked to sovereignty. For these reasons the Conseil d’État proposed the addition of a new Article, 49bis, instead of incorporating a new paragraph in the existing provisions. As for the choice of the wording ‘institutions of international law’, the Conseil d’État had originally proposed the phrase ‘international institutions’. This proposal, however, was not retained by the parliamentary committee, which opted for the term ‘institutions of international law’ in order to encompass not only international but supranational institutions as well.16 Drawing inspiration from the Belgian Constitution of 1831, Article 37 was last modified by the 1956 constitutional amendment. During the revision, Article 37 was at the centre of a debate between the Government and the Conseil d’Etat. The Parliament declared its intention to systemise and integrate all the constitutional provisions on international relations in one chapter, while the Government proposed the addition of a series of new articles to Article 37 in order to exhaustively consolidate all the constitutional provisions on international relations. These proposals were weighed with care by the Conseil d’Etat, which opted for the solution of adding only a new Article 49bis, leaving the structure of the Constitution intact. The 1956 amendment did not alter the principle of the approval of international treaties by Parliament. The constitutional amendment only specified the form that this approval would take, by adding the requirement of approval by a law. In its Opinion of 10 July 1956 the Conseil d’État 15 See ‘Approbation des traites’ in Marc Besch, Traité de légistique formelle (Luxembourg, Conseil d’Etat du GrandDuché de Luxembourg, 2005) 143. 16 Doc Parl No 25 (516), CCIV.

Luxembourg  469 underlined the distinction between the enforceability of the approving law and of the international treaty. They are two different procedures, which may take place simultaneously or separately, but both need to be compatible with the corresponding constitutional provisions. The Conseil d’État adopted, finally, one formal requirement, that of the publication of the approving law. Article 46 of the proposed draft Constitution contains a similar provision to the current Article 37 regarding the ratification of international treaties. Instead of naming the Grand Duke, the provision states that the ‘Head of State’ shall be responsible for concluding international treaties. The draft provision includes a reference to the procedure for terminating a treaty, similar to the procedure foreseen for the conclusion of a treaty, which is not included in the current version of Article 37. The new draft article also states explicitly that the Head of State shall adopt all necessary regulations for the application of EU law in Luxembourg.

D.  Constitutional Amendments Due to EU Integration 1.  Explicit Amendments As the principle of primacy of international law is taken very seriously in the Grand Duchy, the Chambre has repeatedly affirmed its will to prevent discrepancies between international and national law by adapting the respective domestic rules – including if necessary by an amendment of the Constitution. The following four cases show to what extent EU integration has provoked constitutional amendments in Luxembourg. a.  Introduction of Article 49bis and Revision of Article 37 in 1956 Article 49bis was introduced into the Constitution in 1956 in the context of the European integration process. The ECSC and the EDC treaties raised the pivotal question of whether the Luxembourgish legislature, executive and judiciary may transfer powers to international institutions. The Government considered that the existing Constitution would cover the transfer of powers. The Conseil d’État disagreed, but decided not to oppose approval of the Treaties as a ‘matter of expediency’, although it urged the necessity of immediate revision of the Constitution. Thereupon, both Treaties were approved respectively on 23 June 1952 and 24 April 1954, whilst the Constitution was revised only in 1956 with the introduction of Article 49bis: ‘The exercise of the powers reserved by the Constitution to the legislature, executive, and judiciary may be temporarily vested by treaty in institutions governed by international law.’ A proposal made in 2009 to introduce a complete new chapter on the EU was finally not adopted. b.  Articles 9 and 107(2) and (4): Right to Active and Passive Voting in Municipal and European Elections Granted to all EU Citizens in 1994 In order not to risk the dissolution of the Chambre because of a revision of the Constitution pursuant to Article 114, the Conseil d’État argued that Article 8B(1) Maastricht Treaty would not be in conflict with the Constitution because it would not immediately grant rights that individuals could directly invoke before any tribunal (see the second sentence of the Article: ‘This right shall be exercised subject to detailed arrangements to be adopted before 31 December 1994 by the Council, acting unanimously’). Articles 9 and 107 were only revised in 1994. Consequently, there were two years of incompatibility between European and Luxembourgish law, entailing

470  Jörg Gerkrath discussions about the compatibility of the Maastricht Treaty with the Constitution and the procedure of the revision of the Constitution in virtue of Article 114. A new paragraph (3) was added to Article 9: ‘the law may confer the exercise of political rights to non-Luxembourgers’. Article 9(3) confers henceforth to non-Luxembourgers the exercise of political rights, ie the ‘cardinal political right’ of active and passive voting. c.  Former Article 11(2), Now Article 10bis(2): Access to all Posts in the Public Sector for EU Citizens in 1999 The Constitutional legislator decided to revise the Constitution in the aftermath of a decision of the ECJ in 1996. The ECJ held that the general prohibition for non-Luxembourger to work in the public service exceeded the limits of exception provided for in Article 48(4) EC. In not complying with its obligation ‘to open the areas in question to nationals of other Member States by restricting application of the nationality condition to only those posts which actually involve direct or indirect participation in the exercise of powers conferred by public law and duties designed to safeguard the general interest of the State or of other public authorities’, Luxembourg failed its obligations under the Treaty.17 In reaction to a revision draft introduced by the President of the Commission for constitutional Revision, the Conseil d’État did not agree that a revision would be necessary because the Constitution would not formally restrict the access of non-Luxembourgers to employment in the public service and, hence, would not be in contradiction of Article 48(4) of the Maastricht Treaty. Thus, no revision was adopted. However, when the Conseil d’État delivered its recommendation on a draft law opening the public service to EU citizens, it accepted the argument that this draft required a preceding revision of Article 11(2) (now Article 10bis(2)). As a result, Article 11(2) was revised in April 1999. d.  Article 114: Constitutional Revision Procedure In the context of the 1994 revision, there was an attempt to change the procedure of constitutional revision pleading in favour of a duality of procedures: where an article of the constitution would be in conflict with international treaties’ provision, an alleviated procedure should apply. However, the Conseil d’État argued in favour of a unique revision procedure. In 2003, Article 114 was finally revised (following the argument of uniqueness of the procedure), ie renouncing the requirement of dissolution of the Chambre.

2.  Implicit Amendments The Constitution does not determine the value of international treaties within the domestic legal order. Courts and scholars acknowledge, however, full primacy of treaties over national law, including the Constitution.18 According to its founding statute, the Conseil d’État monitors each draft bill for compliance with all rules of ‘higher law’. The latter category includes the Constitution, treaties and general principles of law. In addition, the statute of the Constitutional Court explicitly excludes treaty-approving laws from constitutional review. This is often justified

17 ECJ 02.07.1996 C-473/93 (Commission v Luxembourg) ECLI:EU:C:1996:263. 18 See Patrick Kinsch, ‘Le rôle du droit international dans l’ordre juridique luxembourgeois’, (2010) 36 Pasicrisie luxembourgeoise, 383.

Luxembourg  471 with reference to the primacy of treaties. As a result, the conclusion of a treaty by the Grand Duchy may induce constitutional change without revising the Constitution. This is for instance the case when human right treaties are adding rights to those guaranteed by the Constitution itself or are modifying their scope. Treaties conferring competences, which the Constitution vested in domestic constitutional bodies, to international organisations are subject to a special regime. Their parliamentary approval requires – due to Article 37 – the same qualified majority (of two-thirds) within the Chambre as the vote of a constitutional revision act. Each conferral of competences to the EU, thus, amounts to an informal change of the Constitution.

E.  Limits to European Integration Outside the EU Legal Order There are strictly no material limits to any enterprises of European integration outside the existing treaty framework. The only formal limit foreseen by the Constitution is the requirement of a qualified majority within the Chambre if powers need to be transferred to international bodies. The national laws approving the participation in the rescue mechanisms (EFSM, EFSF, ESM) all passed by a large majority in the Chambre, the only exception being the Fiscal Compact that raised more debate and passed by a smaller majority (46 out of 60 MPs). Luxembourg even gained some ‘benefits’ from the euro crisis. First, the EFSF was established as a public limited liability company under the laws of the Grand Duchy of Luxembourg and, second, the EFSF as well as the European Stability Mechanism (ESM), which is an intergovernmental organisation, are both located in Luxembourg City.

IV.  Constitutional Limits to EU Integration There are no limits ratione materiae with regard to the extent to which powers can be transferred to the EU, and there are no such limits to constitutional amendments in general. Neither the text of the Constitution, nor case law, nor constitutional commentary refers to such limits under Luxembourgish law. Although the Constitution does not formulate any explicit material restriction to the Chambre’s power of constitutional revision, there are two provisions of the Constitution that ought to be mentioned. According to Article 113, ‘no provision of the Constitution may be suspended’. This provision can be considered, on the one hand, as a kind of corollary of the principle that the Constitution can only be revised following the procedure described by it in Articles 114 and 115. On the other hand, it also tends to prohibit derogations to the constitutional functioning of the state and its organs during exceptional situations. Article 113 could also be construed as a prohibition to revise the revision procedure in such a way that it would be identical to the ordinary legislative procedure. The idea of a ‘suspended constitution’ had also been mobilised in the context of the ratification of the Maastricht Treaty.19 It was criticised that ratification intervened without a prior revision (of Article 9 of the Constitution) introducing the right to vote for EU citizens. Some lawyers argued that the Constitution should have been revised before the ratification of the Treaty. The Conseil d’État considered that the Treaty did not directly organise the exercise



19 Alex

Bonn, ‘La Constitution suspendue’, Letzebuerger Land No 24 of 12 June 1992.

472  Jörg Gerkrath of these rights but required the adoption of a directive. Finally the Constitution was revised in December 1994. During the period between the entry into force of the Treaty and the final revision of the Constitution, the latter was somehow suspended. Article 115 provided initially that ‘no change in the Constitution can be done during a regency’. It was revised in 1998 in the sense that from then on, only changes in ‘the constitutional prerogatives of the Grand Duke, his status and the order of succession’ are prohibited. The restriction laid down in Article 115 is simply a limitation ratione tempori not ratione materiae. One can also wonder whether this provision has not lost its purpose since the revision of Article 34 of the Constitution, in which the Grand Duke’s power to sanction laws was removed. Now the Grand Duke will never be able to oppose a revision of these powers, either during regency or in ordinary times.

A.  Limits to the (Further) Transfer of Powers to the EU Through Treaty Amendments The only hypothetical constitutional limit to EU membership is to be found in the ‘Independence clause’ of Article 1, providing that Luxembourg is ‘an independent state’.

1.  ‘Core Competences’ The idea of ‘core competences’ whose exercise should remain in the hands of the domestic constitutional organs has never been introduced in domestic constitutional law. Luxembourg’s Constitution has never been based on a conception of absolute sovereignty. In the absence of any clause on the transfer or delegation of sovereignty, the Conseil d’État stated already in 1952 that ‘a state may and must renounce certain parts of its sovereignty if the public good, the ultimate purpose of the state’s organisation, requires it’.20 As the Constitutional Court may not review the constitutionality of Acts approving international treaties, there is no pertinent case law from this Court. One of the particular characteristics of Luxembourg’s domestic legal order lies in the fact that its very existence results from international law. Established as an independent state by the Final Act of the Congress of Vienna of 9 June 1815, the Grand Duchy’s independence was confirmed by the Treaties of London of 19 April 1839 and 11 May 1867. Therefore, far from constituting a threat to national sovereignty, international law is understood in Luxembourg as a ‘vital guarantee of the existence and survival of the state’.21 Moreover, Luxembourg’s courts have had no difficulty in recognising the pre-eminence of international law and the primacy of EU Law, including in respect of a constitutional provision.22 This state of affairs also explains why it was not considered necessary to write a provision into the Constitution that would more explicitly authorise transfers of competences to the Union.

20 Opinion on the Ratification of the ECSC Treaty, doc parl n° 395/2, p 3. 21 Georges Wivenes, ‘Le droit européen et les constitutions nationales, Rapport luxembourgeois’, in Lord Slynn of Hadley and M Andenas (eds), FIDE XX. Congress vol 1 (London, BIICL, 2002) 267. 22 Kinsch, ‘Rôle du droit’ (n 18) 399; Pierre Pescatore, ‘L’autorité en droit interne, des traités internationaux’, (1962) 18 Pasicrisie luxembourgeoise 97.

Luxembourg  473

2.  The ‘Independent State’ Clause The ‘independent state’ statement appeared in the same place in the previous constitutional texts of 1848 and 1856, which, however, put this independence into perspective by going on to state that the Grand Duchy was ‘part of the German Confederation’. The 1856 text went even further by stating that the Grand Duchy ‘participates in the rights and obligations arising from the Federal Constitution. These rights and these obligations cannot be derogated by the internal legislation of the country.’ The current constitutional declaration of independence has never stopped the Grand Duchy from participating in integration exercises with a constitutional dimension. Moreover, at no time has the preservation of the independence of the state ever been seriously discussed as a possible limit to Luxembourg’s participation in the European integration process.23 The successive incorporation of the Grand Duchy in the German Confederation (1815–66), the Zollverein (1842–1919), the Belgium-Luxembourg Economic Union (from 1921) and Benelux (from 1944) have, on the contrary, enabled it to acquire the experience indispensable for being prepared for the legal implications of its membership of the EC and the EU. As part of the current overhaul, though, the Constitution is envisaged to feature a new Article 5 stating that the Grand Duchy ‘participates in the European Union’ and that ‘the exercise of the State’s powers may be transferred to the European Union … by an act passed by qualified majority.’ Luxembourg’s ‘Member State status’ will then be anchored in constitutional law. We can only regret that it is not in Article 1 that this European clause is intended to be included, as it was the case in the past with membership of the German Confederation.

3.  Respect of Certain Constitutional Features at EU Level As a Member State of the Council of Europe, Luxembourg always made efforts to fully respect the basic principles of democracy, the rule of law, and the protection of fundamental rights in its internal legal order. But due to the age of its Constitution, these principles have rather been implemented in Luxembourg as a result of its participation in European integration in the widest sense, than the other way around. Thus, no requirements regarding the respect of these features by the EU institutions can be identified in domestic law. The only elements that could become an issue are the preservation of the Luxembourgish social model and the multilingualism within the public sector. Both policies represent potential restrictions to free movement rights of EU citizens and the Grand Duchy considers them as part of its constitutional identity in the sense of Article 4 TEU.

B.  Scrutiny of Secondary Legislation, Especially Ultra Vires Doctrine 1.  Against the Yardstick of National Constitutional Requirements The scrutiny of secondary legislation against the yardstick of constitutional requirements is made impossible in Luxembourg by means of the domestic rules of procedure. In the case of conflict of an international or European engagement with the Constitution or legislative acts, the position of the domestic authorities is that national standards should be subject



23 Wivenes,

‘Droit européen’ (n 21).

474  Jörg Gerkrath to revision or amendment before the international commitment is approved. Once approved, the respective international norms enjoy, in the pure monistic tradition, full primacy over rules of domestic law, even of constitutional value.24 This rule also applies to the secondary legislation of the EU.25 All (civil and administrative) courts have accepted the full supremacy and direct effect of EU law in the very terms of the CJEU’s case law, to which they regularly refer. Acts of the Chambre approving treaties are explicitly excluded from the competence of the Constitutional Court (Art 95ter(2)). In Luxembourg, this is considered as a consequence of the primacy of international treaties. Acts of the Chambre transposing or executing obligations deriving from secondary EU legislation are not explicitly excluded from the Constitutional Court’s jurisdiction. Until now, no such Act has however been submitted to the Constitutional Court as a preliminary question by an ordinary court. The opinion seems to prevail that the ‘immunity’ of Acts approving treaties also covers Acts implementing secondary EU legislation.

2.  Against the Yardstick of EU Law There is no reported case where domestic courts would have directly scrutinised the legality of EU secondary legislation under rules of EU primary law. In case of doubt, Luxembourg would initiate an action for annulment and/or the domestic courts would bring the matter before the CJEU as a preliminary question on validity. Actions of annulment have nonetheless remained very exceptional. The database of the CJEU contains only eight cases in which the Grand Duchy sought to obtain the annulment of EU secondary legislation.26 Three of them concerned measures of the Parliament regarding its seat or working place; two were directed against Commission decisions concerning the clearance of EAGGF accounts; and two were directed against EU directives. Article 13 of the Constitution states that ‘no-one may be reassigned against his will to a court other than that designated by law’. This includes the CJEU, which must be considered a ‘legal court’ in the meaning of the Constitution. This raises the ultimate question of whether the obligation of the Luxembourg court to make a reference for preliminary ruling to the CJEU may, where appropriate, be sanctioned by domestic law. The only remedy available under Luxembourg law would then be to introduce an action for liability against the state.27 Speaking generally, the statistics show that Luxembourg’s courts, which do not seem to experience any great difficulty in applying and interpreting Union law, are among those sending the lowest number of referrals for preliminary ruling to the CJEU.28 In the 55 years from 1963 to September 2018, Luxembourg courts introduced altogether 92 preliminary rulings, but only one on validity.29 24 Cour d’appel, arrêt du 13 novembre 2001, n° 396/01V, 12 Annales du droit luxembourgeois 2002, 456. Cour supérieure de justice (assemblée générale), arrêt du 5 décembre 2002, n° 337/02, 13 Annales du droit luxembourgeois 2003, 683. 25 Conseil d’Etat, 21 novembre 1984, 26 Pasicrisie luxembourgeoise 1985, 174. 26 ECJ 10.02.1983 230/81 (Luxembourg v Parliament) ECLI:EU:C:1983:32; ECJ 12.07.1984 49/83 (Luxembourg v Commission) ECLI:EU:C:1984:268; ECJ 10.04.1984 108/83 (Luxembourg v Parliament) ECLI:EU:C:1984:156; ECJ 28.11.1991 joined cases 213/88 and C-39/89 (Luxembourg v Parliament) ECLI:EU:C:1991:449; ECJ 07.11.2000 C-168/98 (Luxembourg v Parliament and Council) ECLI:EU:C:2000:598; ECJ 13.06.2002 C-158/00 (Luxembourg v Commission) ECLI:EU:C:2002:367; and CJEU 12.05.2011 C-176/09 (Luxembourg v Parliament and Council) ECLI:EU:C:2011:290. 27 Cf Georges Ravarani, Rapport luxembourgeois au Colloque des Conseils d’État du Benelux, 10 October 2013, La transposition et la mise en œuvre des actes normatifs de l’Union européenne en droit national, www.conseil-État.public.lu/fr/ actualites/2013/10/C_Colloque_Benelux/index.html. 28 Cf also Jean-Paul Hoffmann, ‘Les questions préjudicielles posées par les juridictions luxembourgeoises à la CJCE’, in R Iglesias et al (eds), Mélanges en hommage à Fernand Schockweiler (Baden-Baden, Nomos, 1998) 5; and Manou Weirich, ‘L’application du droit communautaire au Grand-Duché de Luxembourg’, in Comité de la Conférence Saint-Yves (ed), Diagonales à travers le droit luxembourgeois: livre jubilaire de la Conférence de Saint-Yves 1946-1986 (Luxembourg, Saint Paul, 1986) 982. 29 ECJ 25.06.1997 C-114/96 (Kieffer and Thill) ECLI:EU:C:1997:316.

Luxembourg  475 There are unfortunately no reliable statistics available in Luxembourg concerning the number of cases in which the applicants have requested a preliminary ruling with regard to the validity of EU measures. Between 2000 and 2018, only two cases were identified after a thorough research of the case law of the respective civil, social and administrative courts. First, in a case brought to the Conseil supérieur des assurances sociales, the applicant challenged the validity of Regulation 1408/71 (on the cooperation between social security systems) with regard to former Article 48(2) EC (now Article 45(2) TFEU) on the free movement of workers. It was argued that the application of the residence clause in Article 71 of that Regulation constitutes indirect discrimination on grounds of nationality with regard to unemployed frontier workers.30 Second, in a case brought to the tribunal administratif, the applicants challenged the validity of Directive 2003/87 (from 13 October 2003, establishing a scheme for greenhouse gas emission allowance trading within the Community) based on the ground that the Directive would not sufficiently take into account the specific situation of the steel industry within global competition.31 The courts did not submit a reference for a preliminary ruling to the ECJ in either of these cases.

V.  Constitutional Rules and/or Practice on Implementing EU Law As there are currently no constitutional rules regarding EU membership in general, the reader will not be astonished to learn that there are also no constitutional rules on implementing EU law. Ordinary law, practice, and the internal rules of the Chambre des Députés deal with the matter.

A.  Binding (Parliamentary or Governmental) Resolutions for the Ministers in the Council? Luxembourg’s Constitution does not contain specific provisions concerning the rules governing the participation of the Chambre in EU affairs. Although Prime Minister Juncker suggested such an amendment in the course of the ratification procedure of the Lisbon Treaty in 2008, the Chambre rejected his proposal. It allegedly feared that it was not sufficiently staffed to face this responsibility. Thus it is a simple ‘Memorandum on Cooperation between the Chambre and the Government of the Grand Duchy of Luxembourg in the Field of EU Policy’, which is annexed to the rules of procedure of the Chambre that governs the rights and duties of both institutions in this respect.32 Introduced on 2008, this memorandum contains in particular rules on the right of the Chambre to be informed on issues of European policy. The Lisbon Treaty recognises the right of national parliaments to contribute actively to the good functioning of the Union (Article 12 TEU). From the review of official documents of the European institutions to the transposition of directives into national law, the action of the Chambre in the political sphere of the EU is, thus, manifold. National parliaments are indeed 30 Arrêt du 3 décembre 2007, ADEM 2006/0094. 31 Arrêt du 5 avril 2006, nos 20372 et 20373 du rôle. 32 Aide-mémoire sur la coopération entre la Chambre des députés et le Gouvernement du Grand-Duché de Luxembourg en matière de politique européenne.

476  Jörg Gerkrath involved by monitoring the activity of their respective national governments at the European level. In the Grand Duchy of Luxembourg, a parliamentary committee invites the ministers to its sessions before and after Council meetings. The Chambre contributes to the monitoring of respect of the subsidiarity principle laid down in the Treaty of Lisbon.33 Beyond these two main powers, the Chambre participates in interparliamentary cooperation within the Union. Inter-parliamentary meetings are held mainly within the framework of the Conference of Parliamentary Committees for Union Affairs of Parliaments of the European Union (COSAC) and the Conference of Presidents of Parliaments of the EU. Information on the analysis of EU documents is transmitted between national Parliaments through the platform for EU Interparliamentary Exchange (IPEX) database. The control exercised by the Chambre over the Government in the field of European affairs appears altogether rather modest compared to practices in other Member States. In EU affairs as in domestic affairs the Chambre monitors the Government’s activities mostly by ordinary means of parliamentary questions, the scrutiny of government reports or, if necessary, by creating investigation committees on specific issues.

B.  Participation of the National Parliament in Secondary Law Making The participation of the Chambre des Députés in European affairs has developed outside of the Constitution. Nevertheless, its ‘European function’ is no less real.34 Under Article 12 TEU, all national parliaments are, in effect, called upon to actively contribute to the good functioning of the Union, including ‘by seeing to it that the principle of subsidiarity is respected in accordance with the procedures provided for in the Protocol on the application of the principles of subsidiarity and proportionality’.35 The scrutiny they exercise in this regard on compliance with subsidiarity criteria can be connected both to their role of controlling national executives and to their participation in the exercise of European legislative power. Once again, membership of the Union brings with it the expansion of the powers of a constitutional organ.

C.  Techniques of Implementing Secondary Legislation The transposition and implementation of EU measures can be realised in Luxembourg by acts of Parliament or Grand Ducal decrees. Doing so by Grand Ducal decrees is undeniably seen as a way to increase the speed of implementation, as well as the Grand Duchy’s ranking on the European Commission’s scoreboards. Every year, the Government as well as the Chambre proudly announce progress made in this respect. The matter is currently determined by an Enabling Act (loi d’habilitation) adopted by the Chambre in 1971.36 This Act, however, is limited in scope and

33 See Martin Gennart, Le contrôle parlementaire du principe de subsidiarité. Droit belge, néerlandais et luxembourgeois (Brussels, Larcier, 2013). 34 Cf Martina Mayer, Die Europafunktion der nationalen Parlamente in der Europäischen Union (Tübingen, MohrSiebeck, 2012) and Patrick Dumont and Astrid Spreitzer, ‘The Europeanization of Domestic Legislation in Luxembourg’, in S Brouard, O Costa, T König (eds), The Europeanization of Domestic Legislatures, The Empirical Implications of the Delors’ Myth in Nine Countries (New York/Dordrecht/Heidelberg/London, Springer, 2012) 131. 35 Cf on this point Gennart, ‘Contrôle parlementaire’ (n 33). 36 Loi du 9 août 1971 concernant l’exécution et la sanction des décisions et des directives ainsi que la sanction des règlements des Communautés européennes en matière économique, technique, agricole, forestière, sociale et en matière de transports, Texte coordonné, in Code administratif, 2013, vol 4(as amended).

Luxembourg  477 has not really proven an ‘appropriate and effective response’ to the problems encountered.37 The Draft of the current constitutional amendment procedure foresees the addition of an Article to the Constitution that will explicitly give the executive power the competence to transpose and implement EU measures by Grand Ducal decree. Parliamentary Acts will, thus, lose some of their importance in national legislation because of the perceived need to streamline the procedures for the transposition and implementation of secondary EU legislation. The inclusion of a new Article 47 in the Constitution, as set out in the 2018 Draft and which obligates the Head of State to adopt all the necessary regulations for the application of legally binding acts of the EU, will eradicate a number of legal uncertainties in this area. According to the memorandum on cooperation between the Chambre des Députés and the Government on European policy (in force since 1 July 2008), the Government undertakes to submit annually, during the first semester, a report to the Chambre on the transposition of EU directives and the application of the law of the EU. The 2018 report is the twelfth report on the transposition of EU directives. This is, as in previous editions, a public report. Over the last decade, Luxembourg has gradually managed to improve its results in the transposition of EU directives. The transposition deficit of Luxembourg, which already showed a significant downward trend, from 4 per cent (in May 2005) to 1.1 per cent (in May 2015), is currently turning around the 1 per cent goal set by the European Council in March 2007.

D.  Enforcement Through the Courts, Including the Protection of the EU Charter of Fundamental Rights All courts have accepted the full supremacy and direct effect of EU law in the very terms of the CJEU. In the field of fundamental rights there is a long tradition of directly applying the provisions of the ECHR and since the entry into force of the Lisbon treaty the domestic courts frequently refer to the Charter of Fundamental Rights of the EU (CFR). In a recent case the administrative court even proceeded to an ex officio application of the CFR.38

VI.  Relationship between EU Law and National Law One of the characteristics of Luxembourg’s domestic legal order lies in the fact that, unlike other European countries, it is founded on international law. Established as an independent state by the Final Act of the Congress of Vienna of 9 June 1815, the Grand Duchy’s independence was confirmed by the Treaties of London of 19 April 1839 and 11 May 1867. Therefore, far from constituting a threat to national sovereignty, international law is understood in Luxembourg as a ‘vital guarantee of the existence and survival of the state’.39 Moreover, Luxembourg’s courts have had no difficulty in recognising the pre-eminence of international law and the primacy of Community Law, including in respect of a constitutional provision.40 This state of affairs also explains why it was not considered necessary to write a provision into the Constitution more explicitly authorising transfers of competences to the Union. 37 Paul Schmit, Rapport du Conseil d’État, Colloque des Conseils d’État du Benelux, 10 octobre 2013. La transposition et la mise en œuvre des actes normatifs de l’Union européenne en droit national, 5; and Weirich, ‘L’application’ (n 28) 65. 38 Judgment of 17.12.2015, 36893C. 39 Wivenes, ‘Droit européen’ (n 21). 40 Cf Pescatore, ‘L’autorité’ (n 22) 97; Kinsch, ‘Rôle du droit’ (n 18) 383.

478  Jörg Gerkrath Under Luxembourg law, the Constitution long ago ceased to be the only supreme law. There is now a set of norms ranked supra-legislative and designated by the term ‘higher law.’ It is within the remit of the Conseil d’État to monitor the compliance of draft bills and draft regulations with the rules of higher law. The amended Act of 12 July 1996, reforming the Conseil d’État, states in its Article 2(2), that if it considers a draft bill to be contrary to the Constitution, to international agreements and treaties, or to the general principles of law, the Conseil d’État shall state this in its Opinion. It shall do the same, if it considers a draft regulation to be contrary to a rule of higher law.

As part of the ongoing constitutional revision procedure, this ex ante check on compliance with ‘higher law’ will be included in Article 88 with the addition of ‘legally binding decisions of the European Union’. Article 96 is also planned to state that ‘courts shall only apply acts and regulations to the extent that they comply with rules of higher law’. The Constitution does not contain any explicit rule on the legal value of international or EU Law within the domestic legal order. Nonetheless, according to well-settled case law and the position of the Conseil d’État as well as Luxembourgish scholars, self-executing international treaties enjoy full primacy with regard to the provisions of internal law, including the Constitution itself.41 The case law on this point was developed from the early 1950s when first the Cour de cassation (the highest civil court) and subsequently the Conseil d’État (as the former highest administrative court) reversed the previously defended position that judicial control of the compliance of parliamentary Acts with treaties was not possible because of the principle of separation of powers.42 According to the reference decision of the Conseil d’État in 1951, an international treaty incorporated into domestic law by a law of approval is law of superior essence having a higher origin than the will of an internal organ. It follows that in the case of conflict between the provisions of an international treaty and those of a subsequent national law, international law must prevail over national law.43

The wording of this decision is clearly very wide, as the judgment states without distinction that an international norm prevails over the will of any national organ. It must be noted, however, that in 1956 the Chambre expressly rejected a governmental constitutional amendment bill, which provided that ‘[t]he rules of international law are part of the national legal order. They supersede all other law and national provisions’. According to this draft, primacy would have encompassed constitutional provisions. The Conseil d’État, however, implicitly accepted such a general primacy in an opinion of 26 May 1992 on the draft Act Approving the EU Treaty. It was considered that it should be borne in mind that under the rule of the hierarchy of legal norms, international law takes precedence before national law and, in the case of conflict, the courts shall dismiss domestic law in favour of the Treaty. As it is important to avoid a contradiction between national law and international law, the Conseil d’État insists that the related constitutional amendment take place within due time to prevent such a situation of incompatibility.

In the case of conflict of an international or European engagement with the Constitution or legislative acts, national standards should be subject to amendment before the competent national authorities approve the international commitment. Once approved, the respective international 41 Cf Kinsch, ‘Rôle du droit’ (n 18) 399. 42 Cour de cassation, arrêts des 8 juin 1950, Pasicrisie lux. 15, 41, et 14 juillet 1954, Chambre des métiers c Pagani, Pasicrisie lux. 16, 151; JT 1954, 694, note Pescatore. 43 Conseil d’Etat, (Comité du contentieux), 28 juillet 1951, Dieudonné c Administration des contributions, Pasicrisie luxembourgeoise vol XV, 263.

Luxembourg  479 norms enjoy, in the pure monistic tradition, full primacy over rules of domestic law, even of constitutional value.44 This rule also applies to the secondary legislation of the EU.45

VII.  Miscellaneous: The Current Constitutional Amendment Procedure As mentioned above, the Constitution of the Grand Duchy, one of the oldest constitutional documents in Europe still in force, is currently undergoing a far-reaching revision procedure aiming at a general overhaul.46 According to the parliamentary committee in charge, this revision will finally give birth to a ‘new’ Constitution, meaning that a modified and updated edition of the Constitution shall be published in the official journal and the current Constitution be repealed. The present amendment procedure started officially in April 2009 with the submission of a formal revision proposal by the parliamentary committee on constitutional amendment.47 The proposal aspires to realise a general renovation of the Constitution and to rearrange the whole document. It has been evaluated in depth by the Venice Commission, which delivered an opinion in December 2009, and by the Conseil d’Etat, which published its extensive opinion in June 2012.48 This text of 216 pages appears like a ‘counter proposal’, completed by a comparative table confronting the current wording of the Constitution with the initial revision proposal on one side and the version favoured by the Conseil d’Etat on the other. In the light of this opinion the parliamentary committee profoundly modified its initial proposal and established in January 2013 an internal working document containing a first consolidated version containing all the future amendments to the Constitution. After the break-up of the coalition government formed by the conservative Christian Democrats of Jean-Claude Juncker (CSV) and the Social Democrats (LSAP), the Chambre was dissolved by the Grand Duke and parliamentary elections took place on 20 October 2013. Though the CSV remained, with 23 seats, the strongest party in the Chambre, three smaller parties formed a new coalition government. This coalition of the Democratic Party (DP), led by the new Prime Minister Xavier Bettel, the LSAP and the Green Party (Dei Greng) is supported by a slim majority of 32 seats out of 60. Under the new majority, the renewed committee on constitutional revision went back to work and managed to agree on most of the chapters and provisions of the draft Constitution during 2014. In March 2015 the committee’s draft was published on the website of the Chambre within a special page dedicated to the referendum held on 7 June 2015.49 The work of the parliamentary committee could be followed based on the minutes of its numerous meetings, which are regularly published on the Chambre’s website.50 These minutes show a strong tendency of the committee to find solutions agreeable to a large majority within the

44 Cour d’appel, arrêt du 13 novembre 2001, n° 396/01 V, 12 Annales du droit luxembourgeois 2002 456; Cour supérieure de justice (assemblée générale), arrêt du 5 décembre 2002, n° 337/02, 13 Annales du droit luxembourgeois 2003, 683. 45 Conseil d’Etat, 21 novembre 1984, Pasicrisie lux 26, p 174. 46 Jörg Gerkrath, ‘Some remarks on the pending constitutional change in the Grand Duchy of Luxembourg’, (2013) 13 European Public Law 449. 47 See ‘Proposition de révision portant modification et nouvel ordonnancement de la Constitution’ of 21 April 2009, doc parl N° 6030, www.chd.lu. 48 Avis du Conseil d’Etat du 6 juin 2012, no 48.433, doc parl 6030/06. 49 See www.referendum.lu/Uploads/Nouvelle_Constitution/Doc/1_1_PropositionNouvelleConstitution.pdf. 50 See www.chd.lu/wps/portal/public/RoleEtendu?action=doDocpaDetails&backto=/wps/portal/public&id=6030.

480  Jörg Gerkrath Chambre. The document, which had the highest impact on the work of the committee was clearly the opinion delivered by the Conseil d’Etat in June 2012. Opinions and positions introduced by other bodies such as the judiciary, the Venice Commission or the professional chambers only played a secondary role. The question whether the amended Constitution should be submitted to a referendum was often raised in the past but the different political parties changed their positions over time. However, since December 2013, the project to consult the Luxembourgish people on the new Constitution reappeared vigorously on the agenda of the coalition government. A first referendum was, thus, scheduled on 7 June 2015 by an Act of 27 February 2015 ‘on the organisation of a national referendum on different questions related to the working out of a new Constitution’.51 The members of the parliamentary majority, which authored the proposal of this Act, justify this first referendum, based on Article 51(7) of the Constitution, by the need to consult the people on several essential questions that have remained controversial during the preparatory work of the draft Constitution.52 Both, the authors of the proposal and the Conseil d’Etat insist on the point that this referendum is purely consultative and must not be confused with the referendum foreseen by Article 114 of the Constitution.53 It is a consultative referendum scheduled well before the first vote of the Chambre on the final text. Though the outcome of this referendum is not be legally binding on the Chambre, it will strongly influence the final discussions within the parliamentary committee on constitutional amendment, which has to ‘translate’ this outcome into constitutional provisions. The initial proposal intended to submit four questions to the referendum: first, whether the voting age should be lowered to 16; second, whether foreigners should be allowed to vote in the Chambre elections; third, whether mandates of members of the Government should be limited to 10 years; and fourth, whether the obligation of the state to pay the salaries and pensions of religious officials should end. Due to an agreement reached between the state and the religious communities and a consensus reached within the parliamentary committee to introduce a new article in the draft Constitution on the neutrality of the state in religious matters and the separation between the state and the religious communities, the fourth question was eventually withdrawn from the proposal.54 The result of the referendum has been very clear. All three questions have been answered by ‘No’ with majorities ranging between 70 and 80 per cent. Hence all three reform projects have been withdrawn from the text of the draft Constitution. The final text of the draft is now accessible within the final report of the parliamentary committee for constitutional reform of 6 June 2018. A second decisional referendum, planned for 2019, well after the parliamentary elections of October 2018, was to be organised in order to approve the new Constitution by a direct popular consultation rather than a second vote within the Chambre. As this referendum was announced by both the current and the former governmental majority, it is very likely that the members of the Chambre will vote in favour of convening it. The necessary number of at least 16 members of the Chambre asking for it, as foreseen by Article 114 of the Constitution, should easily be reached.

51 Loi du 27 février 2015 portant organisation d’un référendum national sur différentes questions en relation avec l’élaboration d’une nouvelle Constitution, Mémorial A-35, 4 March 2015, 358. 52 See Proposal of an act on the organisation of a national referendum on issues related to the making of a new Constitution, doc parl 6738, 2. 53 See the Opinion of the Conseil d’Etat of 13 January 2015, doc parl 6738/1, 2. 54 See doc parl 6738/2 of 23 January 2015.

Luxembourg  481

References M Besch, Traité de légistique formelle (Luxembourg, Conseil d’Etat du Grand-Duché de Luxembourg, 2005). A Bonn, La Constitution oubliée (Luxembourg, Imprimerie Centrale, 1968). A Bonn, ‘La Constitution suspendue’, Letzebuerger Land No 24 of 12 June 1992. P Dumont and A Spreitzer, ‘The Europeanization of Domestic Legislation in Luxembourg’, in S Brouard, O Costa, T König (eds), The Europeanization of Domestic Legislatures, The Empirical Implications of the Delors’ Myth in Nine Countries (New York/Heidelberg, Springer, 2012) 131. B Fayot, Les quatre référendums du Grand-Duché de Luxembourg. Essai (Luxembourg, Éditions de la Petite Amérique, 2006). L Frieden, ‘Luxembourg: Parliament abolishes royal confirmation of laws’, (2009) 7 International Journal of Constitutional Law 539. M Gennart, Le contrôle parlementaire du principe de subsidiarité. Droit belge, néerlandais et luxembourgeois (Brussels, Larcier, 2013). J Gerkrath (ed), La jurisprudence de la Cour Constitutionnelle du Luxembourg 1997–2007 (Luxembourg, Pasicrisie luxembourgeoise, 2008). J Gerkrath, ‘Some remarks on the pending constitutional change in the Grand Duchy of Luxembourg’, (2013) 13 European Public Law 449. J Gerkrath (ed), Les 20 ans de la Cour Constitutionnelle: trop jeune pour mourir? (Luxembourg, Pasicrisie luxembourgeoise, 2018). L Heuschling, Le citoyen monarque. Réflexions sur le Grand-Duc, la famille grand-ducale et le droit de vote (Windhof, Promoculture Larcier, 2013). L Heuschling and Ph Poirier, ‘L’opposition politique au Grand-duché de Luxembourg. Regards croisés, de science juridique et de science politique, sur une démocratie consociative’, in JP Derosier (ed), L’opposition politique (Paris, Lexis Nexis, 2016) 225. JP Hoffmann, ‘Les questions préjudicielles posées par les juridictions luxembourgeoises à la CJCE’, in R Iglesias et al (eds), Mélanges en hommage à Fernand Schockweiler (Baden-Baden, Nomos, 1998) 5. HU Jessurun d’Oliveira, ‘The EU and Its Monarchies: Influences and Frictions’, (2012) 8 European Constitutional Law Journal 63. P Kinsch, ‘Le rôle du droit international dans l’ordre juridique luxembourgeois’, (2010) 36 Pasicrisie luxembourgeoise, 383. M Mayer, Die Europafunktion der nationalen Parlamente in der Europäischen Union (Tübingen, MohrSiebeck, 2012). P Pescatore, ‘L’autorité en droit interne, des traités internationaux’, (1962) 18 Pasicrisie luxembourgeoise 97. JH Reestman, ‘The state of the European Union’s monarchies. An introduction to the series’, (2011) 7 European Constitutional Law Journal 270. W Roobol, ‘Twilight of the European Monarchy’, (2011) 7 European Constitutional Law Journal 286. P Schmit, Rapport du Conseil d’État, Colloque des Conseils d’État du Benelux, 10 octobre 2013, La transposition et la mise en œuvre des actes normatifs de l’Union européenne en droit national, www.conseil-État.public. lu/fr/actualites/2013/10/C_Colloque_Benelux/index.html. M Weirich, ‘L’application du droit communautaire au Grand-Duché de Luxembourg’, in Comité de la Conférence Saint-Yves (ed), Diagonales à travers le droit luxembourgeois: livre jubilaire de la Conférence de Saint-Yves 1946-1986 (Luxembourg, Saint Paul, 1986) 982. G Wivenes, ‘Le droit européen et les constitutions nationales, Rapport luxembourgeois’, in Lord Slynn of Hadley and M Andenas (eds), FIDE XX. Congress vol 1 (London, BIICL, 2002) 267.

482

17 Hungary RENÁTA UITZ*

In recent years the Hungarian Government has become famous for its illiberal constitutional transformation questioning shared European constitutional values, its vigorous defence of the Schengen border (with a fence) and Europe’s Christian identity against the invasion of ‘illegal migrants’, as well as for its questionable handling of European Union (EU) funds.1 In May 2018 at his confirmation hearing before the parliamentary European Affairs Committee, Gergely Gulyás, Minister of the Prime Minister’s Office – the Cabinet member responsible for EU affairs – submitted without hesitation that EU Hungary’s membership served the national interest, as its benefits outweighed its drawbacks.2 After some adjustment, the new Hungarian Fundamental Law of 20113 has come to frame this deeply conflicting (and often outright hostile) stance perfectly. The Fundamental Law replaced Hungary’s post-communist Constitution.4 The text of the old Constitution served as an instrument of transition to democracy and Hungary’s EU accession. Hungary was often described as an enthusiastic Member State: it was the first to ratify the Lisbon Treaty, in an urgent vote in Parliament, even before its text was officially available to MPs.5 Such instances of enthusiasm were framed by a more complex array of motivations behind seeking EU accession from the early days of accession negotiations in all Central European countries.6

* Professor of law, Central European University, Vienna. Unless indicated otherwise, all translations from the Hungarian are mine. All websites were last visited on 10 October 2018, and the manuscript was completed immediately afterwards. 1 OLAF, The OLAF report 2017. Eighteenth report of the European Anti-Fraud Office. 1 January to 31 December 2017 (Luxembourg, Publications Office of the EU, 2018). 2 Meeting of the EAC, EUB-1/2018, 15 May 2018, minutes at www.parlament.hu/documents/static/biz41/bizjkv41/ EUB/1805151.pdf (in Hungarian). 3 Original title: Magyarország Alaptörvénye [Fundamental Law of Hungary, hereinafter FLHU]. The consolidated text of the Fundamental Law available in English translation at https://hunconcourt.hu/uploads/sites/3/2018/10/fundamental_law_of_hungary-7.pdf. The Fundamental Law entered into force on 1 January 2012. For a comprehensive treatment of a wide range of issues concerning the adoption of the Fundamental Law see Gábor Attila Tóth (ed), Constitution for a Disunited Nation: On Hungary’s 2011 Fundamental Law (Budapest, CEU Press, 2012). 4 Act no XX of 1949, A Magyar Köztársaság Alkotmánya (Constitution of the Republic of Hungary). Hereinafter: the old Constitution. Available in English translation at www-archiv.parlament.hu/angol/act_xx_of_1949.pdf. 5 Gabriella Ilonszki, ‘The Hungarian Parliament and EU Affairs: A Modest Actor Dominated by the Executive’, in C Hefftler et al (eds), The Palgrave Handbook of National Parliaments in the European Union (London, Palgrave Macmillan, 2015) 531. 6 See eg Heather Grabbe, The EU’s Transformative Power. Europeanization Through Conditionality in Central and Eastern Europe (London, Palgrave Macmillan, 2006), esp ch 3.

484  Renáta Uitz The Fundamental Law has been ambivalent about Hungary’s place in Europe from the start. On the one hand, it had several provisions enabling participation in the EU. In addition to a Europe clause7 the Fundamental Law also includes express provisions on the powers of the Government and the Parliament in EU decision-making,8 on the participation of EU citizens in local and EU parliamentary elections,9 and express articles which envisage cooperation in criminal justice matters.10 At the same time, the Fundamental Law’s preamble, the National Avowal, prefers to situate Hungary first and foremost in Europe’s millennial Christian tradition, and not in the EU. For the Fundamental Law Europe is – first and foremost –a historical artefact: the National Avowal proclaims that ‘We are proud that our people has over the centuries defended Europe in a series of struggles and enriched Europe’s common values with its talent and diligence.’ The Fundamental Law’s ambivalence did not seem to make much difference at first. When the Government asked the Constitutional Court about the constitutional requirements of ratifying the ESM Treaty, the Government presented the issue as undertaking a potential future international obligation stemming from EU membership, but going beyond the scope of the founding treaties.11 The Constitutional Court permitted the ratification of the ESM Treaty, provided that Parliament undertook this new international obligation with a two-thirds majority. It was accepted that the new obligations include limits on the fiscal powers of Parliament without crying sovereignty. As a clear departure, recent amendments to the Fundamental Law clearly emphasise the distinctiveness of the Hungarian constitutional order. The Seventh Amendment adopted in June 2018 makes defending the constitutional identity and Christian culture the duty of all state organs.12 It demarcates ‘Hungary’s inalienable right of disposition relating to its territorial integrity, population, political system and form of governance.’13 It also includes several provisions to fend off efforts by external forces (ie the EU) to settle foreigners (ie asylum-seekers) in Hungary.14 This is a far cry from the commitment underlying the Fifth Amendment, adopted in 2013, to put an end to debates with European constitutional actors about Hungary’s constitutional transformation.15 In practice, voluntary compliance with EU law goes hand in hand with open resistance to ‘Brussels’ in other instances.16 Since 2010 the Hungarian Government has gradually built solid constitutional foundations for its defiance of the Commission and other EU institutions in constitutional matters. The Constitutional Court assisted the government’s defence of

7 FLHU, Art E. 8 FLHU, Art 19. 9 FLHU, Art XXIII. 10 Eg FLHU, Art XXVIII(4) and (6). 11 Alkotmánybíróság (hereinafter AB) decision 22/2012 (V. 11.), para 51: ‘It is a necessary condition that Hungary, as the Member State of the European Union, should be a party to the treaty together with other Member States. It is a precondition that the treaty should cause the joint exercising of further competences, or exercising them through the institutions of the European Union … It is not a condition, however, that the treaty specifies itself as the European Union’s law, and neither is it a requirement that it should belong to the founding treaties of the EU.’ Unless mentioned otherwise, English translations and/or summaries of AB jurisdiction are available at https://­ hunconcourt.hu/translations-summaries-old/. 12 FLHU, new Art R(4). 13 FLHU, Art E(2), as amended in June 2018. 14 Esp FLHU, Art E(2), as amended, and Art XIV, as amended. 15 As stated in the official reasons accompanying bill T/11545 (June 2013). 16 On the problem of symbolic or creative compliance where Hungary’s performance is assessed in a comparative perspective, see Ágnes Bátory, ‘Defying the Commission: Creative Compliance and Respect for the Rule of Law in the EU’, (2016) 94 Public Administration 685.

Hungary  485 sovereignty – often presented in terms of defending national constitutional identity – against European invasion. This chapter traces these constitutional developments and argues that the Hungarian Government takes advantage of procedures supplied by European law for building and legitimising a constitutional regime that mocks the founding values of the EU. In doing so the chapter hopes to contribute to the study of the European constitutional space shaped by ongoing dialogue (and discord) on shared constitutional values and principles.17

I.  Main Characteristics of the National Constitutional System, Including the System of the Judiciary Hungary has a civil law legal system which is based on the supremacy of its written constitution, the Fundamental Law. Similarly to its predecessor, the Fundamental Law can be amended by Parliament through a qualified majority.18 Despite formal constitutional continuity between the old Constitution and the new Fundamental Law, the latter was adopted as a strategic step in a gradual, yet fundamental constitutional transformation seeking to turn Hungary into an illiberal democracy, as explained by Prime Minister Viktor Orbán himself.19 The Fundamental Law provides for a unicameral Parliament (often translated in official documents as the National Assembly)20 and a parliamentary system of government,21 with a Government responsible before Parliament.22 The Government is led by the Prime Minister, elected by Parliament by simple majority.23 The Government exercises executive powers.24 The President of the Republic is indirectly elected by Parliament.25 The President’s constitutional role is predominantly ceremonial; he does not participate in the formation and execution of the Government’s policies. The Hungarian judiciary has four levels of courts, with the Kúria (previously known as the Supreme Court26) the highest court of appeal in criminal, civil, administrative and other matters. The rest of the judicial hierarchy is comprised of 111 district courts, 20 regional courts and five regional courts of appeal.27 In addition to being the highest court of appeal and exercising 17 On this see Renáta Uitz and András Sajó, ‘The Sovereign Strikes Back. A Judicial Perspective on Multi-Layered Constitutionalism’, (2018) 20 European Journal of Law Reform 97. Also Wojciech Sadurski, Constitutionalism and the Enlargement of Europe (Oxford, OUP, 2012); Monica Claes and Maartje de Visser, ‘Reflections on Comparative Method in European Constitutional Law’, in M Adams and J Bomhoff (eds), Practice and Theory in Comparative Law (Cambridge, Cambridge University Press, 2012) 143, 157–61. 18 FLHU, Art S(2). A constitutional amendment cannot be adopted by a referendum (FLHU, Art (83)(a)). Rules on amending the old Constitution (Article 24(5)) were amended to enable the adoption of a new constitution by two-thirds majority in Parliament (instead of the four-fifths majority initially required). The Hungarian Government lost its constitution-making majority in Parliament in February 2015, for much of its second consecutive term. In April 2018 the Government won back its two-thirds parliamentary majority in regular national elections for a third consecutive term. 19 See Renáta Uitz, ‘Reinventing Hungary with Revolutionary Fervor: The Declaration of National Cooperation as a Readers’ Guide to the Fundamental Law of 2011’, in J Kovács and B Trencsényi (eds), Brave New Hungary: Mapping the ‘System of National Cooperation’ (Lanham MD, Lexington Books, 2020) 9. 9. 20 FLHU, Art 1. 21 FLHU, Art 1(2)(f): Parliament elects the Prime Minister and decides on matters of confidence related to the Government. 22 FLHU, Arts 15(1) and 21. 23 FLHU, Art 16(3)-(4). 24 FLHU, Art 15(1). 25 FLHU, Art 1(2)(e): the President of the republic is elected by Parliament. Art 10(1) prescribes a five-year presidential term. 26 FLHU, Art 25(1). 27 See http://birosag.hu/en/information/hungarian-judicial-system (in English).

486  Renáta Uitz review jurisdiction, the Kúria is responsible for ensuring the ‘uniformity of the application of the law by the courts and shall take uniformity decisions which shall be binding on the courts’.28 The Seventh Amendment of the Fundamental Law established a new administrative judiciary, complete with a Supreme Administrative Court (Közigazgatási Felsőbíróság) alongside the Kúria.29 The administration of courts is handled by the National Office for the Judiciary and its president.30 In addition to the four-level judiciary, the Fundamental Law retains a separate Constitutional Court (Alkotmánybíróság) as ‘the principal organ for the protection of the Fundamental Law’.31 The Constitutional Court has 15 justices, elected by two-thirds majority in Parliament for a 12-year, non-renewable term.32 Despite the constitutional reform curbing its powers, on paper the Constitutional Court continues to enjoy considerable powers to perform preliminary constitutional review, abstract constitutional review as well as accepts judicial referrals and individual constitutional complaints.33 Independent constitutional actors with a potential to serve as checks on political powers include the Commissioner of Fundamental Rights (also known as the ombudsman),34 the National Bank, which is a central bank responsible for monetary policy,35 the State Audit Office36 and a Budget Council ensuring the feasibility of the budget.37 The equality body (Equal Treatment Authority) which was established in compliance with EU anti-discrimination law was not granted constitutional status by the Fundamental Law.38 These constitutional checks are mostly symbolic, however, as long as the Government holds control over their appointment through its overwhelming majority in Parliament.

II.  Constitutional Culture Hungary’s recent constitutional transformation has kept European constitutional actors on high alert. In the summer of 2018 the Orbán Government’s self-proclaimed pursuit of illiberal democracy39 was upgraded to a commitment to build a Christian democracy that is ‘by definition, not liberal’.40 It is in this spirit that the exploitation of constitution-making for political gains and its

28 FLHU, Art 25(3). 29 FLHU, Art 25(3), as amended. The implementing legislation establishing the administrative judiciary was expected to be tabled before the end of 2018. 30 FLHU, Art 25(5). 31 FLHU, Art 24(1). 32 FLHU, Art 24(8). Constitutional justices serve until the end of their term and are no longer subject to mandatory retirement at the age of 70. 33 FLHU, Arts 24(2) and 24(5) for the key rules on the jurisdiction of the Constitutional Court. See also Act no CLI of 2011 for the Act on the Constitutional Court, as available in English translation at https://hunconcourt.hu/act-on-the-cc/. 34 FLHU, Art 30. 35 FLHU, Art 41. 36 FLHU, Art 43. 37 FLHU, Art 44. The Budget Council is comprised of the President of the Budget Council, the Governor of the National Bank of Hungary and the President of the State Audit Office; FLHU, Art 44(4). 38 See www.egyenlobanasmod.hu/eng. 39 Hungary’s Prime Minister a Champion for Illiberalism, The Washington Post, 6 August 2014, www.washingtonpost. com/opinions/harold-meyerson-hungarys-prime-minister-is-a-champion-of-illiberalism/2014/08/06/143a53ae-1d9d11e4-82f9-2cd6fa8da5c4_story.html. 40 Prime Minister Viktor Orbán’s speech at the 29th Bálványos Summer Open University and Student Camp, 28 July 2018, Tusnádfürdő (Băile Tuşnad), www.kormany.hu/en/the-prime-minister/the-prime-minister-s-speeches/ prime-minister-viktor-orban-s-speech-at-the-29th-balvanyos-summer-open-university-and-student-camp.

Hungary  487 defiance to European constitutional and human rights standards continues in Hungary in a welldocumented fashion. Although the Commission continues to bring infringement proceedings against Hungary all the way to the Court of Justice of the EU (CJEU) in matters of constitutional significance,41 the most important political development of the fall of 2018 was the European Parliament’s decision to activate Article 7(1) TEU against Hungary.42 The Government’s hostility towards constitutionalism, rule of law, and the European constitutional dialogue was certainly not shared by all constitutional actors in Hungary – at least initially. As soon as it had an opportunity to express an opinion on the idea of a new constitution, in September 2010, in a letter to the parliamentary committee of constitution drafting, the Constitutional Court emphasised that the new Hungarian Constitution has to be part of Europe’s shared constitutional tradition.43 The Court insisted in 2011 in this spirit that the unalterable core of the Hungarian constitution may be derived from ius cogens and general principles of international law expressed in such multilateral treaties as the ICCPR, the European Convention on Human Rights, and those fundamental rights and constitutional principles which are part of the shared European constitutional tradition and are reflected in the documents of the EU and the Council of Europe.44 Thus, the Constitutional Court believed that the constitution-making powers of Parliament were not unlimited. In 2013 the Constitutional Court promised that it would define limits on constitution-making and legislative powers in light of Articles Q and E of the Fundamental Law, with reference to the obligations stemming from Hungary’s EU membership.45 According to the Constitutional Court, these sources give effect to a coherent system of constitutional values which cannot be disregarded by the constitution-making or the legislative power.46 The Constitutional Court also said that national and European constitutional development by necessity influenced the interpretation of the Fundamental Law.47 Although the Constitutional Court showed great willingness to define the unalterable core of the Hungarian constitution with reference to international and European obligations, the Court was not expected to protect this core from invasions of domestic constitutional actors.48 This was due in part to the foreseeable changes in the composition of the Constitutional Court and in part to limitations on its jurisdiction imposed in in response to its early attempts to draw limits of Parliament’s constitution-making powers. One symbolic constitutional battle with practical consequences for the foundation and limits of EU membership concerns the applicability of the Constitutional Court’s old case law to the new Fundamental Law. The Constitutional Court confirmed the continuing applicability of its old case law under the Fundamental Law in 2012, in its judgment on the terms of ratification of

41 Infringement action no 20172076 on the amendment of the higher education act; infringement action no 20172110 on transparency measures affecting civil society organisations. 42 P8_TA-PROV(2018)0340 European Parliament resolution of 12 September 2018 on a proposal calling on the Council to determine, pursuant to Art 7(1) of the Treaty on European Union, the existence of a clear risk of a serious breach by Hungary of the values on which the Union is founded (2017/2131(INL)). 43 The letter signed by Chief Justice Péter Paczolay was published in Alkotmánybírósági Szemle 2010, No 2, 100 (in Hungarian). 44 AB decision 61/2011 (VII. 14.) on the constitutionality of certain constitutional amendments. In the case the Constitutional Court found that it did not have jurisdiction to declare constitutional amendments unconstitutional (in Hungarian). 45 AB decision 12/2013 (V. 24.). 46 Ibid, paras 46 and 48. 47 AB decision 36/2013 (XII. 15.) para 24 (in Hungarian). This reference does not point to EU law directly, only to European constitutional developments more generally. 48 Nóra Chronowski, ‘A magyar alkotmánybíráskodás és a közös európai alkotmányos standardok’ [‘Hungarian Constitutional Court Jurisdiction and Common European Constitutional Standards’], (2014) 20(2) JURA 26, 30.

488  Renáta Uitz the ESM Treaty.49 In response, in 2013 the Fourth Amendment introduced a special rule, invalidating old Constitutional Court jurisprudence.50 This has not prevented the Constitutional Court from following its old case law in cases where a provision of the Fundamental Law is essentially similar to the article of the old Constitution, if the two contexts are similar, and the rules of interpretation of the Fundamental Law permit so in the concrete case.51 Specifically, this means that the Constitutional Court carried over its jurisprudence on the Europe clause to the Fundamental Law. This meant that in the spirit of its old case law the Constitutional Court was – at least in principle – ready to draw limits on the Hungarian Parliament’s constitution-making powers with reference to international and European human rights standards and European constitutional principles will continue to operate under the new Fundamental Law. In December 2016, however, the Constitutional Court departed from its earlier jurisprudence to offer a full-throttled defence of constitutional identity against the EU. The Court announced that its task was to ascertain ‘whether the joint exercise of powers under Article E(2) of the Fundamental Law would violate human dignity, another fundamental right, the sovereignty of Hungary or its identity based on the country’s historical constitution’.52 This sharp change of course happened not even a month after the Government’s bill to insert the protection of constitutional identity in the Fundamental Law failed to receive a two-thirds majority (by two votes) in Parliament.53 Although at the time of the judgment the majority of the Constitutional Court had already slipped to justices appointed by the Fidesz-majority Parliament, the new appointees were not involved in the case. Fidesz’ coalition won a two-thirds majority in Parliament in 2018. In June 2018 it adopted the Seventh Amendment to the Fundamental Law, defending Hungary’s sovereignty and constitutional identity (amended Article E(2) and new Article R(4)). The provisions installed by the Seventh Amendment to prevent ‘external forces’ from settling foreigners (ie asylum-seekers) in Hungary (Article XIV, as amended) go back to October 2016 to a referendum proposed by the Government on the question ‘Do you want the European Union to be able to prescribe the mandatory settlement of non-Hungarian citizens in Hungary even without the consent of Parliament?’54 Although the referendum was invalid due to low turnout, the Government communicated it as a sweeping victory where over 3 million Hungarian voters rejected the will of Brussels. As soon as it had the requisite majority, the Government ensured that the Fundamental Law was amended accordingly. While the Constitutional Court’s adventures under the new Fundamental Law are most notable, two further developments are also visible: • The Commissioner of Fundamental Rights (ombudsman) routinely refers to European Court of Human Rights (ECtHR) and CJEU jurisprudence, and to EU law in applications before the Constitutional Court.

49 AB 22/2012 (V. 11.) (n 11) paras 42–46. 50 FLHU, Closing provisions, point 5: ‘The decisions of the Constitutional Court taken prior to the entry into force of the Fundamental Law are repealed. This provision shall be without prejudice to the legal effects produced by those decisions.’ 51 AB decision 13/2013 (VI. 17.) para 32. 52 AB decision 22/2016 (XII. 5.) (emphasis added). 53 On the failed amendment see Renáta Uitz, ‘National Constitutional Identity in the European Constitutional Project: A Recipe for Exposing Cover Ups and Masquerades’, 11 November 2016, at https://verfassungsblog.de/national-­ constitutional-identity-in-the-european-constitutional-project-a-recipe-for-exposing-cover-ups-and-masquerades/. 54 Matteo Garavoglia, ‘What Hungary’s Referendum Says About Europe’s Politics and Policymaking’, 3 October 2018, www.brookings.edu/blog/order-from-chaos/2016/10/03/what-hungarys-referendum-says-about-europes-politics-andpolicymaking/.

Hungary  489 • Ordinary courts have been rather active in making references for preliminary ruling before the CJEU.55 The openness of the judiciary to engage with EU law together with the Constitutional Court’s new-found enthusiasm for protecting Hungarian constitutional ­ ­identity – enshrined in Article R(4) of the Fundamental Law since 2018 – are likely to encourage Hungarian litigants to export such cases to European fora, such as the CJEU or the ECtHR via domestic litigation.

III.  Constitutional Foundations of EU Membership and Closely Related Instruments A.  The Old Constitution In 1998 in its decision on the Europe Agreement and its implementing rules the Constitutional Court held that Hungary’s accession to the EU required a constitutional amendment to this very effect. In particular, the Court ruled that ‘without an express authorisation by the Constitution, the Parliament may not, in an international treaty, constitutionally extend beyond the principle of territoriality in a field of law covered by the exclusive jurisdiction of the sovereign state’.56 Still, in the long preparation for EU accession, the constitutional dimension has been rather neglected: the project of the Europe clause appeared rather late and by that time it was clear that a minimalist approach would be adopted.57 In the end in 2002 the old Constitution received a Europe clause authorising accession to the Union (Article 2/A), another clause stating that the creation of European Unity was a constitutional objective (Article 6(4)) and a third clause settling the relationship between the government and the parliament in EU matters (Article 35/A, old Constitution). At the time of its adoption, the Europe clause served first and foremost as an accession clause. While scholarship has appreciated the significance of the Europe clause as the foundation of EU membership (integration) early on,58 the Constitutional Court took a long time to come to terms with the constitutional effect of EU membership. The Europe clause was added to supplement the old Constitution’s provision on international law (Article 7).59 In its first decisions concerning EU law the Constitutional Court made it clear that it was not going to treat the EU treaties as international law,60 thus excluding questions of

55 See Flóra Fazekas and Márton Varju, ‘The Reception of European Union Law in Hungary, The Constitutional Court and the Hungarian Judiciary’, (2011) 48 CML Rev, 1945. 56 AB decision 30/1998 (VI. 25.), Reasoning section V.2. 57 Nóra Chronowski, ‘Integrálódó’ alkotmányjog [‘Integrating’ constitutional law] (Budapest and Pécs, Dialóg Campus, 2005) 164, 173. 58 Pál Sonnevend, ‘Alapvető jogaink a csatlakozás után’ [‘Our Fundamental Rights After EU Accession’], (2003) 7(2) Fundamentum, 27, 33. 59 ‘The legal system of the Republic of Hungary accepts the generally recognized principles of international law, and shall harmonize the country’s domestic law with the obligations assumed under international law.’ 60 AB decision 1053/E/2005 (in Hungarian). The position was reaffirmed more recently in AB decision 8/2011 (II. 18.) (dismissal for civil servants). For a selection of Constitutional Court decision on EU matters in English see Márta Dezső, Constitutional Law in Hungary (Alphen aan den Rijn, Kluwer Law International, 2010) 57 f. For a more detailed analysis of key cases see András Bragyova, ‘No New(s), Good News?’, in G Tóth (ed), Constitution for a Disunited Nation: On Hungary’s 2011 Fundamental Law (Budapest, CEU Press, 2012) 335, 340–48.

490  Renáta Uitz compatibility of domestic law with the EU treaties from its jurisdiction.61 The Constitutional Court adopted this stance because Hungary retains a dualist position towards international law. This stance, however, is not tenable due to the primacy, direct effect, and direct applicability of EU law. Thus, much of the Constitutional Court’s ambivalence regarding EU law and EU issues originates from this tension between dualism and monism.62 At the same time the Hungarian Constitutional Court is not unique among its peers when it refuses to see the Europe clause itself as an authorisation to review the compatibility of national law with EU law.63 In the Hungarian literature the Constitutional Court’s lack of jurisdiction to review the compatibility of domestic legal norms in light of EU law is also justified by arguments stemming from EU law. After all, on the basis of Simmenthal,64 domestic law which violates EU law is not applicable and it is not for the Constitutional Court to rule on this inapplicability.65 It is understood that this state of affairs does not violate the constitutional principle of the rule of law.66 It is the task of ordinary courts to find inapplicable domestic legal rules which violate EU law. The Supreme Court has a prominent position to ensure that EU law is properly applied by ordinary courts through directing the uniform application of the law by all courts. The Constitutional Court will not find a violation of EU law by a national legal norm even in a case where ordinary courts have already found the challenged rule inapplicable. Instead, in such a case the Constitutional Court was seen to find such a rule to violate the Constitution only. It was in this spirit that in its early jurisprudence the Constitutional Court said explicitly that national authorities are not required to enact measures in order to implement EU law under the Constitution’s Europe clause.67 In a subsequent case the Constitutional Court emphasised that it will rule on the failure to transpose EU law when Parliament overlooks that constitutional (ie non-EU) obligation.68 While in a scholarly contribution Judge András Bragyova suggested that the Constitution’s rule of law clause (Article 2(1), old Constitution) may impose a requirement of Europarechtsfreundlichkeit on Hungarian constitutional actors,69 the Constitutional Court has never adopted this approach. This stance on EU law was in sharp contrast with how the Constitutional Court viewed the task of the government towards international obligations.70

61 AB decision 72/2006 (XII. 15.). The Constitutional Court has and had jurisdiction to find that domestic legal norms violate international obligations. See Act no XXXII of 1989 on the Constitutional Court, Art 1(c) for rules under the old Constitution, and FLHU Art 24(2)(f) on the rules currently in force. 62 See László Blutman, ‘A magyar Lisszabon-határozat: Befejezetlen szimfónia luxemburgi hangnemben’ [‘The Hungarian Lisbon Decision: Unfinished Symphony in Luxemburg Tone’] (2010) 1(2) Alkotmánybírósági Szemle, 90, 91 on the consequences of dualism for the reception of EU law in Hungary. A treaty to which the EU is a member qualifies as regular international law, and not as EU law, according to the Constitutional Court. That was the case with the Agreement between the European Union and the Republic of Iceland and the Kingdom of Norway on the surrender procedure between the Member States of the European Union and Iceland and Norway, in AB decision 32/2008 (III. 12.) (known as the ‘EUIN decision’). 63 See Darinka Piqani, ‘The Role of National Constitutional Courts in Issues of Compliance’, in M Cremona (ed), Compliance and Enforcement of EU Law (Oxford, OUP, 2012) 133, 137. 64 ECJ 09.03.1977 106/77 (Amministrazione delle Finanze dello Stato v Simmenthal SpA) ECLI:EU:C:1978:49. 65 Péter Paczolay, ‘A magyar alkotmány jövője és az uniós csatlakozás’ [‘The Future of the Hungarian Constitution and Accession to the Union’], 13 Politikatudományi Szemle (2004) 31, 38. 66 Art 2(1) of the old Constitution. László Trócsányi and Lóránt Csink, ‘Alkotmány v közösségi jog: Az Alkotmánybíróság helye az Európai Unióban’ [‘Constitution v Community Law: the Status of the Constitutional Court in the EU’], (2008) 63(2) Jogtudományi Közlöny, 63, 67. 67 AB 1053/E/2005 (online betting, also known as sportingbet) (n 60). 68 AB 72/2006 (XII. 15.) (n 61) Part III. 10. The Constitutional Court found a constitutional omission earlier in the case. 69 Bragyova, ‘No New(s)’ (n 60) 342. Judge Bragyova emphasised that the obligation is not stemming from the Europe clause (Art 2/A(1), old Constitution). 70 AB decision 16/1993 (III. 12.). The international treaty was the Paris Peace Treaty in the case.

Hungary  491 In short, the Hungarian Constitutional Court ‘tended, in practice to avoid the c­ onstitutionalisation of EU law’.71 The Court’s approach to EU issues has long been characterised by ambivalence edging on outright hostility. Most famously, on the eve of Hungary’s accession to the Union in a case on agricultural surplus stock the Constitutional Court invalidated the rules transposing the EU regulation solely on (domestic) constitutional grounds, arguing that the invalidation of the Hungarian transposing rules does not affect the validity of EU rules.72 This was taken as a sign that the Court was still learning how to engage in cooperative constitutionalism on the European level.73 When discussing cooperation in EU matters, it is worth noting that in addition to the Europe clause the old Constitution provided in its Article 6(4) that ‘The Republic of Hungary shall take an active part in establishing a European unity in order to achieve freedom, well-being and security for the peoples of Europe.’ This clause received very little attention in the early case law of the Constitutional Court on EU issues, although in an essay the future Chief Justice of the Constitutional Court Péter Paczolay indicated that the principle of loyal cooperation as enshrined in the EC Treaty (ex-Article 10, now Article 4[3] TEU) could be an important source of constitutional interpretation in EU issues.74 While Article 6(4) of the old Constitution could have been read as the source of a duty of loyal cooperation to the farthest extent,75 the Constitutional Court hesitated to take this path. The Constitutional Court discovered the potentials of EU law in 2010 at a time when the Fundamental Law was already on the drafting board.76 In the Lisbon Treaty decision the Court visibly attempted to create tighter bonds between the Hungarian Constitution and EU obligations.77 Relying on Article 6(4) of the old Constitution the Constitutional Court noted that participation in the EU is not an aim in itself, but was ‘meant to serve the expansion of human rights, well-being and security’.78 In defining the goals of EU membership the Constitutional Court curiously substituted the provision’s original reference to ‘freedom of peoples’ by ‘human rights’. This slight slip of the tongue may indicate that the Hungarian Constitutional Court came to associate the protection of human rights with EU membership, a somewhat more accessible concept for a court than the ‘freedom of peoples’ prescribed by the language of Article 6(4) of the old Constitution. Following this line of reasoning in subsequent judgments first under the old Constitution and then under the Fundamental Law the Constitutional Court insisted that the Fundamental Law and its constitutional regime form part of the European constitutional

71 Allan Tatham, Central European Constitutional Courts in the Face of EU Membership. The Influence of the German Model in Hungary and Poland (Leiden, Brill, 2013) 191. 72 AB decision 17/2004 (V. 25.). For English language analysis see Renáta Uitz, ‘EU Law and the Hungarian Constitutional Court: Lessons from the First Encounter’, in W Sadurski, J Ziller and K Żurek (eds), Après Enlargement: Legal and Political Responses in Central and Eastern Europe (2006) 41. For a similar judgment more recently see AB decision 142/2010 (VII. 14.), invalidating for lack of sufficient time to prepare provisions of the law transposing EU law on the SPS scheme. 73 András Sajó, ‘Learning Co-operative Constitutionalism the Hard Way: the Hungarian Constitutional Court Shying Away from EU Supremacy’, (2004) 2 Zeitschrift für Staats- und Europawissenschaften 351. 74 Paczolay, ‘A magyar alkotmány jövője’ (n 65) 38. 75 Márta Dezső and Attila Vincze, Magyar alkotmányosság az európai integrációban [Hungarian Constitutionalism in European Integration], 2nd edn, (Budapest, HVG-ORAC, 2012) 65. 76 See Parliamentary Resolution 47/2010 (VI. 29.) OGY, establishing the parliamenary committee on constitution-making. 77 AB decision 143/2010 (VII. 14.) (Lisbon Treaty). 78 Ibid, ABH (Alkotmánybíróság Határozatai, the official gazette of the Constitutoinal Court, available at www.­kozlonyok. hu/kozlonyok/valaszt.htm) 2010, 698, 708 (in Hungarian; emphasis added).

492  Renáta Uitz tradition, and that the unalterable core of the constitution builds on EU obligations and EU law, among other sources.79 The Constitutional Court, however, was not unanimous in the Lisbon Treaty decision in its newly found allegiance to EU law. Judge Béla Pokol in a strongly worded dissenting opinion to the decision on student contracts warned of the danger of Hungarian society’s melting away in the EU.80 This was an indication that the EU-friendly attitude of the Constitutional Court may not survive the impending change of the composition of the Court. The Constitutional Court’s unexpected change of heart about EU law may have been ­influenced by the newly started constitutional transformation. Just a week before the Lisbon Treaty decision was handed down the Hungarian parliament amended the old Constitution and changed the procedure for electing Constitutional Court justices.81 The amendment enabled the parliament to fill the two empty seats on the Constitutional Court without having to compromise with the opposition in the nomination phase, and thus enabling the instant installation of two new justices.

B.  The Fundamental Law Similar to its predecessor, the new Fundamental Law has contained a Europe clause (Article E) and a separate clause on international law and international obligations (Article Q) from the start. By and large, these provisions were meant to have similar roles as their counterparts had in the old Constitution.82 Article E initially read as follows: 1) Hungary shall take an active part in establishing a European unity in the pursuit of freedom, well-being and security for the peoples of Europe. 2) In its role as a Member State of the European Union and by virtue of international treaty, Hungary may – to the extent necessary for exercising its rights and fulfilling its obligations stemming from the Founding Treaties – exercise certain competences deriving from the Fundamental Law, together with the other Member States, through the institutions of the European Union. 3) General binding rules of conduct may be laid down in European Union legislation within the framework set out in Paragraph (2). 4) The votes of two-thirds of all Members of Parliament shall be required to authorise the recognition of an international treaty referred to in Paragraph (2) as binding in scope.

While the making of the Fundamental Law provided the perfect opportunity to rethink and refine the Europe clause, to turn it into a proper integration clause (to use the terminology of former Constitutional Court Judge Imre Vörös83), and to ensure that the freshly adopted Europe

79 AB 61/2011 (VII. 14.) (n 44); AB 12/2013 (V. 24.) (n 45). 80 AB decision 32/2012 (VII. 4.) on student contracts in higher education, para 86 (Judge Béla Pokol, dissenting) (in Hungarian). 81 Constitutional amendment of 6 July 2010. 82 Nóra Chronowski and Erzsébet Csatlós, ‘Judicial Dialogue or National Monologue? The International Law and Hungarian Courts’, (2013) 1 ELTE Law Journal 7, 8. 83 Imre Vörös, Csoportkép Laokoónnal. A magyar jog és az alkotmánybíráskodás vívódása az európai joggal [Tableau with Laokoon, The Struggle of the Hungarian Legal System and Constitutional Justice with European law] (Budapest, HVG-ORAC, 2012) 104 f.

Hungary  493 clause reflects the altered constitutional reality of the EU itself,84 such a transformation did not take place. For its part, the Government was of the opinion that the two Europe clauses are ­essentially similar.85 In June 2018 the Seventh Amendment revised Article E(2) to include the following limitation: The exercise of powers under this Paragraph must be consistent with the fundamental rights and freedoms set out in the Fundamental Law, and it must not be allowed to restrict Hungary’s ­inalienable right of disposition relating to its territorial integrity, population, political system and form of governance.

C.  The Constitutional Court’s Case Law In its early judgments under the Fundamental Law the Constitutional Court followed a more EU-friendly approach to constitutional interpretation.86 Importantly, it found that the Europe clause in its initial iteration permitted ratification of the ESM Treaty without any substantive limits on sovereignty transfer. When applying Article E, the Constitutional Court followed the premises and principles of its old case law developed under Article 2/A of the previous Constitution.87 The only condition set by the Constitutional Court was that the treaty is ratified by a two-thirds majority in Parliament (hardly a limitation, in the circumstances). As expected, in November 2016 the majority of the Constitutional Court was comprised of appointees of the Fidesz Government. In December 2016, the Constitutional Court announced that it was ready to protect Hungary’s sovereignty and constitutional identity under the Europe clause.88 In doing so the Constitutional Court claimed to follow its European peers.89 It was in this spirit that in July 2018 the Constitutional Court blocked Hungary’s ratification of the Agreement on the European Patent Court as an unconstitutional limitation of judicial power (and ultimately, sovereignty) under the Fundamental Law.90 This is a clear departure from the Court’s attempt in 2010 to build an unamendable core to the Hungarian Constitution comprising, inter alia, EU law. At this point one may start wondering whether the Constitutional Court in its current composition would have let the ESM Treaty pass as it did in 2012. Meanwhile, a question which remains open is the status of EU law in the Hungarian constitutional regime. As noted, the Constitutional Court insists that founding EU treaties as well as secondary EU law do not qualify as regular international law for the purposes of constitutional analysis. The Constitutional Court has stated that EU primary as well as secondary law are ‘part

84 Flóra Fazekas, ‘Az uniós tagság alkotmányos alapjai az Alaptörvény előtt és után’ [‘The Constitutional Foundations of EU Membership Before and After the Fundamental Law’], (2015) 1 ProFuturo, 29, 40. The Europe clause could have also been amended to apply to transfer of sovereignty in other international organisations, and not only in the EU (ibid, 30). 85 See AB 22/2012 (V. 11.) (n 11) paras 3 and 36 on the Government’s position. 86 Eg AB 32/2012 (VII. 4.) (n 80), where the Constituional Court used EU Law to make up for the lack of a general right to higher education in the Fundamental Law (Art XI[4]). 87 AB 22/2012 (V. 11.) (n 11) para 46. 88 AB 22/2016 (XII. 5.) (n 52). 89 Ibid, paras 32-46. 90 AB decision 9/2018 (VII. 9.), esp para 53. See further in Katalin Kelemen, ‘The Agreement on a Unified Patent Court Cannot Be Ratified by Hungary, the Constitutional Court Says’, 1 August 2018, www.diritticomparati.it/ agreement-unified-patent-court-cannot-ratified-hungary-constitutional-court-says/.

494  Renáta Uitz of ’ Hungary’s ‘internal law’.91 The practical implications of this statement, nonetheless, are far from clear. Article E(3) of the Fundamental Law which declares that ‘the law of the European Union may stipulate generally binding rules of conduct subject to the conditions set out in paragraph (2)’ is of little help. In the past, the internal division of the Constitutional Court on this issue may have been the reason for the insistence of the Court to review the constitutionality of transposing or implementing Hungarian legal rules and not the original EU norm whenever possible. This may well change now, with a clear majority preferring to defend Hungarian constitutional identity from European invasion. The Constitutional Court made it clear that the Europe clause does not confer fundamental rights. Therefore it cannot be invoked via an individual constitutional complaint.92

IV.  Constitutional Limits to EU Integration In its foundational 1998 decision on the Europe Agreement the Constitutional Court indicated that there were limits to the transfer of sovereignty. In particular it said that state authority exercised by the Parliament embodying sovereignty through representatives is not an unlimited power; the Parliament may only act in the framework of the Constitution and the provisions of the Constitution limit its powers. This rule certainly applies, as appropriate, to other organs empowered to exercise state sovereignty as well. … The democratic legitimacy of exercising public authority is a constitutional requirement for both internal and external acts of sovereign power aimed at the determination of international relations or resulting in international obligations.93

In this early decision the Constitutional Court was concerned with the manner of transfer of sovereignty within the EU context. The Europe clause of the old Constitution, as carried over initially to the Fundamental Law did not prescribe any substantive limits to the transfer of sovereignty. Until December 2016 the Constitutional Court showed little interest94 in developing substantive limitations to the transfer of sovereignty.95 The Court’s newly found willingness to protect constitutional identity was confirmed in June 2018 by the Seventh Amendment to this effect. Until its change of heart in December 2016, the Court focused on procedural and formal requirements, seeking to ensure that the transfer of sovereignty rests on democratic legitimacy in the form of voting in Parliament.96 This approach was in line with the old Constitution that required a referendum for accession to the Union, while leaving the subsequent transfer of competences to a two-thirds (ie constitution-making) majority vote in Parliament.97 In its 2012 decision on the ESM Treaty under the Fundamental Law the Constitutional Court

91 See AB 72/2006 (XII. 15.) (n 61) Reasoning section III.11., confirmed in AB 143/2010 (VII. 14.) (n 77), ABH 2010, 698, 702 (in Hungarian). 92 AB decision 3151/2016 (VII. 22.) para 18 (in Hungarian). 93 AB 30/1998 (VI. 25.) (n 56), Reasoning section V.3. 94 For an exception, see AB 32/2008 (III. 12.) (n 62) (preliminary review of the Iceland-Norway agreement extending the reach of the European Arrest Warrant). Ultimately, the Constitutional Court noted that the unconstitutionality of the act could be remedied by the earlier than planned entry into force of the constitutional amendment which had already been passed. See part XI, 2, pp 14–15. 95 Blutman, ‘Lisszabon-határozat’ (n 62) 97. Also Flóra Fazekas, ‘EU law and the Hungarian Constitutional Court’, in M Varjú and E Várnay (eds), The Law of the European Union in Hungary: Institutions, Processes and the Law (Budapest, HVG-ORAC, 2014) 32, esp 60 f. 96 See AB 143/2010 (VII. 14.) (n 77), part IV.2.3.2. 97 Chronowski, Integrálódó (n 57) 193 f.

Hungary  495 read the Europe clause of the new Fundamental Law to require that Parliament takes a decision with a two-thirds majority (supermajority) on any transfer of sovereignty related to EU membership, and not only for the adoption or amendment of the founding EU treaties.98 The super-majority requirement in Article E(4) ‘is not required for the consent to be bound by an international treaty, if the treaty would not result in exercising, jointly with the institutions of the European Union or with other member states, new competences originating from the Fundamental Law’.99 The Court emphasised that first and foremost, it is the Government’s (and not the Court’s) constitutional duty to determine whether a treaty requires approval by supermajority and that this duty is fulfilled when the government prepares the bill to be adopted by Parliament.100 The Constitutional Court emphasised in the Lisbon Treaty decision that a complete transfer of sovereignty would be unconstitutional,101 as the Europe clause of the old Constitution (similarly to the Europe clause of the new Fundamental law) permitted the transfer of sovereignty ‘to the extent necessary to exercise the rights and fulfil the obligations set out in the founding treaties, exercise some of its competences deriving from the Fundamental Law jointly with other Member States, through the institutions of the European Union’ (Article E(2), emphasis added). The Constitutional Court reserved itself the power to assess the extent to which the transfer of sovereignty is necessary.102 In this respect the Court noted that it prefers to determine the necessity of the transfer of sovereignty through a preliminary review of the international agreement.103 In the Lisbon Treaty decision then Judge László Trócsányi’s concurring opinion offered a discussion on further, substantive limitations to the transfer of sovereignty. His premise is that the EU Member States cannot give up the essence of their sovereignty and statehood, their constitutional identity.104 According to him, the transfer of sovereignty cannot result in the alteration of key constitutional institutions (without further constitutional amendments to this effect) such as the manner of election of parliamentary deputies and the composition and powers of Parliament, Government, and the judiciary. This approach may not have been accepted by the majority at the time because the Constitutional Court had not been ready to develop a substantive standard for the constitutionally permissible limitation of sovereignty.105 Judge Trócsányi left the Constitutional Court in October 2010 to serve as ambassador to France, and from April 2014 as Minister of Justice, supporting the defence of national interests and national constitutional identity in the EU through using the legal tools made available by EU law.106 The scholarly literature has engaged with identifying formal as well as substantive limits to the transfer of sovereignty from the early days and then partly in response to the decisions of the

98 AB 22/2012 (V. 11.) (n 11). 99 Ibid, ruling, point 2. 100 Ibid, para 58: ‘the draft Act of Parliament on the promulgation of the international treaty shall be presented by the Government. Therefore it is the primary obligation of the Government, as the presenter of the draft, and of the Parliament – as the legislator of the Act promulgating the international treaty – to form an opinion about whether an international treaty falls under the scope of Article E) paras (2) and (4) of the Fundamental Law.’ 101 AB 143/2010 (VII. 14.) (n 77), ABH 2010, 698, 710 (my translation). 102 AB 143/2010 (VII. 14.) (n 77), ABH 2010, 698, 705. 103 Ibid. 104 Ibid, ABH 2010, 698, 713 (J. Trócsányi, concurrring). Trócsányi developed this argument into a theory to protect national constitutional identity as a basis of participation in the EU, in László Trócsányi, Az alkotmányozás dilemmái. Alkotmányos identitás és európai integráció [Dilemmas of Constitution-Making. Constitutional Identity and European Integration] (Budapest, HVG-ORAC, 2014) 79 f. 105 Blutman, ‘Lisszabon-határozat’ (n 62) 97. 106 EUB-1/2018 (n 2).

496  Renáta Uitz Constitutional Court.107 There is widespread agreement in the literature that Hungary cannot transfer powers to the EU which it does not have (nemo plus iuris).108 Some scholars argued in a compelling manner that the essential content of fundamental constitutional rights is a limit to transfer of sovereignty under the old Hungarian Constitution.109 Since December 2016 the Constitutional Court appears more eager to invoke constitutional identity defined by the historic constitution in order to limit transfer of sovereignty in the EU context. This is a clear change from the earlier position of the Constitutional Court on potential limitations of sovereignty transfer. Since June 2018 the Seventh Amendment has included a clear textual basis on the limits of transfer of sovereignty (Article E(2), as amended), fortified by a duty to defend national constitutional identity (Article R(4)). The Fundamental Law sets further, indirect obstacles to the full enjoyment of the benefit of EU law. Article 37(4) of the Fundamental Law limits the jurisdiction of the Constitutional Court to review the constitutionality of tax rules ‘solely as pertaining to inherent rights to life and human dignity, the right to the protection of personal data, the right to freedom of thought, freedom of conscience and freedom of religion, or the rights in connection with Hungarian citizenship, and may annul such laws only in the case of any infringement of these rights’. This means that excessive taxes that violate property rights cannot be challenged before the Constitutional Court.110 Applicants with a property claim against excessive taxes can (and do) turn to the CJEU via preliminary reference.

V.  Constitutional Rules and Practice on Implementing EU Law A.  Applicable Provisions of the Fundamental Law In June 2018 the Seventh Amendment introduced an express constitutional obligation on all constitutional actors to defend Hungary’s constitutional identity (Article R(4)). The practical impact of this provision cannot be assessed yet. What follows is an overview of other constitutional provisions, the operation of which may be affected by the new Article R(4). Article 19 of the Fundamental Law regulates the powers and responsibilities of government and parliament in EU affairs as follows: Parliament may request information from the Government on its position to be represented in the decision-making procedures of those institutions of the European Union that require government participation, and may take a stand on the draft placed on the agenda. In the course of the decisionmaking process within the European Union, the Government shall represent the position taken by the Parliament.

107 See Fruzsina Gárdos-Orosz, ‘Újabb áttekintés az uniós jog alkalmazásának magyarországi alkotmányos kereteiről’ [‘Recent Overview on the Constitutional Framework of the Application of Union Law in Hungary’], (2013) 6(4) Közjogi Szemle, 44, 49 ff. 108 Sonnenvend, ‘Alapvető jogaink’ (n 58) 33; Dezső/Vincze, Magyar alkotmányosság (n 75) 66; Tamás Kende and Tamás Szűcs (eds), Európai közjog es politika [European public law and policy] (Budapest, Wolters Kluwer, 2006), 772 f. 109 Sonnenvend, ‘Alapvető jogaink’ (n 58) 33 f. The article was inspired to a large extent by the fact that the EU Charter offered weaker protection to a number of fundamental constitutional rights than the old Consitution did (28 f). Dezső and Vincze import this reading for the application of the new Fundamental Law, Dezső/Vincze, Magyar alkotmányosság (n 75) 74. 110 AB decision 26/2013 (X. 4.) (in Hungarian), confirmed in AB ruling 3090/2017 (IV. 28.) paras 35 f (in Hungarian).

Hungary  497 Most details on the tasks of the government and parliament are contained in the act on the National Assembly111 and in the parliamentary Rules of Procedure (standing orders).112 These rules were adopted after the adoption of the Lisbon Treaty.113 In addition to EU-specific scrutiny mechanisms, Parliament may also use its regular means of scrutiny and political responsibility in EU matters. Within the Government the coordination of EU matters is entrusted with the Minister heading the Prime Minister’s Office.114 The portfolio includes all EU matters, with the exception of legal harmonisation.115 In EU decision-making processes the Hungarian position is prepared by the Government’s Inter-ministerial Committee for European Coordination (Európai Koordinációs Tárcaközi Bizottság – EKTB). After the elections of 2018 the Minister of the Prime Minister’s Office, Gergely Gulyás announced at his parliamentary hearing in the European Affairs Committee (EAC) the plan to have monthly cabinet meetings on EU affairs.116 Parliamentary scrutiny over EU affairs is concentrated in the European Affairs Committee, a permanent standing parliamentary committee117 that speaks on behalf of the whole of Parliament on EU matters. Parliament and its EAC are perceived to have an advisory role and as such, not a force to be reckoned with. It is telling that the Hungarian name of the parliamentary scrutiny procedure is not ‘control’ (ellenőrzés), but ‘coordination’ (egyeztetés). Reportedly, in 2012 the Government rejected an initiative sponsored by the Committee by claiming that the Parliament cannot instruct the Government.118 All in all, EU affairs are driven by the Government, while the Parliament is entitled to receive information on the Government’s position.119 As a rule, the Hungarian position is prepared by the Government. The Parliament’s opinion on the Government’s position is assessed by the EAC, and is not a matter for the plenary.120 Before attending meetings of the European Council, the Prime Minister is required by law to brief the Parliament’s Consultative Body on European Union Affairs.121 The Consultative Body is comprised of the Speaker, the leaders of parliamentary party factions, the chairman and vice-chairman of the Committee on European Affairs, the chairman of the Committee on Constitutional Affairs and the chairman of the Committee on Foreign Affairs, and others invited by the Speaker.122 The in camera briefing is followed by a debate in

111 Act no XXXVI of 2012 on the National Assembly. Available in English translation at http://parlament.hu/ documents/125505/138409/Act+XXXVI+of+2012+on+the+National+Assembly/b53726b7-12a8-4d93-acef-140feef44395. 112 Parliamentary Resolution 10/2014 (II. 24.) OGY on certain provisions of the Rules of Procedure, available in English at www.parlament.hu/documents/125505/138409/Resolution+on+certain+provisions+of+the+Rules+of+Procedure/96 8f2e08-f740-4241-a87b-28e6dc390407?version=1.2&inheritRedirect=true. 113 For a comprehensive English language overview see Ilonszki, ‘Hungarian Parliament’ (n 5). 114 94/2018 (V. 22.) Korm. decree on the responsibilities and jurisdiction of members of the Government, Art 14(1) (4) and Art 18. For an English language analysis of the old system (also covering the pre-accession origins of relevant mechanisms) see Ágnes Bátory, ‘The National Coordination of EU Policy in Hungary: Patterns of Continuity and Change’, (2012) 90 Public Administration 922. 115 Decree 94/2018 (n 114) Art 18(2)(d). 116 EUB-1/2018 (n 2) 36. 117 The minutes of the meetings of the Committee are available at www.parlament.hu/web/europai-ugyek-bizottsaga/abizottsag-ulesei (in Hungarian). 118 Here I follow Angéla Juhász-Tóth, ‘European Union Law and the Hungarian Parliament: Wasted Opportunities?’, in M Varjú and E Várnay (eds), The Law of the European Union in Hungary: Institutions, Processes and the Law (Budapest, HVG-ORAC, 2014) 77, 99. 119 Act no XXXVI of 2012, Art 64. 120 Resolution 10/2014 (II. 24.) OGY (fn 112), Art 140. 121 Act no XXXVI of 2012 on the National Assembly, Art 69(2). See Parliament’s dedicated English language website on these European Consultation Meetings at www.parlament.hu/web/house-of-the-national-assembly/ consultative-body-on-eu-affairs-and-reports-of-the-prime-minister. 122 Act no XXXVI of 2012, Art 69(3). On the in camera deliberation: Resolution 10/2014 (II. 24.) OGY (n 112), Art 139(2).

498  Renáta Uitz the committee. The Prime Minister is usually accompanied by a representative of the Ministry of Foreign Affairs at this briefing.123 Parliament may also formulate its own position on EU matters in the course of the scrutiny procedure.124 Such a procedure may be started by the chairperson of the EAC or a two-fifths minority (up to four times per ordinary sessions).125 If this position is formulated at the committee level, the committee meeting shall take place in camera.126 Note that the EAC’s role is advisory: the Government is not bound by its position.127 The Parliament’s position has a somewhat more significant weight if it concerns a matter that requires a supermajority decision in Parliament. In this case the Government may only depart from the Parliament’s position in a ‘justified case’.128 Detailed rules for the ex-ante and ex-post control of subsidiarity are included in both the act on the National Assembly (ie Parliament)129 and in the parliamentary Rules of Procedure.130 The primary responsibility for the control of subsidiarity in parliament rests with the EAC. If in an ex-ante procedure the Committee finds a violation, it shall transmit its opinion to the plenary, which is expected to decide on it within 15 days.131 In the procedure for the ex-post control of subsidiarity the Committee may ask the government to file a case with the CJEU.132 As even this brief summary suggests, parliamentary scrutiny is not particularly robust or transparent in Hungary.133 In addition, Parliament appears to display moderate interest in EU affairs in general. As of June 2018, according to the Parliament’s own website, the last time the Government offered its annual report on EU affairs to Parliament (required by law134) was for 2016.135

B.  Parliamentary Scrutiny and Early Warning Mechanism in Practice Initially, the Hungarian Parliament showed little interest in engaging with the European legislative process. It devoted time to issues that were of immediate national concern (such as the regulation of tobacco products136 or the European Public Prosecutor137). The Hungarian Parliament has

123 Juhász-Tóth, ‘EU Law and Hungarian Parliament’ (n 118) 97. 124 Act no XXXVI of 2012, Art 65. Parliament’s website keeps track of scrutiny procedures on a dedicated page. The English version is at www.parlament.hu/web/house-of-the-national-assembly/ongoing-scrutiny-procedures. 125 Resolution 10/2014 (II. 24.) OGY (fn 112), Art 140(3). The regular sessions are between 1 February to 15 June and from 1 September to 15 December each year. 126 Act no XXXVI of 2012, Art 65(4). 127 Ibid, Art 65(5). 128 Ibid, Art 65(6). 129 Ibid, Art 71. 130 Resolution 10/2014 (II. 24.) OGY (n 112), Arts 142 f. 131 Ibid, Art 142(3). 132 Ibid, Art 143(1). 133 For a European comparison see Katrin Auel, ‘The Europeanization of National Parliaments’, in J Magone (ed), The Routledge Handbook of European Politics (Abingdon-on-Thames, Routledge, 2014) 366, 370 f. 134 Act no. XXXVI of 2012, Art 69(5). 135 www.parlament.hu/web/house-of-the-national-assembly/report-of-the-government. 136 Proposal for a Directive of the European Parliament and of the Council on the approximation of the laws, regulations and administrative provisions of the Member States concerning the manufacture, presentation and sale of tobacco and related products, COM/2012/0788. The opinion of the National Assembly is available at http://ec.europa.eu/dgs/ secretariat_general/relations/relations_other/npo/hungary/2012_en.htm. 137 Proposal for a Council Regulation on the establishment of the European Public Prosecutor’s Office, COM/2013/0534. The opinion of the National Assembly is available at http://ec.europa.eu/dgs/secretariat_general/relations/relations_other/ npo/hungary/2013_en.htm.

Hungary  499 been criticised for being rather reactive: issues get discussed and concerns get raised when the Council has already prepared a consolidated draft, ie too late for a Hungarian objection to make its way to the European level.138 Over time the Hungarian Parliament became somewhat more active in the EU law-making process.139 The EAC started formally reflecting on the Commission’s legislative programme in 2014.140 By June 2018, the Parliament had raised subsidiarity concerns (in a parliamentary resolution) six times.141 The Hungarian Parliament, even in its least active phase, participated in all three yellow card attempts through the Early Warning Mechanism (EWM).142 The apparent appeal of the EWM is the opportunity that the subsidiarity check offers to defend national interests in the vocabulary of sovereignty. While Hungarian resistance to a common European asylum regime is the loudest contribution, the opportunities offered by the EWM to defend national sovereignty were clearly demonstrated in the Hungarian opposition to the European Public Prosecutors Office (EPPO).143 In the heat of the EPPO controversy, after the famous last-minute attempt of Commissioner Vera Jurova to sweeten the EPPO deal in the Hungarian EAC on 5 December 2016,144 the chairperson of the EAC informed the plenary of Parliament that EPPO endangered Hungary’s national identity and constitutional architecture as it blurred the line between the powers of national prosecutors and the planned European office.145 In particular, Hungary was concerned about transferring control over the investigation of VAT fraud to Europe, as VAT is a key source of the national – and not the EU – budget (thus, the protection of national interests is warranted). This position was fully supported by the Ministry of Justice.146 Despite the grand narrative of constitutional identity, Hungary’s special interest in fighting against the EPPO are clear: Hungary is on the top of investigations for mismanaging EU funds,147 to the point where a street lighting project awarded to Prime Minister Orban’s son-in-law made the narrative report for 2017.148 Predictably, the next battlefield for defending national constitutional identity is the Commission’s proposal to extend the powers of OLAF for Member States not joining the EPPO.149 The Hungarian Government – alongside Poland and Romania – can also be expected to oppose the Commission’s reform proposal to tie EU funding to rule of law performance on

138 Juhász-Tóth, ‘EU Law and Hungarian Parliament’ (n 118) 94. 139 A complete list of scrutiny processes concluded in 2014–18 is available on the Parliament’s website at www.parlament. hu/lezarult-egyeztetesi-eljarasok-2014-2018 (in Hungarian). 140 Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions. Commission Work Programme 2015. A New Start COM/2014/0910 final. See also http://ec.europa.eu/dgs/secretariat_general/relations/relations_other/npo/docs/hungary/2014/com20140910/ com20140910_assembly_opinion2_en.pdf. 141 See a list at www.parlament.hu/szubszidiaritas-vizsgalat (in Hungarian). 142 These are the so-called Monti II proposal (on tobacco products) (COM/2012/0788, n 136), the European Public Prosecutor’s Office (EPPO) (COM/2013/0534, n 137) and the Proposal for a Directive of the European Parliament and of the Council amending Directive 96/71/EC of The European Parliament and of the Council of 16 December 1996 concerning the posting of workers in the framework of the provision of services, COM/2016/0128. 143 www.ipex.eu/IPEXL-WEB/dossier/document/COM20130534.do#dossier-APP20130255. Also Parliamentary Resolution 87/2013 (X. 22.) OGY on asserting the principle of subsidiarity in relation to the draft regulation on establishing the European Public Prosecutor’s Office. 144 www.parlament.hu/documents/static/biz40/bizjkv40/EUB/1612051.pdf. 145 6 December 2016, plenary speech no 20. 146 6 December 2016, plenary speech no 21. 147 2017 OLAF Report (n 1). 148 2017 OLAF Report (n 1) 15. More broadly see eg www.theguardian.com/world/2018/feb/12/how-hungarianpms-supporters-profit-from-eu-backed-projects. 149 http://europa.eu/rapid/press-release_IP-18-3862_en.htm.

500  Renáta Uitz the national level for the next budget cycle (2021–27).150 Although the unanimity requirement in the Council in principle empowers Hungary to veto the EU budget, in practice that move would also prevent the Government from receiving much-needed EU funds – even if with ties attached.

C.  Legal Harmonisation In Hungary, legal harmonisation (including transposition of EU law) is the duty of the Government, not the Parliament. It is seen as a largely technical and bureaucratic affair, the origins of which date back to the times of preparation for EU accession.151 From the earliest days the legal harmonisation portfolio has remained with the Ministry of Justice,152 together with the representation of Hungary in infringement procedures before the Commission,153 and subsequently, before the CJEU.154 The Ministry of Justice bears overall responsibility to ensure that all government bills and government decrees comply with EU law.155 Official reasons accompanying bills must include a statement of conformity with EU obligations.156 In addition to ensuring the overall compatibility of all newly introduced legal norms with EU law, the Ministry of Justice is responsible for legal harmonisation, including the transposition of EU directives. The so-called programming of legal harmonisation is governed by a dedicated government decree.157 Building on the experiences of accession, Hungary follows a centralised model of legal harmonisation, with individualised programming of tasks for each EU norm. The Ministry of Justice has overall responsibility for coordination, while line ministries (typically the ministry which negotiated the act at the EU level) are responsible for preparing a timetable for approximation and further substantive measures. The programming of transposition also accounts for developments in CJEU jurisprudence, and there are several reported instances where preventive amendments were passed following CJEU judgments against legal solutions in other Member States.158 In the case of EU regulations, approximation mostly focuses on deregulation of incompatible national rules. In the case of EU directives, national rules are prepared depending on the constitutional rules requiring statutory or executive level regulation on a certain issue. Hungary follows the method of organic transposition, where provisions of directives are integrated into several relevant domestic legal norms, to enable the effective integration of EU rules.159 To the extent possible and practicable, transposition localises EU terminology, to the point where the

150 Eg Proposal for a Regulation of the European Parliament and of the Council on the protection of the Union’s budget in case of generalised deficiencies as regards the rule of law in the Member States, 02.05.2018, COM/2018/324 final, 2018/0136 (COD). 151 Here I follow Réka Somssich, ‘The Process and Methodology of Coordinating the Transposition of EU Law in Hungary’, in M Varju and E Várnay (eds), The Law of the European Union in Hungary (Budapest, HVG-Orac, 2014) 112. 152 Decree 94/2018 (n 114) Art 112(9)(b). 153 Ibid, Art 112(9)(bf). 154 Ibid, Art 112(9)(c). 155 Act no CXXX of 2010 on lawmaking, Art 16(2) on government bills and government decrees, and Art 16(3) on decrees of line ministries. 156 Ibid, Art 18(2). 157 302/2010 (XII. 23.) Korm. decree on the performance of tasks related to the preparation of legal rules in compliance with European Union law, as required by Act no CXXX of 2010, Art 30(1)(b), available at http://jogharmonizacio.gov.hu/ (in Hungarian). 158 See eg Somssich, ‘Process and Methodology’ (n 151) 144. 159 Ibid, 123 f.

Hungary  501 foreign origin of EU rules is indistinguishable.160 To indicate the EU origin of a legal norm, Hungarian law requires the inclusion of an approximation clause,161 through the inclusion of a standard formula on approximation.162 The Ministry of Justice maintains an online approximation database.163 The correlation tables prepared during programming for the preparation of bills and other legal rules are, however, not part of this database, nor are they submitted with the bills.164 The constitutional significance of legal harmonisation/assessment of compatibility with EU law is far from clear. In the case on student contracts in higher education the Constitutional Court took brief note of the applicable EU rules and noted that the challenged government decree lacks an approximation clause corresponding to the one in the act on higher education.165 Although the lack of this formal element in the Cabinet decree was noted by the Constitutional Court, the Court did not elaborate on it further and it was not a major reason for the ultimate unconstitutionality of the challenged rules. Since the government decree in question did not transpose EU law, the inclusion of an approximation clause was not required contrary to the opinion of the Constitutional Court.166 It would still have been incumbent on the Ministry of Justice to indicate the decree’s collision with EU law as part of its general obligation to screen all legal measures originating from the government for their compatibility with EU law. As the current Minister of Justice, László Trócsányi, explained in his investiture hearing at the EAC in May 2018, the number of infringement actions brought by the Commission against Hungary is not outstanding in European comparison.167 The Government clearly distinguished between what they regard as ‘politically charged processes’ and ordinary business. The more recent allegedly politically charged processes include amendments to the Higher Education Act, aiming to oust the Central European University from Hungary,168 or the NGO law, using money laundering rules to undermine civil society organisations169 in a fashion familiar from Russia’s and Israel’s laws requiring foreign-funded NGOs to declare themselves foreign agents. Challenges against these laws were also brought before the Constitutional Court for violations of the Fundamental Law. Yet, after the Commission closed the dialogue phase of the infringement action and turned to the CJEU, the Hungarian Constitutional Court suspended in the summer of 2018 the proceedings in two of its cases in the name of European judicial dialogue.170 This suspension may be a little too polite, as the cases before the Constitutional Court affect the validity of these amendments under the Fundamental Law, and not EU law. Thus, the Constitutional Court’s resolutions can be seen as a delaying tactic to hold off on its own decision till the latest possible point, instead of attempting to defend fundamental rights under the Fundamental Law.

160 Ibid, 125. 161 61/2009 IM decree on editing of legal rules, Art 88(2). 162 61/2009 IM decree, Arts 91–95. 163 Required by decree 302/2010 (n 157) Art 17. 164 Somssich, ‘Process and Methodology’ (n 151) 130. 165 AB 32/2012 (VII. 4.) (n 80) para 43. 166 Somssich, ‘Process and Methodology’ (n 151) 135. 167 EUB-1/2018 (n 2). 168 See infringement action no 20172076 (n 41). 169 Infringement action no. 20172110 (n 41). 170 AB ruling II/1036/2017, 4 June 2018 (re infringement no 20172076); AB ruling IV/1810/2017, 4 June 2018 (re infringement no 20172110) (both in Hungarian).

502  Renáta Uitz

D.  Ordinary Courts: Requests for Preliminary Rulings Ordinary courts have long played a major role in giving effect to EU law in individual cases.171 The Supreme Court’s (now: Kúria) guidance to ensure the uniform application of the law has played an important role in this respect. The reception of EU law in ordinary courts is marked by a genuine aspiration for compliance, where shortcomings do not result from intentional non-compliance.172 Nonetheless, human rights cases under EU law are marginal in the j­ urisprudence.173 In recent years the CJEU has rejected a number of preliminary references in Charter cases where Hungarian courts were not mindful of the limited applicability of the Charter (Article 51). In an article the Hungarian CJEU judge, Endre Juhász attributed these cases to the referring judges’ lack of attention to detail.174 In recent years the Constitutional Court started receiving complaints contesting the refusal of courts to request a preliminary ruling from the CJEU175 as a violation of fair trial rights and access to a lawful judge under the Fundamental Law.176 The Constitutional Court has held consistently that there is no automatic right of access to the CJEU through Hungarian courts. Accordingly, trial courts have discretion in making a reference for a preliminary ruling, and the Constitutional Court is not ready to review the exercise of judicial discretion in this respect, so long as the trial court provides reasons.177 The Constitutional Court held that the Hungarian Parliament violated the fair trial guarantees under the Fundamental Law when it failed to require courts to provide reasons for refusing a request for preliminary ruling.178 This unconstitutional omission was remedied a month after the deadline with an amendment of the Code of Civil Procedure (Article 155/A (2)–(3)).179 Regarding its own powers in the preliminary reference context, the Constitutional Court found that it cannot make a reference for a preliminary ruling in a procedure for constitutional complaint.180

VI.  Resulting Relationship between EU Law and National Law The relationship between EU law and Hungarian law is best described as a dynamic exchange, frequently ridden with conflicts and contestation. Hungary is certainly not the only Member 171 Hungarian preliminary references are leading the statistics of those who joined the Union in 2004. See Endre Juhász, ‘Magyarország és az Európai Unió Bírósága’ [‘Hungary and the CJEU’], (2014) 61(4) Közgazdasági Szemle 373, 383. 172 Márton Varju, ‘The Judicial Reception of EU Law’, in M Varju and E Várnay (eds), The Law of the European Union in Hungary (Budapest, HVG-Orac, 2014) 183, 187. 173 Ibid 195 f. 174 Juhász, ‘Magyarország’ (n 171) 378. 175 Note that whether the refusal of trial courts to make a reference for preliminary ruling to the CJEU in violation on the CILFIT criteria violates fair trial rights (ECHR, Art 6) is the central issue in a Hungarian case concerning disability pension before the ECtHR (ECtHR 28.08.2018 Application No 60934/13 (Somorjai v Hungary) ECLI:CE:ECHR:2018:08 28JUD00609341). 176 FLHU, Art XXVII(1). 177 AB ruling 3110/2014 (IV. 17.) para 24; AB ruling 3165/2014 (V. 23.) para 16; AB ruling 3037/2015 (II. 20.) paras 22–24; AB ruling 3050/2015 (III. 2.) para 17; AB ruling 3082/2016 (IV. 18.) paras 34 f; AB decision 3003/2017 (II. 1.) paras 28–30; AB ruling 3082/2016 (IV. 18.) paras 32-34, confirmed in AB ruling 3353/2018 (XI. 12.) para 15 (all in Hungarian). 178 AB decision 26/2015 (VII.21.) (in Hungarian). The deadline for parliament for passing the missing rules was set at December 31, 2015. 179 Act no CLXXX of 2015 on amendments to the Code of Civil Procedure, Art 2. 180 AB 3165/2014 (n 177) para 19. The Constitutional Court left another request for preliminary ruling unaddresed for unrelated reasons in AB ruling 3286/2017 (XI. 14.) para 13.

Hungary  503 State that has an ambivalent relationship with the EU’s constitutional consensus181 or where various constitutional actors seek to negotiate the terms of reception of EU law.182 There are numerous instances where the Government and the Parliament do not fight EU institutions and EU law (especially if compliance results in access to funds) and the Hungarian judiciary also appears to be keen to display its European credentials. Hungarian courts also routinely make references for preliminary rulings in cases with an EU law dimension. The Hungarian Government’s attitude is special in that it uses European processes in order to defy shared European constitutional values. In situations the Hungarian Government finds it politically salient, it prefers narratives built on defending national constitutional identity and – more and more openly – sovereignty in the face of European imposition. This conclusion resonates with Tanja Börzel and Ulrich Sedelmeier’s finding – well taken in scholarship on EU policy studies – that ‘[i]lliberal governments can exploit the difficulties of mobilizing the sanctions contained in Article 7 TEU while otherwise complying with the acquis to continue enjoying the benefits of EU membership’.183 The gap, of course, is explained by the difference in powers (and leverage) available to European actors in rule of law compliance on the one hand, and compliance with the acquis on the other. Bridging this gap is all the more important in light of ample evidence that strategic national actors pursue questionable political goals through procedures and tools make available to them by EU law on the European level.

References K Auel, ‘The Europeanization of National Parliaments’, in J Magone (ed), The Routledge Handbook of European Politics (Abingdon, Routledge, 2014) 366. Á Bátory, ‘The National Coordination of EU Policy in Hungary: Patterns of Continuity and Change’, (2012) 90 Public Administration 922. Á Bátory, ‘Defying the Commission: Creative Compliance and Respect for the Rule of Law in the EU’, (2016) 94 Public Administration 685. L Blutman, ‘A magyar Lisszabon-határozat: Befejezetlen szimfónia luxemburgi hangnemben’ [‘The Hungarian Lisbon Decision: Unfinished Symphony in Luxemburg Tone’], (2010) 1 (2) Alkotmánybírósági Szemle 90. T Börzel and U Sedelmeier, ‘Larger and More Law Abiding? The Impact of Enlargement on Compliance in the European Union’, (2017) 24 Journal of European Public Policy 197. A Bragyova, ‘No New(s), Good News?’, in G Tóth (ed), Constitution for a Disunited Nation: On Hungary’s 2011 Fundamental Law (Budapest, CEU Press, 2012) 335. N Chronowski, ‘Integrálódó’ alkotmányjog [‘Integrating’ constitutional law] (Budapest and Pécs, Dialóg Campus, 2005). N Chronowski, ‘A magyar alkotmánybíráskodás és a közös európai alkotmányos standardok’ [‘Hungarian Constitutional Court Jurisdiction and Common European Constitutional Standards’], (2014) 20 (2) JURA, 26.

181 See eg Ulrich Sedelmeier, ‘Anchoring Democracy from Above? The European Union and Democratic Backsliding in Hungary and Romania’, (2014) 52 Journal of Common Market Studies 105 (explaining the differences in the success of EU measures). 182 Generally Sadurski, Constitutionalism (n 17); Franz C Mayer, ‘Constitutional comparativism in action. The ­example of general principles of EU law and how they are made – a German perspective’, (2013) 11 International Journal of Constitutional Law 1003. 183 Tanja Börzel and Ulrich Sedelmeier, ‘Larger and More Law Abiding? The Impact of Enlargement on Compliance in the European Union’, (2017) 24 Journal of European Public Policy 197, 211.

504  Renáta Uitz N Chronowski and E Csatlós, ‘Judicial Dialogue or National Monologue? The International Law and Hungarian Courts’, (2013) 1 ELTE Law Journal 7. M Claes and M de Visser, ‘Reflections on Comparative Method in European Constitutional Law’, in M Adams and J Bomhoff (eds), Practice and Theory in Comparative Law (Cambridge, Cambridge University Press, 2012) 143. M Dezső, Constitutional Law in Hungary (Alphen aan den Rijn, Kluwer Law International, 2010). M Dezső and A Vincze, Magyar alkotmányosság az európai integrációban [Hungarian Constitutionalism in European Integration], 2nd edn (Budapest, HVG-ORAC, 2012). F Fazekas, ‘EU law and the Hungarian Constitutional Court’, in M Varjú and E Várnay (eds), The Law of the European Union in Hungary: Institutions, Processes and the Law (Budapest, HVG-ORAC, 2014) 32. F Fazekas, ‘Az uniós tagság alkotmányos alapjai az Alaptörvény előtt és után’ [‘The Constitutional Foundations of EU Membership Before and After the Fundamental Law’], (2015) 1 ProFuturo, 29. F Fazekas and M Varju, ‘The Reception of European Union Law in Hungary, The Constitutional Court and the Hungarian Judiciary’, (2011) 48 CML Rev 1945. M Garavoglia, ‘What Hungary’s Referendum Says About Europe’s Politics and Policymaking’, 3 October 2018, www.brookings.edu/blog/order-from-chaos/2016/10/03/what-hungarys-referendum-says-abouteuropes-politics-and-policymaking/. F Gárdos-Orosz, ‘Újabb áttekintés az uniós jog alkalmazásának magyarországi alkotmányos kereteiről’ [‘Recent Overview on the Constitutional Framework of the Application of Union Law in Hungary’], (2013) 6(4) Közjogi Szemle, 44. H Grabbe, The EU’s Transformative Power. Europeanization Through Conditionality in Central and Eastern Europe (London, Palgrave Macmillan, 2006). G Ilonszki, ‘The Hungarian Parliament and EU Affairs: A Modest Actor Dominated by the Executive’, in C Hefftler et al (eds), The Palgrave Handbook of National Parliaments in the European Union (London, Palgrave Macmillan, 2015) 531. E Juhász, ‘Magyarország és az Európai Unió Bírósága’ [‘Hungary and the CJEU’], (2014) 61 Közgazdasági Szemle 373. A Juhász-Tóth, ‘European Union Law and the Hungarian Parliament: Wasted Opportunities?’, in M Varjú and E Várnay (eds), The Law of the European Union in Hungary: Institutions, Processes and the Law (Budapest, HVG-ORAC, 2014) 77. K Kelemen, ‘The Agreement on a Unified Patent Court Cannot Be Ratified by Hungary, the Constitutional Court Says’, 1 August 2018, www.diritticomparati.it/agreement-unified-patent-court-cannot-ratifiedhungary-constitutional-court-says/. T Kende and T Szűcs (eds), Európai közjog es politika [European public law and policy] (Budapest, Wolters Kluwer, 2006). FC Mayer, ‘Constitutional comparativism in action. The example of general principles of EU law and how they are made – a German perspective’, (2013) 11 International Journal of Constitutional Law 1003. OLAF, The OLAF report 2017. Eighteenth report of the European Anti-Fraud Office. 1 January to 31 December 2017 (Luxembourg, Publications Office of the EU, 2018). P Paczolay, ‘A magyar alkotmány jövője és az uniós csatlakozás’ [‘The Future of the Hungarian Constitution and Accession to the Union’], (2004) 13 Politikatudományi Szemle 31. D Piqani, ‘The Role of National Constitutional Courts in Issues of Compliance’, in M Cremona (ed), Compliance and Enforcement of EU Law (Oxford, OUP, 2012) 133. W Sadurski, Constitutionalism and the Enlargement of Europe (Oxford, OUP, 2012). A Sajó, ‘Learning Co-operative Constitutionalism the Hard Way: the Hungarian Constitutional Court Shying Away from EU Supremacy’, (2004) 2 Zeitschrift für Staats- und Europawissenschaften 351. U Sedelmeier, ‘Anchoring Democracy from Above? The European Union and Democratic Backsliding in Hungary and Romania’, (2014) 52 Journal of Common Market Studies 105. R Somssich, ‘The Process and Methodology of Coordinating the Transposition of EU Law in Hungary’, in M Varju and E Várnay (eds), The Law of the European Union in Hungary (Budapest, HVG-Orac, 2014) 112.

Hungary  505 P Sonnevend, ‘Alapvető jogaink a csatlakozás után’ [‘Our Fundamental Rights After EU Accession’], (2003) 7(2) Fundamentum, 27. A Tatham, Central European Constitutional Courts in the Face of EU Membership. The Influence of the German Model in Hungary and Poland (Leiden, Brill, 2013). GA Tóth (ed), Constitution for a Disunited Nation: On Hungary’s 2011 Fundamental Law (Budapest, CEU Press, 2012). L Trócsányi, Az alkotmányozás dilemmái. Alkotmányos identitás és európai integráció [Dilemmas of Constitution-Making. Constitutional Identity and European Integration], (Budapest, HVG-ORAC, 2014). L Trócsányi and L Csink, ‘Alkotmány v közösségi jog: Az Alkotmánybíróság helye az Európai Unióban’ [‘Constitution v Community Law: the Status of the Constitutional Court in the EU’], (2008) 63(2) Jogtudományi Közlöny, 63. R Uitz, ‘EU Law and the Hungarian Constitutional Court: Lessons from the First Encounter’, in W Sadurski, J Ziller and K Żurek (eds), Après Enlargement: Legal and Political Responses in Central and Eastern Europe (2006) 41. R Uitz, ‘National Constitutional Identity in the European Constitutional Project: A Recipe for Exposing Cover Ups and Masquerades’, Verfassungsblog, 11 November 2016, https://verfassungsblog.de/ national-constitutional-identity-in-the-european-constitutional-project-a-recipe-for-exposing-coverups-and-masquerades/. R Uitz, ‘Reinventing Hungary with Revolutionary Fervor: The Declaration of National Cooperation as a Readers’ Guide to the Fundamental Law of 2011’, in J Kovács and B Trencsényi (eds), Brave New Hungary: Mapping the ‘System of National Cooperation’ (Lanham MD, Lexington Books, 2020) 9. R Uitz and A Sajó, ‘The Sovereign Strikes Back. A Judicial Perspective on Multi-Layered Constitutionalism’, (2018) 20 European Journal of Law Reform 97. M Varju, ‘The Judicial Reception of EU Law’, in M Varju and E Várnay (eds), The Law of the European Union in Hungary (Budapest, HVG-Orac, 2014) 183. I Vörös, Csoportkép Laokoónnal. A magyar jog és az alkotmánybíráskodás vívódása az európai joggal [Tableau with Laokoon. The Struggle of the Hungarian Legal System and Constitutional Justice with European law] (Budapest, HVG-ORAC, 2012).

506

18 Malta PETER G XUEREB 1 AND MIREILLE M CARUANA 2

I.  Main Characteristics of the National Constitutional System A.  The Constitution of Malta The Constitution of Malta3 is the supreme law of the land. Malta was a British colony from 1813 to 1964 and as a consequence constitutional and administrative law in Malta have been heavily influenced by the English common law position, even though English common law was never formally part of Maltese law. Under the 1964 Constitution – amended in 1974 and 1987 – Malta is a democratic republic with a unicameral House of Representatives of at least 65 members. The country has proportional representation using the single transferable vote system. A party which obtains a majority of votes but a minority of seats is allocated additional seats to give it an overall majority of one. The House is composed of an odd number of members elected for a single legislative term of five years: five members are returned from each of 13 electoral districts using the single transferable vote electoral system, but additional members are elected in cases of disproportionality (where a party with an absolute majority of votes fails to win an absolute majority of seats and where only candidates from two parties are elected). The House may not sit for longer than five years. The President is the head of state and is elected for a fiveyear period of office by the House of Representatives. The incumbent has executive authority but must act on the cabinet’s advice and the position is therefore largely ceremonial. The Prime Minister and Leader of the Opposition are both appointed by the President. The cabinet is appointed by the President on the advice of the Prime Minister. All appointees must be members of Parliament.

1 The author is grateful for the granting of permission to use certain sections from his report ‘The Constitution of Malta: Reflections on New Mechanisms for Synchrony of Values in Different Levels of Governance’, in A Albi and S Bardutzky (eds), The Role of National Constitutions in the European and Global Governance (The Hague, TMC Asser Press, 2019) 141. Professor Xuereb is jointly responsible for sections II and III and mainly responsible for sections IV–VI. 2 Dr Caruana is mainly responsible for section I and jointly responsible for sections II and III. 3 The Constitution of Malta (hereinafter MTC) is available at www.justiceservices.gov.mt/DownloadDocument. aspx?app=lom&itemid=8566.

508  Peter G Xuereb and Mireille M Caruana The first chapter of Malta’s Constitution is dedicated to underlining fundamental characteristics applicable to the state of Malta, as follows: Article 1. The Republic and its Territories. Article 2. Religion. Article 3. National Flag. Article 4. National Anthem. Article 5. Language. Article 6. Constitution to be supreme law.

Of note among these are the clauses on neutrality and religion. As regards the neutrality clause, this has been the subject of contested interpretation at a political level over the years,4 as well as controversy in the context of on-going discussions on Constitutional reform.5 The relevant sub-article6 states that Malta is a neutral state actively pursuing peace, security and social progress among all nations by adhering to a policy of non-alignment and refusing to participate in any military alliance.

Under Article 2 of the Constitution, Roman Catholic Christianity is the state religion and must be taught in state schools: (1) The religion of Malta is the Roman Catholic Apostolic Religion. (2) The authorities of the Roman Catholic Apostolic Church have the duty and the right to teach which principles are right and which are wrong. (3) Religious teaching of the Roman Catholic Apostolic Faith shall be provided in all State schools as part of compulsory education.

The Maltese Constitution follows the British model, with one major difference: while under the British system Parliament is supreme and a Parliament can never bind a future Parliament, under the Maltese Constitution, Parliamentary sovereignty is limited by the supremacy clause of the Maltese Constitution, namely Article 6 of the Constitution, which reads: Subject to the provisions of sub-articles (7) and (9) of article 47 and of article 66 of this Constitution, if any other law is inconsistent with this Constitution, this Constitution shall prevail and the other law shall, to the extent of the inconsistency, be void.7

B.  The System of the Judiciary The Maltese judicial system is essentially a two-tier system with a court of first instance presided by a judge or magistrate, and a Court of Appeal composed of three judges when the appeal is from a court presided by a judge (superior appeal), or one judge when the appeal is from a court presided over by a magistrate (inferior appeal). The courts presided by a judge are the superior 4 See eg Kurt Sansone, ‘Labour U-turn on Partnership for Peace’, Times of Malta 04.09.2011, www.timesofmalta.com/ articles/view/20110904/local/Labour-U-turn-on-Partnership-for-Peace.383100. More recently, see Jacob Borg, ‘Joseph Muscat faced mini-revolt in PL during 2013 election campaign’, The Malta Independent on Sunday 24.05.2015, www. independent.com.mt/articles/2015-05-24/local-news/Joseph-Muscat-faced-mini-revolt-in-PL-during-2013-electioncampaign-6736136058. 5 See eg Michael Grech and Charles Miceli, ‘Our neutrality clauses’, Times of Malta 30.09.2014, www.timesofmalta. com/articles/view/20140930/opinion/Our-neutrality-clauses.537835. 6 MTC Art 1(3). 7 MTC Art 6.

Malta  509 courts, the courts presided by a magistrate are the inferior courts. There are also various tribunals for specialised areas with varying degrees of competence (eg tax, social security, electronic communications); in almost all cases there is a right of appeal on points of law from decisions of these tribunals to the Court of Appeal (composed of a single judge). When Malta attained independence in 1964, the Constitutional Court was established as the appellate court in matters relating to the Constitution. The Constitutional Court, designated as a Superior Court, is composed of three judges, who are normally those judges constituting one of the chambers of the Court of Appeal. It is made up of the Chief Justice and such number of other judges as may be prescribed by any law in force in Malta.8 The Constitutional Court hears inter alia appeals from judgments of the First Hall of the Civil Court, in its constitutional jurisdiction,9 concerning any alleged breach of fundamental human rights, as protected in the Constitution and/or in the European Convention on Human Rights and Fundamental Freedoms (ECHR), as transposed to Maltese law by virtue of the European Convention Act.10 The Constitutional Court also hears issues regarding the validity of laws, but there is no jurisdiction of the kind that exists in other Member States11 according to which an opinion can be sought ex ante from the Court as to constitutionality or otherwise of any proposed law. Article 95(2) of the Constitution of Malta establishes the Constitutional Court and reads as follows: (2) One of the Superior Courts, composed of such three judges as could, in accordance with any law for the time being in force in Malta, compose the Court of Appeal, shall be known as the Constitutional Court and shall have jurisdiction to hear and determine … (c) appeals from decisions of the Civil Court, First Hall, under article 46 of this Constitution [fundamental rights and freedoms of the individual as protected under Chapter IV of the Constitution]; (d) appeals from decisions of any court of original jurisdiction in Malta as to the interpretation of this Constitution other than those which may fall under article 46 of this Constitution; (e) appeals from decisions of any court of original jurisdiction in Malta on questions as to the validity of laws other than those which may fall under article 46 of this Constitution; and (f) any question decided by a court of original jurisdiction in Malta together with any of the questions referred to in the foregoing paragraphs of this sub-article on which an appeal has been made to the Constitutional Court: Provided that nothing in this paragraph shall preclude an appeal being brought separately before the Court of Appeal in accordance with any law for the time being in force in Malta.

Under Article 95(2)(f) therefore the Constitutional Court has extensive jurisdiction if such issues are linked to Constitutional questions. According to Article 3(4) of the European Convention Act12 the Constitutional Court shall in addition to the jurisdiction conferred on it by Article 95 of the Constitution, have jurisdiction to hear and determine all appeals under this Act and exercise all such powers as are conferred on it by this Act. 8 MTC Art 95(2) and (6). 9 The jurisdiction of the First Hall of the Civil Court is defined in the Code of Organisation and Civil Procedure, Art 32. 10 Act XIV of 1987. 11 Such as exists in Art 26 of the Irish Constitution. 12 Cap 319 of the Laws of Malta, www.justiceservices.gov.mt/DownloadDocument.aspx?app=lom&itemid=8795&l=1.

510  Peter G Xuereb and Mireille M Caruana There have to date been no cases wherein our Courts were called upon to recognise or deny the final character of the decisions of the Court of Justice of the EU.

II.  Constitutional Culture The Constitution of 1964 has proved itself to be a living Constitution based on the parliamentary model. Parliament is organised into committees that feed into the parliamentary sittings. Procedures are governed by Westminster Parliament-inspired rules, standing orders and conventions. It has seen major amendment, especially in the 1974 conversion of Malta into a Republic. The system is one of division of powers with checks and balances, and a strong commitment to human rights and individual freedoms. The President of Malta is the highest authority in the land and has important powers centred around the operation of democracy, including the power to dissolve parliament in specified cases, but otherwise, as already stated above, the role is largely ceremonial. He or she is expected to give consent to all legislation passed by Parliament. A Constitutional Court ensures the application of the Constitution and is now itself so constituted as to ensure that it will always be composed, sitting and accessible.

A.  Interpretation of the Constitution An unresolved question is the issue of self-referential interpretation of the Constitution by the Constitutional Court, meaning the question whether certain constitutional provisions dictate how others will be interpreted. For example, Article 2 privileges the Roman Catholic religion as the declared religion of Malta. Other religions are freely allowed to operate under the fundamental rights clauses; however, there appears to be this intended preference for the Roman Catholic religion, and as long as Article 2 exists in this form one might construct an argument that other provisions of law and of the Constitution itself are to be interpreted and applied against its backdrop and context. This could, for example, impact on the interpretation of the right to life as enshrined in the Constitution and possibly be brought up in the context of any proposed new rules on abortion or euthanasia. Just by way of hypothetical example, then, Article 2 may be seen as providing a limit to (a) the capacity of the Maltese state to embrace plurality of values and/or (b) the ability of the state institutions to agree to transfer certain powers or adopt certain policies and/or to implement rules or obligations dictated by externally-made undertakings that went counter to its spirit, and/or (c) the capacity of the Maltese institutions, including the Constitutional and other courts to interpret union law as required by the Court of Justice when that interpretation conflicts with the essence of Article 2, and/or (d) the capacity of the Maltese institutions, including the Courts, to regard EU law which conflict with Article 2 or even with an interpretation of other clauses of the Constitution themselves read in its light (eg hypothetically, the alleged right to life of the unborn child) as applicable and binding – despite the quasi-constitutional nature of the European Union Act of 2003 whereby Malta accepted the effectiveness of the Union acquis. Much would depend on the interpretative effect of Article 2 on the rest of the Constitution. Does it colour the interpretation of other provisions of the Constitution, including the possible

Malta  511 legitimate scope of Article 65 itself – ie the ability of the state under the Constitution to transfer ‘powers’ to the Union, for example in regard to family policy, or abortion and so on?

B.  National Identity Much also depends on the link between provisions such as Article 2 and (especially) the human rights provisions in Title IV and the notion of ‘national identity’. After all, Article 4 of the Treaty on European Union (TEU) preserves and safeguards the national identity of Member States against encroachment by the Union. The Maltese Constitution, in common with many other Member State Constitutions, but differently from some others, does not contain a clause that expressly safeguards the national identity. There is no definition of national identity.13 It is true that there is a definition (arguably a partial and rather narrow one) of national identity in Article 4 of the TEU itself. However, this does not of itself dictate the position at national level, and it has not prevented some Member states from seeking to conceptualise, define, entrench and safeguard national identity in broader terms (Germany and Italy are cases in point). It is instructive and thought-provoking to look across at the practice of other Member States, especially those with which Malta had an affinity – social, cultural and legal as well as constitutional. Ireland certainly goes well beyond anything to be found elsewhere in focusing on the main tenets of the Catholic religion and imbuing the Preamble of the Constitution of Eire, and presumably the rest of the Constitution, with a specific religion’s ethos. Yet, divorce and abortion, and more recently same-sex marriage, have been provided for by law. So also in Malta while abortion is always illegal, divorce and same-sex marriage have recently been put in place, and other laws in the pipeline further the notion of a secular state, embracing not only pluralism of religion but plurality of values and value systems. If this continues, a rather more flexible and open view of the import of Article 2 of the Constitution is inevitable. It arguably already is the case that the road is closed to a ‘fundamentalist’ interpretation of the effect of Article 2. At the very least, Article 2 (it would appear) cannot now be read as colouring, far less driving, the interpretation of the rest of the Constitution and imposing on the Courts an obligation to apply such a rule of interpretation. Parliament has already adopted legislation that is at variance with the tenets of the Catholic religion with no challenge to its constitutional validity being mounted in the Constitutional Court. Any objections of a principled kind were based on ­religious sentiment and personal positioning for each voting member of Parliament, occurred in the parliamentary debates, and were put to the vote in the ordinary way without any constitutional amendment being considered as a necessary anterior step. Another indication may (or may not) be the campaign mounted by the lobby group ‘Gift of Life’ to have the right to life of the unborn child entrenched in the Constitution, unsuccessfully but not clearly for the reason that it was accepted that the right to life as currently framed actually already protects such life, thus rendering any amendment redundant. The position is ambiguous. The real question is whether in any proper political debate the people would support a dilution of Article 2 to read more generally and not as exclusively privileging one religion or its removal entirely, thus bolstering secularisation as a principle.



13 Unlike,

for example, the German Constitution.

512  Peter G Xuereb and Mireille M Caruana

C.  Culture and National Identity – And a Human Rights Culture that is Culturally Sensitive One key related question is whether, apart from the historical dominance of the Catholic religion, ongoing neutrality and human rights provisions, inter alia, possibly are of their nature essential components of the concept of identity, albeit mediately through the concept of culture. If so, they can be protected mediately through that concept. Otherwise they may need to be spelled out in a revised Article 1, for example. Unless one provides for a narrow notion of constitutional identity – which arguably is the way the TEU itself currently uses the concept by referring in the main to structural constitutional and organisational elements – then it would be a matter of protecting essential diversity – in line with a possibly revised Article 4 TEU which broadened the current concept – by providing a safeguard for a cultural-social-legal concept of national/European identity that encapsulated elements such as culture, religious heritage, the public administration/ administrative model, the social model (in so far as this concept can be satisfactorily defined), and very importantly the values of a society, which are most often reflected in the way in which that culture mediates conflicting rights. This often translates into the mediation of competing fundamental rights. A word here about fundamental rights under the Maltese Constitution would be pertinent. It has already been noted that since 1961, Malta’s constitutional documents have featured a list of fundamental rights with corresponding action for enforcement. The Constitution of 1964 is the main document but the list of rights was enhanced by the European Convention Act of 1987, which extended the same right of action to the new rights derived from the ECHR. As to the general principles, they have tended to emerge almost as a matter of custom in the gurisprudenza (judgments) of the courts, especially of the First Hall of the Civil Court in its Constitutional jurisdiction, and of the Constitutional Court. General legal principles are not applied in a way that they could be considered as sources of fundamental rights protection. Article 3214 of the Constitution provides as follows: Whereas every person in Malta is entitled to the fundamental rights and freedoms of the individual, that is to say, the right, whatever his race, place of origin, political opinions, colour, creed, sex, sexual orientation or gender identity, but subject to respect for the rights and freedoms of others and for the public interest, to each and all of the following, namely (a) life, liberty, security of the person, the enjoyment of property and the protection of the law; (b) freedom of conscience, of expression and of peaceful assembly and association; and (c) respect for his private and family life, the subsequent provisions of this Chapter shall have effect for the purpose of affording protection to the aforesaid rights and freedoms, subject to such limitations of that protection as are contained in those provisions being limitations designed to ensure that the enjoyment of the said rights and freedoms by any individual does not prejudice the rights and freedoms of others or the public interest. [emphasis added].

This pre-ambular set of limitations is reflected in specific provisions legitimising derogation in the case of each right (except for protection from inhuman or degrading treatment), such as ‘by a provision of law which is reasonably required in the interest of public safety, public order, public morality or decency, public health or the protection of the rights and freedoms of others’. Several

14 Which, while entrenched, is at the same time declared by Art 46 MTC to be non-justiciable, unlike the rest of the provisions in Chapter IV MTC, which are declared to be justiciable.

Malta  513 provisions speak both of ‘reasonably required’ and ‘reasonably justifiable in a democratic society’. The standard is substantially similar to that of the ECHR.

D.  Fundamental Rights/Identity/ Culture and the Economic Freedoms of the Treaties Issues can arise where economic freedoms clash with classic rights as interpreted and applied in Malta in certain cases. For example, abortion, seen as an assault on ‘life’, remains a criminal offence under Maltese law. On this basis attempts to provide this ‘service’ (as described by the Court of Justice) in Malta or to procure or otherwise arrange for it to be obtained outside Malta have met with police or other action. There has been one difficult case in which a father attempted to stop his partner from proceeding abroad for an abortion.15 The applicant requested the First Hall of the Civil Court to declare that his partner’s imminent deportation by the authorities (on the ground of her being present in Malta illegally, but against a background of her declaring that if returned to Russia she would procure the abortion of her unborn child in Russia, where this was legal) would be in violation of the unborn child’s right to life and that his own fundamental human right to protection of family life would be violated. Another example involved the surgical separation of twins that were joined at birth where one of them (the more viable one) stood a chance of surviving if the operation were carried out, but with the certain knowledge that the other would die ‘during’ any operation to separate them, and that both would certainly die without any surgical intervention.16 The matter was finally decided in an English court (where the twins had been taken) in favour of the operation (carried out in the UK), but the outcome would conceivably have been different in a Maltese court (and hospital), and any economic right or freedom of movement right in play might possibly have been relegated to second place by a Maltese court. As far as social rights are concerned, Malta is a republic ‘founded on work and fundamental rights and freedoms’ (Article 1(1)) and ‘committed to social progress among nations’ (Article 1(3)), and the Constitution seeks to protect social principles by declaration (non-­ justiciable principles) in Chapter III. This creates a very strong presumption in favour of these principles and a strong rule of interpretation that all legislation must be in conformity with them. Thus, the Maltese courts are called upon to ensure balance between economic aims and rights, and social aims and individual rights. Nevertheless, as has been strongly asserted by a group of NGOs in their recent replies and recommendations in response to a government consultation exercise on the protection of human rights, second-generation rights sadly remain non-justiciable as such, and third-generation rights are totally absent from the Constitution. Any future constitutional review should seek to remedy this situation. It is this flexibility in terms of interpretation added to the possibility of amendment of most provisions of the Constitution that renders the Constitution a living Constitution that could adapt in 2003, for example, to membership of the European Union with a minimum of amendments. It is clear that the Constitution both sets out essential procedural requirements and also allows for the practical resolution of problems that may arise in practice, not least in the latter case by the provision of a court structure capable of delivering interpretations in line with Malta’s international obligations. 15 ‘Man claims planned abortion would breach his rights’, Times of Malta 17.12.2003, www.timesofmalta.com/articles/ view/20031217/local/man-claims-planned-abortion-would-breach-his-rights.134109. 16 Clare Dyer, ‘Conjoined twins separated after long legal battle’, (2000) 321 British Medical Journal, 1175.

514  Peter G Xuereb and Mireille M Caruana

III.  The Constitutional Foundations of EU-Membership A.  Practice and Doctrinal Debate In pre-membership Malta, the questions were posed: ‘Does membership involve per se a conflict with the Constitution so that an amendment of the Constitution is required, bearing in mind that it would be a Maltese Act of Parliament which brought in the Treaties?’ Or, more precisely, would a law ratifying the Treaties (Treaty) be ‘inconsistent with the Constitution’ as per Article 6? One implicit subquestion was whether the ‘transfer of law-making power’ to the Union institutions per se involved a conflict with the Constitution. The second implicit subquestion was whether, apart from this first issue, ‘there was any present or foreseeable future unresolvable conflict between the Acquis Communautaire (now the acquis of the Union), including the Constitutional acquis, and the Constitution?’ In the end the view prevailed that membership of the Union did not require or involve the repeal of Article 6. Another question is still as relevant today as it was in 2001. Can the Maltese Constitution be interpreted as providing or (if deemed necessary) be made to provide an even more secure constitutional basis for the adoption of the constitutional and other acquis coupled with the preservation of the principle of supremacy of the Constitution, the Constitution reflecting Malta’s international obligations as signatory to the Treaties while remaining the expression of the ultimate sovereignty of the people?17 Further constitutional development of the European Union can and should be absorbable if this also includes further provision for solidarity vis-à-vis the essential identity, and essential interests, of each Member State.18 If we all wish the Union to become stronger and more effective, we must adopt mechanisms that ensure input and output legitimacy and fairness in all decision-making processes. Hence, the authors repeat a call made at the time of the debate on the future of Europe and of the accession of Malta to the Union, for the enshrining of an active procedural and substantive principle of solidarity.19 As it turned out, the only such clause to be inserted was a discreet and specific Article 80 TFEU, which is only being employed as originally intended some 10 years later. Arguably, a general clause on those lines would have provided a solid basis for much more solidary action in the intervening years. Until the amendment introduced by the European Union Act of 200320 whereby the Treaty of Accession21 was ratified according to Maltese law, the Constitution made no reference either to the international legal order in general or to the effects in law of the eventuality of Malta joining the European Union. The Maltese legal position was fully dualist. The matter of incorporation of treaties was governed by the Ratification of Treaties Act22 which provides in Article 3(3) that ‘No provision of a treaty shall become, or be enforceable as, part of the law of Malta except by or under an Act of Parliament.’ In effect, parliamentary approval is necessary via ratification or 17 Peter G Xuereb, ‘Constitutional Questions Raised by the Proposed Accession of Malta to the European Union in the General Context’, in A Kellerman, J de Zwaan and J Czuczai (eds), EU Enlargement: The Constitutional Impact at EU and National Level (The Hague, Asser Press, 2001) 229. 18 Peter G Xuereb, ‘The Future of Europe: Solidarity and Constitutionalism. Towards a Solidarity Model’, (2002) 27 European Law Review 643. 19 Ibid. 20 Act V of 2003 (Cap 460 of the Laws of Malta), www.justiceservices.gov.mt/DownloadDocument.aspx?app=lom&ite mid=8926&l=1. 21 [2003] OJ L 236. 22 Act V of 1983 (Cap 304 of the Laws of Malta), www.justiceservices.gov.mt/DownloadDocument.aspx?app=lom&ite mid=8783&l=1.

Malta  515 authorisation of ratification of any international treaty. The Maltese courts would only be able to give effect to international agreements insofar as these might have been incorporated into Maltese law by, or by virtue of, legislation passed by the House of Representatives for the purpose. The main relevant parts of Article 3 of the Ratification of Treaties Act states that (1) Where a treaty to which Malta becomes party after the coming into force of this Act is one which affects or concerns (a) the status of Malta under international law or the maintenance or support of such status, or (b) the security of Malta, its sovereignty, independence, unity or territorial integrity, or (c) the relationship of Malta with any multinational organization, agency, association or similar body, such treaty shall not enter into force with respect to Malta unless it has been ratified or its ratification has been authorised or approved in accordance with the provisions of [the Ratification of Treaties Act]. (2) A treaty to which subarticle (1) applies shall be ratified or shall have its ratification authorised or approved as follows: (a) where such treaty concerns a matter referred to in subarticle (1) (a) or (b) or contains any provision which is to become, or to be enforceable as, part of the law of Malta, by Act of Parliament; (b) in any other case, by Resolution of the House of Representatives. (3) No provision of a treaty shall become, or be enforceable as, part of the law of Malta except by or under an Act of Parliament.

Malta acceded to the European Union in April 2003 and a new Article 65 ‘Powers and Procedure of Parliament’ was introduced into the Constitution of Malta: (1) Subject to the provisions of this Constitution, Parliament may make laws for the peace, order and good government of Malta in conformity with full respect for human rights, generally accepted principles of international law and Malta’s international and regional obligations in particular those assumed by the treaty of accession to the European Union signed in Athens on the 16th April, 2003.

The Constitution contains no other reference to the constitutional effects which accession was intended to have on the Maltese legal system. No specific reference to the ‘limitation’ or ­‘transfer’ or ‘pooling’ of sovereignty or the exercise of sovereignty as linked to membership of the Union, as found in the Constitutions of some of the other Member States of the Union, was made. The effect of the amendment to Article 65 of the Constitution would appear to be that should Parliament enact any laws which are inconsistent with Malta’s obligations as a Member State of the Union, such laws will, to the extent of any incompatibility, be deemed null and void as being in breach of Article 65. In effect, Union law is made ‘supreme’ over any law passed by the Maltese Parliament including after the coming into force of the European Union Act and including over any subsequent Maltese law (ie no lex posterior rule would apply). The remaining question is as to any conflict that may occur between Union law (assuming it to be included in ‘any other law’ referred to in Article 6 of the Constitution) and one or more provisions of the Constitution itself.23 23 Peter Xuereb, ‘The Maltese Constitutional and Legal Order and EU Membership’, in A Kellermann et al (eds), The Impact of EU Accession on the Legal Orders of New EU Member States and (Pre-) Candidate Countries. Hopes and Fears (The Hague, Asser Press, 2006) 409.

516  Peter G Xuereb and Mireille M Caruana The general consensus in Malta is that there is no existing conflict between the acquis and any particular provision of the Constitution of Malta. The hypothetical question arises as to what the position would be should such a conflict occur in the future. It is not envisaged that the Constitution will be changed in order to create a conflict that did not previously exist. In the unlikely event that this were done deliberately, the grounds would presumably be that the matter was of the essence of ‘national identity’ (Article 4 TEU) and there would no doubt be questions as to whether it was not the case that the Union had in some way encroached on essential matters. In other words, there might be a question as to just who was failing in a duty of loyalty. But this is of course to argue hypothetically.24 Leaving aside questions of encroachment on national identity, the question resolves itself into the question as to what the position would be as a matter of Maltese Constitutional law should new Union law cause an irreconcilable (via interpretation and full dialogue) conflict to arise. Article 65 might then be turned to for an answer as it may be read as expressing a duty on Parliament to try to bring the Constitution into line with Union law in such an event, if necessary via an amendment of the Constitution, despite the enabling terminology (‘may’) used in Article 65; this in virtue of general principles, also referred to in Article 65, including those of international law, which oblige Malta to abide by its international obligations, but presumably saving always the protection of the essence of national identity.25 It might therefore be argued that the enactments and amendments referred to above have secured to most intents and purposes, and for most foreseeable realistic eventualities, the effective resolution of any conflict of legal provision in favour of the application of Union law, in line with Malta’s obligations under the Treaties. This appears to be the carefully expressed view of at least one Maltese judge who has written that ‘if a law were to be enacted by a European Union institution that would conflict with our Constitution, Article 65 of the Constitution read in conjunction with Article 3 of the European Union Act could probably be referred to [authors’ italics] so as to safeguard the principle of supremacy of European Union law’.26 This kind of openness may partly explain the rather accepting approach of Parliament to later apparent diminutions of sovereignty. In recognition of the dualist understanding of international law, the Parliament of Malta ­legislated for the direct application of EU law through the European Union Act27 which came into force upon accession on 1 May 2004. In doing so, Malta was following the UK example.28 This means that by the power of the Act the acquis would have the power of law as Maltese law, s­ olving the problem caused by the dualist approach. Article 3 of the European Union Act provides: (1) From the First day of May 2004, the Treaty and existing and future acts adopted by the European Union shall be binding on Malta and shall be part of the domestic law thereof under the conditions laid down in the Treaty. (2) Any provision of any law which from the said date is incompatible with Malta’s obligations under the Treaty or which derogates from any right given to any person by or under the Treaty shall to the extent that such law is incompatible with such obligations or to the extent that it derogates from such rights be without effect and unenforceable.29 24 Ibid. 25 Ibid. 26 David Paul Scicluna, ‘EU Law – A National Judge’s Perspective’, in E Buttigieg (ed), Enforcing One’s Rights under EU Law (La Valetta, Gutenberg Press, 2011) 37, 38. 27 Cap 460 of the Laws of Malta. 28 The UK, after signing and ratifying its Accession Treaty in 1972, gave internal legal effect to Community (now EU) Law by means of an Act of Parliament: the European Communities Act 1972. 29 EU Act 2003, Art 3.

Malta  517 The ‘supremacy’ of EU law over Maltese law emanates from this article of the European Union Act. To what extent this would apply if there were to be a conflict with the provisions of the Constitution is debatable. Parallelism can be drawn to the theoretical (and highly unlikely) scenario in which the ECHR, as incorporated into Maltese law by virtue of the European Convention Act, Act XIV of 1987, would be found to conflict with the Maltese Constitution. The same force afforded to the European Convention Act by the Maltese Court would probably be afforded to the European Union Act. Sammut opines that, however, given the unique nature of EU law and the rights and obligations that it entails, the fact that Malta voluntarily accepted to join the Union should be enough to convince any Maltese Court that, should this theoretical scenario happen in reality, as long as Malta wants to be part of the Union, EU law is supreme and should prevail even if there were to be a conflict with the Constitution. The sharing of sovereignty is voluntary and unlike a federation, if a country feels that it should no longer share its sovereignty with other Member States, then legally speaking, either opt-out or a withdrawal from the Union should be negotiated. Unlike a federation, the EU does not compel Member States to stay in the Union and in theory a Member State does not give up any sovereignty, but simply shares it with the rest of the Member States.30 Article 4(1) of the European Union Act aims to make the concepts of direct applicability and direct effect part of the Maltese legal system: All such rights, powers, liabilities, obligations and restrictions from time to time created or arising by or under the Treaty, and all such remedies and procedures from time to time provided for by or under the Treaty, that in accordance with the Treaty are without further enactment to be given legal effect or used in Malta, shall be recognised and available in Law, and be enforced, allowed and followed accordingly.

It deems law that under the EU Treaties is to be given immediate legal effect, to be directly enforceable in Malta. Sammut opines that accordingly, Maltese courts, which on the orthodox domestic approach to international law may not directly enforce a provision of an international treaty or a measure passed thereunder, are directed by this article to enforce any directly effective EU measure. In his words, there is no need for a fresh Act of incorporation to enable Malta to enforce each EU Treaty provision, regulation or directive that according to EU law has direct effect. Just as in the cases of France, Germany and Italy, the supremacy of EU law is recognised in Malta by virtue of a domestic legal process and legal theory – by means of an Act of Parliament.31 The European Union Act also provided for any international treaty concluded by the European Union through its external relations powers. The procedure laid down in Article 4 provides that with regard to treaties and international conventions which Malta may accede to as a Member State of the European Union, and treaties and international conventions which Malta is bound to ratify in its own name or on behalf of the European Union by virtue of its membership of the European Union, these would come into force one month following their being submitted in order to be discussed by the Standing Committee on Foreign and European Affairs.32 Also any financial obligations arising out of the Treaty obligations are to be a charge against the Consolidated Fund.33 30 Ivan Sammut, ‘The EU and Maltese Legal Orders: What Kind of Marriage Between Them?’, in P Xuereb (ed), Malta in the European Union: Five Years On and Looking to the Future (Malta, European Documentation and Research Centre, University of Malta, 2009) 97, 101 f. My emphasis. 31 Ibid. 32 EU Act Art 4(2). 33 EU Act Art 4(5).

518  Peter G Xuereb and Mireille M Caruana On one interpretation, the reference to ‘any provision of any law’ in Article 3 of the European Union Act arguably has to be read subject to Article 65 of the Constitution which provides that ‘Subject to the provisions of this Constitution …’ Parliament may make laws in conformity with Malta’s international obligations.34 The argument is that Article 65 operates on the one hand to enable Parliament to pass such a law as the European Union Act, bringing EU law into the Maltese legal order, and at the same time as a limit that safeguards the Constitution from erosion by such a law. The same would be argued for any later law making any new Treaty effective in Malta. However, this is purely hypothetical, as it is most unlikely that an EU law would directly conflict with anything contained in the Constitution. It follows from the majority opinion in legal circles in Malta that the provisions of Maltese law, including those of the European Union Act of 2003, should in almost all foreseeable ­circumstances guarantee the effectiveness, including the direct effect, of relevant Union law in Malta according to the terms of applicability and effect of Union law under Union law. This approach is not yet fully tested, however, and some might disagree about the effect of Article 65, so that the matter must be considered to be still open at this stage. As for the relationship between the Maltese courts and those of the European Union, the European Union Act provides that for the purposes of any proceedings before any court or other adjudicating authority in Malta, any question as to the meaning or effect of the Treaty, or as to the validity, meaning or effect of any instruments arising therefrom or thereunder, shall be treated as a question of law. If the issue is not referred to the Court of Justice of the European Union (CJEU), it must be determined as such in accordance with the principles laid down by, and any relevant decision of, the CJEU or any court attached thereto.35 This allows the preliminary reference procedure under Article 267 TFEU from the point of view of Maltese law. As for the judgments handed down by the EU Courts, judicial notice is taken of the Treaty, of the Official Journal of the European Union and of any decision of, or expression of opinion by, the CJEU or any court attached thereto on any such question as aforesaid, and the Official Journal is admissible as evidence of any instrument or any other Act of the European Union or of any of its institutions as may have been communicated thereby.36

B.  Charter of Fundamental Rights of the EU and General Principles of Law Since 1961, Malta’s constitutional documents have featured a list of fundamental rights with corresponding action for enforcement. The Constitution of 1964 is the main document but the list of rights was enhanced by the European Convention Act of 1987, which extended the same right of action to the new rights derived from the ECHR. The Charter of Fundamental Rights (CFR) may have been given an even stronger reach than is required by the Treaty on the European Union as it seems the Maltese courts will bring it into play even where the issue goes beyond the implementation of Union law.37



34 Xuereb,

‘Maltese Constitutional Order’ (n 23). Act Art 5(1). 36 EU Act Art 5(2). See also Sammut, ‘EU and Maltese Legal Orders’ (n 30). 37 Civil Court First Hall 20.11.2010(Busuttil Linda et v Muscat Dr Josie et). 35 EU

Malta  519

C.  European Integration Outside the EU Legal Order and its Limits The constitutionality of the European Stability Mechanism (ESM) was not overtly the primary issue for the Maltese Parliament. However, constitutional questions were especially at the root of interventions in Parliament made by former Prime Minister Alfred Sant.38 These included accountability to Parliament for use made of the European Financial Stability Facility, with the ratification treaty including a new clause that Malta’s representatives on the ESM Boards of Governors and Directors were to report annually to the House Public Accounts Committee. Otherwise, the Government of the day argued that ratifying the ESM was in the national­ interest as it offered a better financial alternative, with a surer cap on Malta’s exposure (of some 160 million euros)39 than the alternatives employed or which might be employed. The financial calculations appeared to dominate the parliamentary debate, but with underlying considerations of viability and control playing their part. There was no debate about the irreversibility or unconditionality of the commitment, as the issue was treated mainly as an issue of manageability. General discussion about the constitutionality of other proposed measures has taken place internally in the financial institutions, the financial press and some conferences. Amending the Constitution to allow for the Fiscal Compact’s ‘golden rule’ required some debate,40 but in the end the consensus not to act by amendment of the Constitution prevailed.41 Malta has not been subject to a bailout/austerity programme. The author’s view is that many in Malta would share much of the reasoning of the Constitutional Court of Portugal. It is worth recalling that the social progress of all nations is a fundamental principle and policy objective of the Constitution of Malta.42

IV.  Constitutional Limits to EU-integration A.  Limits to the (Further) Transfer of Powers Through Treaty Amendments 1.  Core Competences The starting point is that there is no limit expressly set out in the Constitution on the possible acceptance of wider competence for the Union, whether ‘exclusive’ or ‘shared’ in terms of the TEU. The position on ‘transfer of powers’ is a follows. Although there was arguably no need to make provision in the Constitution for EU membership, it was thought prudent to do so. The result was the amendment of Article 65(1) of the Constitution, as given above, which took this

38 ‘EU Financial Treaty Ratified’, The Times of Malta 05.07.2012, https://timesofmalta.com/articles/view/EU-financialtreaty-ratified.427491; and Miriam Dalli, ‘Malta supports ESM bailout for Spain’, Malta Today 19.06.2012, www. maltatoday.com.mt/news/national/18986/malta-supports-esm-bailout-for-spain-20120619#.XT6-4IDVLIU. 39 For the Budget for 2015, the totals of Government revenue and expenditure are projected at around 3.5 billion euros and 3.7 billion euros, respectively (www.act.com.mt/articles-publications/Budget-Measures-Implementation-2015.php). 40 See eg Silvio Attard and Alexander Demarco, Evolution of Economic Governance in the European Monetary Union (Central Bank of Malta Working paper 03/2013). 41 For a full run-down of the Fiscal Compact ratification and related arguments see Malte Kroeger, Constitutional Change through Euro Crisis Law: Malta (Florence, EUI, 2013). 42 MTC Art 1(3).

520  Peter G Xuereb and Mireille M Caruana current form by the European Union Act, Act V of 2003, Article 7. It is in virtue of this Act that Malta acceded in due course to subsequent treaties, including the Lisbon Treaty in 2007. Presumably this formula can be applied for all future agreed Treaty amendments. However, it is clear that Article 65(1) cannot operate as a permanent commitment to perpetual membership of the Union for the reason that Article 65(1) can be amended by a majority of all the Members of the House under Article 66(5) of the same Constitution. It was presumably thought that the Constitution should contain some reference to the Union, inasmuch as joining the Union was a highly controversial national political issue, not least for the reason that a considerable transfer to the Union institutions of, and limitation on, legislative power for the future was involved. Yet ultimate sovereignty is clearly retained in the Constitution in the form of the maintenance of the alterability of this constitutional provision. The Constitution was amended by majority in Parliament, following a non-binding referendum in favour of membership in turn followed by a national election won on the basis of a mandate to take Malta into the Union. The Constitution was amended by Act No V of 2003, titled the European Union Act of 2003. In the first place, this provision encapsulates the transfer of legislative power to the Union effected by the signature of the Treaties and their ratification by Malta. The accepted analysis is that this transfer occurred with signature and then ratification by the passage and coming into effect of the European Union Act, in virtue of Article 65 of the Constitution. Second, this provision encapsulates the ‘supremacy’ of Union law (this by effect of paragraph 2 of Article 3 of the European Union Act). There have been no statements by the Maltese Constitutional Court itself. The authors’ view is that there has been, objectively speaking, a transfer pro tempore (for as long as Malta remains a Member State and the European Union Act remains in force) of legislative power to the institutions of the Union in the fields of competence of the Union and subject to the Treaty procedures for the exercise of that competence, and that Maltese constitutional theory must be regarded as accommodating that. However, precisely how this would be articulated by the Constitutional Court is an open question in the absence of any indications from it on this matter. It is arguable that there are limits at every point in time. It is further arguable that these would apply as a matter of interpretation of the Constitution at any point in time. One possible approach is to argue that the – or at least some – limits are to be found embedded in the wording of Article 65 of the Constitution, but in our opinion it cannot be assumed that unless therein contained then no (other) limits apply, on some interpretation of exclusiveness, for the possibility of such being held was not in our view envisaged at the time of the amendment of Article 65 to give it some semblance of a transfer of power clause. If those limits that are expressed are conditions, then they of course apply as conditions for the transfer. Moreover, it can be said that the Article 65 ‘limits’ are not particularly strict, and so are unlikely to be held to be the only limits. For example, although this clause does not refer to other elements such as respect for human dignity as such, the principle of democracy, the rule of law and the protection of fundamental rights, as is the case for example with the German Constitution,43

43 Grundgesetz, Art 79.3, which makes amendment of these essentials by amendment of the Basic Law ‘inadmissible’. The German model combines this internal provision with an external one that describes the foundations of the Union being joined, namely Art 23.1 Basic Law ‘Germany … shall participate in the development of the EU that is committed to democratic, social and federal principles, to the rule of law, and to the principle of subsidiarity, and that guarantees a level of protection of fundamental rights essentially comparable to that afforded by this Basic Law.’ Any change in the foundations of the Union is then made subject by Art 23.1.para 3 to Art 79 (2) and (3) of the Basic Law.

Malta  521 that which has been said above, including in regard to Article 1, must surely apply as a matter of interpretation of Article 65 itself. In this way, the good government of Malta must be read as government in accordance with the foundational principles of the state of Malta. Yet it needs to be said that these foundational principles are not identified in terms in the Constitution and it is a matter of interpretation of the Constitution both as to what they are and as to their scope.

2.  Independent State Having said that, it is most unlikely that the Constitutional Court would second-guess the ­determination of Parliament, except where the effect of the parliamentary resolution was to undermine the very fabric or foundation of the Maltese state and residual sovereignty itself. But these are possible limit(s) that do not emerge textually from Article 65, but might be invoked as part of an alternative approach to the interpretation of Article 65 which would read this Article as being subject to the fundamental elements of the Constitution. This would be because ultimate and entire sovereignty continues to reside in the Maltese state as constituted by the Constitution; as a state under international law, Malta, it could be argued, cannot cease to be such by virtue of any action taken by Parliament under Article 65 as it stands. However, there is a major caveat, since, in theory at least, there is the legal possibility of amendment of every constitutional provision, although the applicable procedure can vary and would follow one or other of three different modes, but with one of them available for each provision of the Constitution. Even Article 1, constitutive of the republican state of Malta, and Article 6 establishing the supremacy of the Constitution could (at least textually and theoretically) be amended by special majority in Parliament under Article 66(2). Article 1 of the Constitution provides that (1) Malta is a democratic republic founded on work and on respect for the fundamental rights and freedoms of the individual. (2) The territories of Malta consist of those territories comprised in Malta immediately before the appointed day, including the territorial waters thereof, or of such territories and waters as Parliament may from time to time by law determine. (3) Malta is a neutral state actively pursuing peace, security and social progress among all nations by adhering to a policy of non-alignment and refusing to participate in any military alliance. Such a status will, in particular, imply that: (a) no foreign military base will be permitted on Maltese territory; (b) no military facilities in Malta will be allowed to be used by any foreign forces except at the request of the Government of Malta, and only in the following cases: (i)

in the exercise of the inherent right of self-defence in the event of any armed violation of the area over which the Republic of Malta has sovereignty, or in pursuance of measures or actions decided by the Security Council of the United Nations; or (ii) whenever there exists a threat to the sovereignty, independence, neutrality, unity or territorial integrity of the Republic of Malta; (c) except as aforesaid, no other facilities in Malta will be allowed to be used in such manner or extent as will amount to the presence in Malta of a concentration of foreign forces; (d) except as aforesaid, no foreign military personnel will be allowed on Maltese territory, other than military personnel performing, or assisting in the performance of, civil works or ­activities, and other than a reasonable number of military technical personnel assisting in the defence of the Republic of Malta;

522  Peter G Xuereb and Mireille M Caruana (e) the shipyards of the Republic of Malta will be used for civil commercial purposes, but may also be used, within reasonable limits of time and quantity, for the repair of military vessels which have been put in a state of non-combat or for the construction of vessels; and in accordance with the principles of non-alignment the said shipyards will be denied to the military vessels of the two superpowers.

So that it seems clear that any major overhaul of the state of Malta as conceived by the Constitution – whether regarding the Article 1 elements of republican status, or the ‘work’ or ‘respect for fundamental rights’ or ‘neutrality’ or ‘foundations’ would at the very least require a two-thirds majority vote in Parliament, in some cases with a referendum result in favour, and one might even say that in practical political terms it is almost surely the case that full consensus bordering on unanimity among the citizens of Malta would be needed. On the other hand, it seems to follow from the alterability of the provision that at least some of the characteristics therein listed are in fact not of ‘any unalterable essence’. Indeed this appears to be the logical conclusion from the fact of alterability (though difficult in practice with a two-party system that pits the parties into permanent opposition). While it is hardly arguable that the framers of the Constitution envisaged a two-thirds majority deciding to abolish statehood, it is likely that it was considered quite possible that a time might come when the neutrality clause might come in for alteration or even abolition. It is a question whether the promised constitutional revision exercise (which has not as yet properly commenced) will in course of time propose a greater direct role for the people, or further entrench the several elements expressly set out in article 1 – as well as others alluded to above and protected in other European constitutions – such as by requiring unanimity in Parliament or a super-majority in Parliament combined with a high threshold pro-vote by popular referendum. The dynamics of the constitutional development of the Union itself – and of Malta with it – may well play a role in any debate on these lines. The same applies to neutrality as set out in Article 1.

3.  Respect for Certain Constitutional Features at EU Level The authors’ view is that further constitutional development of the European Union can and should be absorbable if this were also to include (in part through an elaboration of Article 4 TEU) further provision for solidarity vis-à-vis the essential identity, and essential ­interests, of each Member State. The possibility of wider transfer of powers, pace any necessary amendment of the Constitution, raises the question as to which provisions might require amendment. Certainly, in the unlikely event of the proposal for the creation of a federal European state, such a proposal would make the greatest demands. At the very least a debate would be needed as to the meaning of ‘state’, ‘sovereignty’, and ‘independence’ in Article 1(3)(b)(ii) of the Constitution. Any reference to ‘sovereignty’ would either need to disappear or be qualified in some new way. The theoretical question would be raised as to how one would frame any amendments, even if a right of withdrawal from Membership of the Union were retained, although it would be likely that this right and power would disappear in the case of full federalisation. Would there be a transfer of actual sovereignty (as opposed to just powers)? Would there be only a ‘temporary’ or­ pro tempore assignment of sovereign powers? Short of outright Union federalisation, we have recently witnessed the need for constitutional accommodation in most Member States to permit new Union powers or the unforeseen exercise of power at Union level. This is emerging as something of a trend. In the authors’ view – and this

Malta  523 view was expressed at the time of the future of Europe debate at the turn of the century, but resonates even more powerfully today – the final understanding and adoption of the principle of solidarity (enforcing the value claimed by the Treaty) is the key. With true and active solidarity even full integration can be achieved, but without it every step will prove tortuous and fraught with backtracking and avoidance. An active principle of procedural and substantive solidarity – not to be confused with a redistributive policy – would be based on a multi-level dialogic approach that applied – through the institutions – the practice of input and output legitimacy based on the taking into full account of the different national interests, weighing them against the common good at EU level, and then proceeding to balance the proportionate pursuit of the EU level common good by the safeguarding or accommodating of each national interest. This is a combined Solidarity44 and Common Goods45 approach to the future parallel constitutional evolution of the Treaties and of national constitutions. It is not only about respecting the essence of national constitutions. It goes beyond that to respect for legitimate national interests. This practice of real solidarity requires a constitutional underpinning at both Union and national level. If put in place, it would allow for the pursuit of ambitious aims at EU level while securing maximum agreement to the required action at EU level – precisely through the solidary respect for legitimate individual national interests. Legislative and other decision-making procedures would be adopted to provide for this. Mechanisms of review and challenge would ensure that objection could be made to the use of power otherwise than in accordance with this principle of solidarity. For example, the subsidiarity yellow card procedure should be extended to the exercise of the solidarity principle (and duty). If a certain number of Member States felt that a legislative proposal did not evince a reasonable degree of solidarity, the procedure would send the proposal back for further consideration at that early stage. In fact, other grounds of a constitutional nature might be added, so that the monitoring and review by national parliaments would be made to cover other constitutional issues such as proper observance of human rights. Such mechanisms would ensure early consultation and dialogue, and ultimately pre-empt the possibility or even certainty in some cases of ‘post-adoption challenge’ to EU measures not only before the Court of Justice but also before national supreme courts. In the latter case any challenge would have to have its basis in national law, meaning that matters would have come to a clash between Union legislation and national (possibly constitutional) law. Far better to recognise principles at EU level and ensure their application or enforcement at that level (with full dialogue between levels) than adopt legislation or other measures that give rise to chagrin at national level and legal challenges at that level than needs must be settled by reference to national law. Far more likely that the necessary accommodations be found through a process that ensures solidarity ex ante rather than one that relies on raw majoritarianism with the possibility of subsequent challenge on grounds that require of the Court something that a sensitive Commission and political process are far better able to deliver. It has been written elsewhere that ‘As recent events in the Union have shown, state sovereignty needs to be constantly re-evaluated. Otherwise national interest will be pursued not through cooperation, but through the power of the veto, and even when this obstructs the pursuit of the wider common good. This makes it a constitutional issue for all the Member States and the Union as a whole.’46

44 See Xuereb, ‘Solidarity and Constitutionalism’ (n 18). 45 See Peter G Xuereb, ‘A Common Goods’ Understanding of Europe and the World’, in L Bekemans (ed), A ValueDriven European Future (Brussels, Peter Lang, 2012) 159. 46 Ibid 169.

524  Peter G Xuereb and Mireille M Caruana

4.  Neutrality (as a Component of National Identity?) One can think also of the neutrality clause (Article 1) as one that might require amendment, if it were considered necessary that the Union were given wider powers to bind each Member State in the relevant context. It is almost certainly the case that this clause requires review and updating in any event, as a matter simply of having been superseded by world and regional Mediterranean events since its inception (although history has a habit of repeating itself and the conditions that made earlier action necessary), and a much-needed rethink of Malta’s foreign policy stance could lead – even without Treaty amendment – to a new configuration of Maltese active neutrality. But legal and political concepts of neutrality differ across Member States and across the wider international community. The time may come for rationalisation within the Union. Before that, Malta may engage with its own constitutional concept as part of an internal constitutional review process. As Xuereb has written elsewhere, it is arguable that the second clause of paragraph (e), in particular, is to be regarded as redundant in light of the ending of the Cold War (at least the first one, for we may be on the threshold of the second). It has also been argued that the entire provision centres around a policy of ‘non-alignment’ (as between ‘the superpowers’ – which may now be more than the ‘old’ two, at least in prospect), rather than strict neutrality, and that nonalignment is not a fixed concept in international law but must bear reference also to the climate of the time when the provision was adopted. This type of reasoning calls much into question, also in light of the evolution of the role and competence of the Union in the sphere of Foreign and Security Policy. One recalls the duty of loyalty under the Treaty but also the objectives of the Union especially post-Lisbon. Indeed, the Maltese provision speaks of an ‘active’ neutrality, one that is intended to actually permit Malta to be pro-active in the pursuit of peace, security and social progress among all nations. This does not indicate a passive stance. While participation in a military alliance is excluded (Malta was told in the course of membership negotiations that the EU was and was not to become such), the ban on the use of Maltese facilities by foreign military forces is open to a number of exceptions, one of which is the pursuance of measures or actions decided by the Security Council of the United Nations. This all shows that the matter is not clearcut, and that some think that an attempt at ‘clarification’ of Malta’s international active neutrality status should be made in any constitutional review exercise that may be undertaken if this status is to be generally understandable within the context of our membership of the EU and within the new global context. Therefore, the neutrality clause is one clause that has been identified as being in need of review/revision in light of developments relating to Malta’s EU role and to changes on the international scene. Nor is it clear that it is in any sense to be considered that the Maltese brand of neutrality is immutably part of any national identity. For one thing, it is far too recent to claim such status.

5.  A Formula for Mutual Loyalty? Many have argued for a multi-level dialogue process based on mutual loyalty and respect for autonomy and identity. It might be argued by some that this is what we currently have, but counter-argued that this is not unequivocally the case. Article 4 TEU seems constitution-­structure rather than constitution-values oriented. At the same time the Union is able to encroach in various ways – directly and indirectly – on national ‘culture’ – and the big question is: especially if greater powers are to be conferred on the Union, do national constitutions need to think about protecting (by advertising to the EU itself) their ‘no go’ areas – by reference to a clearer concept of identity, and culture. Another way of putting it is to ask whether a future Union Constitution

Malta  525 will see it as its task and duty to protect diversity of identity, or rather to tend towards diminishing (if not seeking directly to expunge) this by ‘creating’ a more clearly defined concept of European identity and especially a strong ‘common’ and universally binding interpretation of human and fundamental rights under the Charter. The Union might seek at least somehow to diminish the capacity of national identities to maintain their separateness and difference in the face of the exercise of greater powers by the Union. In this light, the authors are of the view that a full debate on the national values/identity implications of these possibilities (eg through total secularisation; neutrality on religious and philosophical concepts of moral goodness, or a convergence policy on family ethics) is well and truly indicated, at least at national level in the course of the promised constitutional review exercise in Malta. The debate as to ‘social Europe’ at the time of the referendums on the Treaty establishing a Constitution for Europe (the so-called and ill-fated ‘Constitutional Treaty’) was followed closely in Malta. The social dimension of the Treaties is important in Malta’s eyes. The accountability of the institutions, and of all decision-makers, including in the economic sphere, and of the new ‘European’ regulators, is a matter of real interest, which many in Malta trust will continue to be addressed at all levels and especially throughout the sphere of economic and financial governance. It would be well for the principles and liabilities to be expressly spelt out and for appropriate investigative mechanisms to be put in place, thus mirroring political accountability under the national constitutions. There is every reason to extend this, possibly by EU initiative, to non-EU, global financial and economic institutions. What seems clear to the authors is that for any future constitutional settlement to work across Europe, as long as societies were not homogeneous or in some way to be directed towards becoming so, we in Europe would need to have dialogue and work towards a concept of identity that could at the same time feature as the basis for a ‘future Article 4 TEU’ as well as the basis for the counterpart safeguard clauses in the Member State constitutions. Of course the latter might come first, being incorporated into the national Constitution as a safeguard measure also against any future unwarranted centralisation of power or the over-enthusiastic exercise of power at EU ‘level’.

B.  Judicial Review of EU Measures 1.  Access to Justice and the Standard of Review There are several heads under which there can be judicial review of Union law and of national law and administrative acts. Moreover, redress can be obtained through delegated legislation, including by action of the Prime Minister, a rather unusual procedure whereby the latter may make regulations deleting any instrument having the force of law from the statute book or remove any inconsistency between it and the Constitution or the European Convention Act of 1987. This after a court judgment declaring such to be the case. Similarly, ultra vires is another ground for action, and can apply to a law which runs counter to the Constitution, in accordance with Article 6 of the Constitution, as well as judicial review under the Code of Organisation and Civil Procedure or under Article 3 of the European Convention Act.47 Remedies can include a declaration of nullity, a prohibitory order, a mandatory order and in many cases, compensation. 47 This is a brief indication of the remedies and actions available. For a full review on this point see Kevin Aquilina, ‘Remedies under Maltese Law for Wrongful State Action in Breach of EU Law: A Case of Crying over Spilt Milk’, in E Buttigieg (ed), Enforcing One’s Rights under EU Law (La Valetta, Gutenberg Press, 2011) 88.

526  Peter G Xuereb and Mireille M Caruana With regard to indirect judicial review of Union acts via the preliminary reference procedure of Article 267 TFEU, there has been one case where a question was raised as to the validity of Regulation 530/200848 and of Article 7(2) of Regulation 2371/2002.49 The Court of Justice found partial invalidity in Regulation 530/2008.50 The authors share the view that the CJEU has demonstrated a propensity in the past to apply a liberal standard of review to EU measures. The national standard on the other hand requires a searching enquiry into the matter on the basis of general principles of law. Certainly, any review of an allegation of a breach of fundamental rights will apply a strict standard of compliance, and it may be that the latest Kadi judgment,51 among some others, brings the CJEU closer to the national ‘norm’. Importantly, it seems to indicate the possible emergence of a ‘new constitutionalism’ in the EU legal order, with human rights occupying a (more) central place in EU legal order discourse. At the same time, as has been noted by some, this seems to imply a (welcome) shift in focus from the market integration imperative to a broader constitutional, political and social underpinning for the rule of law in the EU legal order. The Constitutional Court has exercised the power under Article 6 of the Constitution to uphold the supremacy of the Constitution, including its human rights provisions, relatively vigorously. Yet it has been criticised52 for its stance that even where the European Court of Human Rights has ruled against the Government in regard to the application of a law found to be in breach of human rights, the Constitutional Court considers itself unable to affect the (legislative) position for the future by declaring the legislative act in question ‘null’, which would apparently be seen as ‘abrogating’ the act and as something that only Parliament is empowered to do. Of course, the Constitutional Court will declare anti-constitutional laws and acts to be ‘to the extent of the inconsistency, void’, in accordance with Article 6 of the Constitution. All apparently turns on a claimed distinction between the meaning and effect of the terms ‘null’ (not used in the constitutional provision) and ‘void’ (the term actually used). It is considered by the Constitutional Court to be a matter for Parliament to repeal the offending legislation. This could well be a vestige of the British constitutional influence, and may on this score provide a unique source of contrast with the position generally held in continental constitutional law. There is no Constitutional Court ruling in point on the question of the review of measures that implement EU legislation. The view is expressed that the Maltese Court would review the Maltese legislation/measures first and foremost under the Constitution, while interpreting the relevant EU legislation sought to be implemented according to EU norms, including the general principle of protection of human rights and the CFR, but would carry out the review primarily in the context of the Constitution and of the European Convention Act of 1987. It would be theoretically (and very hypothetically) open to an applicant to argue that the Court should apply a higher standard than was being argued was the EU standard of protection, but he would need to do so by reference to a clear opposing prescription in the Constitution. It is not suggested that the standard of review in Europe is considered to be necessarily lower than it is in Malta. The Maltese Courts will, in practice, review measures implementing 48 Commission Regulation (EC) No 530/2008 of 12 June 2008 establishing emergency measures as regards purse seiners fishing for bluefin tuna in the Atlantic Ocean, east of longitude 45 °W, and in the Mediterranean Sea, [2008] OJ L 155/9 f. 49 Council Regulation (EC) No 2371/2002 of 20 December 2002 on the conservation and sustainable exploitation of fisheries resources under the Common Fisheries Policy, [2002] OJ L 358/59–80. 50 CJEU 17.03.2011 C-221/09 (AJD Tuna) ECLI:EU:C:2011:153. 51 CJEU 18.07.2013 Joined Cases C-584/10 P, C-593/10 P and C-595/10 P (Commission and Others v Kadi) ECLI:EU:C:2013:518. 52 What follows is taken from Xuereb, ‘Constitution of Malta’ (n 1).

Malta  527 EU law for alleged breach of human rights. Also, it is probably true that the Maltese courts are likely to presume equivalence of protection, but will apply any higher standard as mandated by EU law and the Constitution itself. It is unclear what the courts would do if a lower standard of protection were indicated by EU law. This is hypothetical, as the ECHR provides a minimum standard in both spheres, and it has not been the subject of a statement by the Constitutional Court. Arguably, however, the Court would be obliged under the Constitution to apply the higher national standard according to the Constitution and the European Convention Act of 1987, and certainly up to the minimum standard of the ECHR. The broad issue of differential treatment has not been raised. In the authors’ view, it is highly unlikely that different standards would be applied in the absence of objective justification in a democratic society (if such were possible). The domestic protection would operate as the minimum standard at all events, as the human rights provisions apply to all persons present in Malta.

2.  Scrutiny of Secondary Legislation: Enforcing the Limits In any event, the constitutional case law of some Member States has clearly stated that it is ­ultimately for the national courts to review and decide, after any preliminary reference to the Court of Justice that may be needed, whether an EU act exceeds the limits to the surrender of sovereignty entailed in acts of accession or domestic ‘recognition’ Acts (such as the European Union Act). This is the case clearly for Germany, Denmark, France and Italy, for example. The Maltese Constitutional Court could, and would almost certainly, take the same line as these other supreme courts. It might even be suggested that the Maltese constitutional review exercise (once it begins in earnest) could go further and, indeed, make express constitutional provision for judicial review in this regard.53 In the end, the avoidance of conflict of constitutions depends on synchronising the various ‘levels’ or ‘spaces’. As has been written elsewhere the general opinion in Malta would be that it is vital for constitutionalism – in the sense of limited government – at national level to be intertwined with constitutionalism – in the same sense – at regional and global level. In other words, that these other levels must themselves be ‘appropriately’ constitutionalised, especially if they are to be further empowered. The role for the national constitution does not change, and the global constitutional picture must be reflected in the national constitution but this also applies in reverse. The real challenge then is to devise the regional and global constitutional compact under the form of constitutionalism and most crucially to build into the regional and global constitutionalism appropriate mechanisms of voice and solidarity (in both the procedural and substantive senses). At regional European level, the Lisbon Treaty (building on the Constitutional Treaty) quite clearly set out to do more of this, but a gap remains. We have a much larger gap at global level. At the same time, constitutional tolerance is said to be a key value. Power is increasingly exercised at supranational level because it is being transferred to that level or even simply taken at that level. It is up to States to balance the transfer of powers to that level with added constitutionalism and constitutional values operating at that level. In the author’s view, changes to national constitutions would best be designed in a context which involved the constitutionalisation of the exercise of power at the ‘higher’ level according to a multilevel – or as I prefer – a ‘horizontal’ constitutionalism model. Both levels or spheres should operate in synchrony according to the same core fundamental values and principles. New mechanisms and instruments need to be devised for the purpose.54

53 See on the Danish position, Helle Krunke, ‘Case note: The Danish Lisbon Judgment’, (2014) 10 European Constitutional Law Review 452, 550. 54 Xuereb, ‘Constitution of Malta’ (n 1) 149 f.

528  Peter G Xuereb and Mireille M Caruana These thoughts have been developed in principle as centring around a procedural and substantive principle of solidarity between the Union and the Member States and between the Member States themselves, a constitutional principle that may well serve to balance the future needs of the Union and of the Member States in the matters of identity and power.55 National constitutions need to (a) expound their essence clearly and provide for effective safeguards for that essence, whether by clarifying also the limitations on the capacity of the state to transfer power or other limits, and (b) provide for the effectiveness of international obligations in their territory otherwise; while the ‘wider’ constitutional level(s) – such as the EU constitutional space or level – need to (a) clearly set out the essence of the common objectives and relative powers and the limits on those powers, and (b) provide for the effective pursuit of those objectives while respecting the essence of the national level and observing the limits applying to their own level. The key synchronising principle is mutual respect and loyalty (expressed through the exercise of active and justiciable solidarity), and strict observance of the limits of the/each other’s spheres of competence. At each level, the highest courts would act as the gatekeepers, acting in dialogue with each other through a pre-emptive Opinion-giving jurisdiction or competence and/ or a preliminary ruling mechanism/jurisdiction.

V.  Constitutional Rules and/or Practice on Implementing EU Law A.  Binding (Parliamentary or Governmental) Resolutions for the Ministers in the Council? According to Standing Order 120F of the House of Representatives, the Standing Committee on Foreign and European Affairs (FEAC) has the following functions: (i) to deal with matters relating to foreign and European Union affairs which may be referred to it by the House or by the Standing Committee on House Business. (ii) in the context of European Union issues and measures to be taken by the Council of Ministers of the European Union, to scrutinise on its own initiative (a) any proposals under the Community treaties for legislation by the Council or the Council acting jointly with the European Parliament; (b) any document which is published for submission to the European Council, the Council or the European Central Bank; (c) any proposal for a common strategy, a joint action or a common position under Title V of the Treaty on European Union which is prepared for submission to the Council or to the European Council; (d) any proposal for a common position, framework, decision or a convention under Title VI of the Treaty on European Union which is prepared for submission to the Council; (e) any document (not falling within (b), (c) or (d) above) which is published by one Union institution and which does not relate exclusively to the consideration of any proposal for legislation; and (f) any other document relating to European Union matters placed on the Table of the House by any Minister; and (iii) to represent the House of Representatives of Malta at the Conference of European Affairs Committees (COSAC).

55 See

Xuereb, ‘Solidarity and Constitutionalism’ (n 18).

Malta  529 Therefore, in terms of section 5(ii) of Standing Order 120F, the FEAC is tasked with the­ scrutiny, on its own initiative, of legislative proposals and non-legislative communications ­originating from the European Institutions. The basis used for scrutiny is the Government’s ­position on the dossier, which is contained in the relevant Explanatory Memorandum. In order to carry out the task of EU scrutiny, a number of Working Groups are set up within the remit of the FEAC.

B.  Subsidiarity Scrutiny by Parliament As to subsidiarity scrutiny, the FEAC will debate any issue flagged by the secretariat and as ­necessary commission independent expert reports or hear interested parties. It may take into account the views of civil society as expressed before it or in another forum such as the Malta Council for Economic and Social Development (MCESD). So far, it can be said that as a practical matter, Parliament’s resources have not been adequate to the task of ongoing and in-depth scrutiny of all EU documentation, with the result that it has been difficult, though not impossible, for Parliament to be presented with totally original and fully researched reports such as may provide a totally different exercise to that carried out by government and communicated to Parliament for consideration by the FEAC. This has also meant that in the past, FEAC had come to a consideration of a measure somewhat late in the day and with only a few days to spare before the need for transmission of the Parliament’s reasoned opinion under the Protocol on the monitoring of the Application of the Principles of Subsidiarity and Proportionality (Protocol No 2 TEU). It follows that in practice the position of Parliament has tended in some cases at least to follow very closely the view expressed in the Explanatory Memorandum prepared by the Government and its own experts.

C.  Techniques of Implementing Secondary Legislation, Especially of Directives: Laws, Administrative Regulations, Others There has been an increase in, and possibly also a shift to, the use of ministerial power, through subsidiary legislation, for the implementation of EU directives. The result is one of less transparency, as well as less awareness, for all involved in the legislative process and afterwards. It may also be the overall outcome that in some areas the resulting complex of legislation lacks cohesiveness or even coherence. This was the case for example for a long time in the area of gender equality law, and this remains to some extent the case despite attempts to rationalise the various pieces of secondary legislation enacted via piecemeal ministerial regulations. To this may be allied the initial tendency, now thankfully diminished, to ‘copy/paste’ directives into national law. Again there are implications here for the democratic process, and ultimately the perception of EU legitimacy. Another issue may be the apparent willingness of the CJEU to defer the application of the principle of proportionality to national laws which come up for scrutiny as possibly infringing this principle of EU law, as well as fundamental Treaty law, to the national courts. The gaming sector is an example. The argument against this, at least in some circumstances, runs like this: if the rule of law means the uniform application in practice of fundamental treaty rights in all Member States, then to leave the ultimate determination of the proportionality of national measures (which allegedly unjustifiably restrict the exercise of those rights) to the national courts

530  Peter G Xuereb and Mireille M Caruana could militate seriously against the proper and uniform application of EU law, to the detriment of individuals seeking to exercise their rights under EU law.

D.  Enforcement Through the Courts, Including Protection of EU Charter of Fundamental Rights Although the number of preliminary references to the Court of Justice under Article 267 of the TFEU has been lower than initially anticipated, the Courts have demonstrated full awareness of the possibility, and, as it arises, the obligation to make a reference on a question of EU law.56 In one important case the reference resulted in the partial annulment of a Regulation by the Court of Justice.57 As to the Charter of Fundamental rights, the First Hall of the Civil Court has held that the Charter is to be applied to all Maltese legislation irrespective as to whether or not the law in issue was intended to implement EU law, holding that all Maltese legislation should be interpreted in conformity with the provisions of the Charter.58

VI.  Resulting Relationship between EU Law and National Law As the legal order of a dualist state, the Maltese legal order is able to resist the interpenetration of international norms. It is by virtue of the Constitution and laws made under it that EU law, regarded in principle as a separate system of law, is integrated into the Maltese legal order. However, EU law has been integrated strongly, first by virtue of the European Union Act, which makes EU law effective law in Malta on the terms of the latter legal order, and secondly by virtue of strong judgments of the Maltese courts, which can be said to be based on a strong interpretation in favour of application of EU law of the relevant Constitutional provisions and of the European Union Act itself. Article 6 of the Constitution, declaring the supremacy of the Constitution over other laws, is considered a hypothetical obstacle only to the full implementation of the Treaties and EU legislation, as well as the rulings of the Court of Justice. However, it remains a real one, for dualism is modified by Article 65 of the Constitution and the European Union Act, but it is not abandoned. The Treaties and the Charter will be given effect by virtue of Article 65 and the European Union Act, and all Maltese law will be interpreted in conformity with Malta’s international obligations wherever possible as a result, but in the event of a real conflict there is always the possibility that interpretative convergence will not be possible and in such a case Article 6 of the Constitution may come into play. By far the rule is loyalty towards the supremacy of EU law.59 As to human and fundamental rights, up until the Charter, the Court of Justice has appeared willing to respect the protection of afforded rights by even a minority of Member States, and

56 See generally, Eugène Buttigieg (ed), Enforcing One’s Rights under EU Law (La Valetta, Gutenberg Press, 2011). 57 CJEU C-221/09 (n 50). 58 Civil Court First Hall, Busuttil v Muscat (n 37). 59 For case law in Malta on this point see generally, Ivan Sammut, ‘Effectiveness of EU Law: The Maltese Experience’, in E Buttigieg (ed), Enforcing One’s Rights under EU Law (La Valetta, Gutenberg Press, 2011) 17, 28 f.

Malta  531 to give credence to different ‘scales of values’ also in the context of internal market legislation, and the relevant Treaty provisions relating to the exercise or limitation of Treaty freedoms. As long as the Charter does not substantially affect this approach of the Court of Justice, so that there is a large measure of constitutional fundamental rights tolerance from both sides, there is unlikely to be a problem in practice. However, it has been argued above that increasing EU activity in the area of development/family policy may well bring about something of a clash between the two legal orders. What may save this – but only to a degree – is in fact a new liberal tendency in Maltese politics, culture and law, as society adapts to what were in the recent past still seen as over-liberal approaches to concepts of family, marriage and same-sex unions, sexual identity and related issues such as adoption, divorce, contraception, abortion, euthanasia, and so on. In the development field, clashes have already occurred at the political level in the European Parliament. Any legislation proceeding from the EU which clashed diametrically with values that were textually or interpretationally capable of finding support in the Constitution would raise the issue of possible safeguard (a) first and foremost through the operation of safeguard clauses in the Treaties such as Article 4 paragraph 3 TEU (national identity), recourse to a margin of appreciation argument, exceptions in relevant Treaty articles (such as Article 36 TFEU) or recognised by the CJEU Justice in its case law, and other techniques available at EU level (for example, the subsidiarity ‘defence’), but (b) ultimately through the kind of ‘conditional supremacy of EU law’ approach we have seen from the German, Danish, Italian and French supreme courts among others. This is to say that it is perfectly possible that the Maltese Constitutional Court would, in what would be an exceptional case, rely ultimately on Article 6 of the Constitution to safeguard an essential constitutionally protected value. As has been said, the proper interpretation of Article 65 is that the Maltese Parliament may make laws, for as long as Malta is a Member State, only in conformity with its EU obligations. This is supported by the European Union Act, which it is suggested is properly regarded as an Act of Parliament with special quasi-constitutional status.60

References K Aquilina, ‘Remedies under Maltese Law for Wrongful State Action in Breach of EU Law: A Case of Crying over Spilt Milk’, in E Buttigieg (ed), Enforcing One’s Rights under EU Law (La Valetta, Gutenberg Press, 2011) 88. S Attard and A Demarco, Evolution of Economic Governance in the European Monetary Union (Central Bank of Malta Working paper 03/2013). E Buttigieg (ed), Enforcing One’s Rights under EU Law (La Valetta, Gutenberg Press, 2011). M Kroeger, Constitutional Change through Euro Crisis Law: Malta (Florence, EUI, 2013) http://eurocrisislaw. eui.eu/. H Krunke, ‘Case note: The Danish Lisbon Judgment’, (2014) 10 European Constitutional Law Review 452. I Sammut, ‘The EU and Maltese Legal Orders: What Kind of Marriage Between Them?’, in P Xuereb (ed), Malta in the European Union: Five Years On and Looking to the Future (Malta, European Documentation and Research Centre, University of Malta, 2009) 97. I Sammut, ‘Effectiveness of EU Law: The Maltese Experience’, in E Buttigieg (ed), Enforcing One’s Rights under EU Law (La Valetta, Gutenberg Press, 2011) 17. DP Scicluna, ‘EU Law – A National Judge’s Perspective’, in E Buttigieg (ed), Enforcing One’s Rights under EU Law (La Valetta, Gutenberg Press, 2011) 37.



60 See

also Scicluna, ‘Judge’s Perspective’ (n 26) 38.

532  Peter G Xuereb and Mireille M Caruana P Xuereb, ‘Constitutional Questions Raised by the Proposed Accession of Malta to the European Union in the General Context’, in A Kellerman, J de Zwaan and J Czuczai (eds), EU Enlargement: The Constitutional Impact at EU and National Level (The Hague, Asser Press, 2001) 229. P Xuereb, ‘The Future of Europe: Solidarity and Constitutionalism. Towards a Solidarity Model’, (2002) 27 European Law Review 643. P Xuereb, ‘The Maltese Constitutional and Legal Order and EU Membership’, in A Kellermann et al (eds), The Impact of EU Accession on the Legal Orders of New EU Member States and (Pre-) Candidate Countries. Hopes and Fears (The Hague, Asser Press, 2006) 409. P Xuereb, ‘A Common Goods’ Understanding of Europe and the World’, in L Bekemans (ed), A Value-Driven European Future (Brussels, Peter Lang, 2012) 159. P Xuereb, ‘The Constitution of Malta: Reflections on New Mechanisms for Synchrony of Values in Different Levels of Governance’, in A Albi and S Bardutzky (eds), The Role of National Constitutions in the European and Global Governance (The Hague, TMC Asser Press, 2019) 141.

19 The Netherlands MONICA CLAES

I.  Main Characteristics of the Constitutional System The current Constitution of the Netherlands is generally considered to date back to the one adopted in 1814, establishing a constitutional monarchy in the Netherlands, and its 1815 amendment, implementing the Congress of Vienna.1 The 1848 revision, adopted against the backdrop of revolutions in Europe, introduced a system of parliamentary democracy, and defines the constitutional system of the Netherlands to date. It was with that revision that Article 115 was adopted, stating that ‘statutes are inviolable’. In 1983 the text of the Constitution was modernised and simplified, and a bill of rights was introduced in the opening chapter. Article 115 was reformulated, turned into Article 120 and now provides that ‘the constitutionality of statutes and treaties shall not be reviewed by the judiciary’. The Dutch Constitution can be characterised as an open constitution: it does not set up a closed system. It is generally agreed that the broader constitution is made up of several components, of which the written document titled ‘Constitution for the Kingdom of the Netherlands’ is only a part, albeit the central part. In the Dutch language, a distinction can be made between the document containing the formal Constitution (grondwet (‘basic law’, hereinafter GW)) and the wider constitution. The latter is the category of fundamental rules and principles governing the relationship between the state organs, between the domestic and the international legal order and between the state and individuals (constitutie). This resembles the distinction between Grundgesetz and Verfassung. The constitutie includes, in addition to the grondwet, unwritten principles, constitutional conventions, some international treaties, most importantly human rights treaties, the Statute of the Kingdom,2 and certain organic laws.3 As will be demonstrated in detail below, the Constitution is also open to international law and the international legal order. The constitutional provisions on international relations and the effect of treaties are contained in section 2 (entitled ‘miscellaneous provisions’) of Chapter 5

1 The Republic of the United Netherlands (1581–1795) did not have a single written constitution in the modern sense of the word, although there were several constitutional documents and charters such as the Union of Utrecht (1579) and the Act of Abjuration (1581). 2 The Statute operates as the basic law of the Kingdom of the Netherlands, which includes the country in Europe (the Netherlands), as well as (currently) Aruba, Curaçao and Sint Maarten. It ranks higher than the Constitution. 3 These are normal Acts of Parliament and do not have a higher rank or stronger force of law. They are considered ‘organic’ because they concern the organisation of the state. Examples are the Act on Municipalities and the Act on the Provinces.

534  Monica Claes on ‘Legislation and Administration’. The section opens with a pledge that the Government shall promote the development of the international legal order, prescribes the procedure for the conclusion of treaties and the transfer of powers to international institutions, and governs the effect of written international law in the domestic legal order. These provisions are notoriously accommodating to international cooperation and international law. It furthermore contains provisions dealing with armed conflict and war. The text of the Constitution is sober and short. It contains no grand concepts such as sovereignty,4 no preamble, no invocatio dei, no Staatsziele. It does not mention the symbols of the state, the flag, the anthem or the motto, or any other elements that would be considered Verfassungskitsch. The main reason for this is that the Constitution is not revolutionary: it was not adopted to make a clear break with the past, sanctioning the sovereignty of the state (against a former oppressor), of the people (against the King) or establishing a new regime (post-authoritarian). It is rather an incremental, ‘long-term’ constitution, based on more or less incremental historical developments, aimed at codifying a state of affairs, rather than modifying the political system for the future.5 In 2006, a National Convention promoted the idea of a preamble to the Constitution, in order to give expression to some evident fundamental values of Dutch society, and to strengthen the Dutch national identity.6 The 2009–10 Constitutional Commission, however, did not see the added legal and political value of such preamble, and pointed to the fact that it was unlikely that consensus would be found on the content of such preamble, for instance on whether such preamble should contain an invocatio dei.7 This has to do with the fact that the Netherlands sees itself as a country of ‘minorities’ that keep each other in check. This consociationalism implies that Dutch politics and governance are characterised by a common striving for broad consensus on important issues, within both the political community and society as a whole.8 There has never been a real reason to definitively decide on a number of fundamental issues, such as the ultimate source of power. The 2009–10 Constitutional Commission did recommend the adoption of a general clause that would express the most fundamental features of the Constitution – these have never been written in the constitutional text before. Namely, that the Netherlands is a democratic state under the rule of law, that the Government respects and protects human dignity, fundamental rights and fundamental principles, and that public authority can be exercised only pursuant to law. This part of the advice of the Committee was picked up by the Upper House, and a bill adding a general preliminary clause to the text of the Constitution was introduced in Parliament. Yet, the text proposed reads very

4 On the (absence of) the concept of sovereignty in the Dutch constitutional discourse on European integration, see Bruno De Witte, ‘Do Not Mention the Word: Sovereignty in Two Europhile Countries: Belgium and The Netherlands’, in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 359; Leonard Besselink, ‘An Open Constitution and European Integration: The Kingdom of the Netherlands’, (1996) 44 SEW, Tijdschrift voor Europees en economisch recht 192; see more recently Jan Willem van Rossem, Soevereiniteit en pluralisme [Sovereignty and Pluralism] (Alphen aan den Rijn, Kluwer, 2014), ch 3. 5 Leonard Besselink, Constitutional Law of the Netherlands (Nijmegen, Ars Aequi Dutch Law, 2004) 18. 6 The National Convention was a temporary advisory body, tasked to advise the Government on institutional change in order to restore public trust in institutions, and to prepare a concept for a 21st-century Constitution; Nationale Conventie, Hart voor de publieke zaak. Aanbevelingen van de Nationale Conventie voor de 21e eeuw [Heart for the Public Cause. Recommendations from the National Convention for the 21st Century], September 2006, 42 f. 7 A Constitutional Commission is a temporary advisory body, mandated to advise the Government on constitutional change. This particular committee was asked to advise on several issues, including the introduction of a preamble and a new procedure for the approval of European treaties. 8 The term was coined by Arend Lijphart, The Politics of Accommodation: Pluralism and Democracy in the Netherlands (Berkeley CA, University of California Press, 1968).

The Netherlands  535 differently, stating merely that ‘The Constitution guarantees fundamental rights and democracy under the rule of law.’9 The Netherlands is a parliamentary democracy with a bicameral Parliament. It has been described as a consociational state and has a long tradition of coalition governments. The country uses a perfect system of proportional representation, without thresholds and without separate constituencies. This has resulted in a multi-party system with coalition governments: no single party has ever had a majority in Parliament since 1900. The formation of a coalition government of often three parties is always necessary. This is a tedious and long process that is regulated by tradition and convention, with only the final appointment process specified by law. One of the most striking provisions of the Dutch Constitution is Article 120, which precludes the courts from reviewing the constitutionality of statutes and treaties. There is no constitutional court. While courts do have the competence under Article 94 of the Constitution to review all Dutch legislation in light of directly effective provisions of international treaties,10 they generally tend to be deferent to the legislature and the executive. The courts have interpreted the constitutional ban as including also a prohibition to review statutes and treaties in the light of general principles and in the light of the Statute of the Kingdom of the Netherlands.11 The prohibition of constitutional review is deeply ingrained in the minds of Dutch judges and lawyers. Any interference in legislative business is considered a violation of the principles of separation of powers and of parliamentary supremacy and democracy.12 Courts are not supposed to cross the divide between law and politics, and they must refrain from ‘judicial law-making’. The courts will not order the state to act when it is the legislature who is required to act.13 This remains the evergreen of Dutch constitutional law. Nevertheless, in 2002 a bill was tabled to amend Article 120 of the Constitution and allow all judges to review statutes in the light of certain fundamental rights contained in Chapter 1 of the Constitution.14 The bill passed both Houses of Parliament in the first reading, but was never tabled for a second reading, which requires a two-thirds majority for the act to be adopted.

9 ‘De Grondwet waarborgt de grondrechten en de democratische rechtsstaat.’ The Bill was introduced in the Lower House on 9 July 2016, TK [Tweede Kamer (Second Chamber) Parliamentary Documents] 34 516, Nos 1–3; it was passed in both Houses and awaits approval by a two-thirds majority in both Houses in the next Parliament. 10 Leonard Besselink, ‘Constitutional Adjudication in the Era of Globalization: The Netherlands in Comparative Perspective’, (2012) 18 European Public Law 231. 11 Hoge Raad [Dutch Supreme Court, hereinafter HRNL] 14.04.1989, 13 822 (Harmonisatiewet) ECLI:NL:HR:1989:AD5725. The Statute operates as the basic law of the Kingdom of the Netherlands. It is higher in rank than the Constitution, and does not contain a provision excluding judicial review. Nevertheless, the HRNL took the view that the Constitution excluded any possibility of reviewing legislation in the light of any national higher rule whatsoever. 12 In recent years, several decisions of administrative and civil law courts have had a great impact on policy-making. Take for instance the Urgenda case, which has become one of the leading cases worldwide on climate change, acclaimed by some as an example of judges taking responsibility where politicians don’t. It is remarkable that the case was decided in a country that stands out for the absence of constitutional review, and a dislike of judicial intervention in politics. Yet, this was also the main criticism of public lawyers commenting on the decision, that the courts had overstepped the boundaries of the judicial function. Remarkably, private lawyers were much less critical, as the case is based on a tort action. See eg Rob van Gestel, ‘Urgenda: een typisch gevalletje rechter, wetgever of politiek?’, (2015) 30 (5) RegelMaat, 384–96. 13 See Geerten Boogaard, Het Wetgevingsbevel. Over constitutionele verhoudingen en manieren om de wetgever tot regelgeving aan te zetten [Court Orders to Legislate. On Constitutional Relations and Ways to Entice the Legislature to Act] (Tilburg, Wolf Legal Publishers, 2013); and Jerfi Uzman, Constitutionele remedies bij schending van grondrechten – Over effectieve rechtsbescherming, rechterlijk abstineren en de dialoog tussen rechter en wetgever [Constitutional Remedies for Fundamental Rights Violations – on Effective Judicial Protection, Judicial Inaction and the Dialogue Between Judge and Legislature], (Alphen aan den Rijn, Kluwer, 2013). 14 Maurice Adams and Gerhard van der Schyff, ‘Constitutional Review by the Judiciary in the Netherlands. A Matter of Politics, Democracy or Compensating Strategy?’, (2006) 66 Zeitschrift für ausländisches öffentliches Recht und Völkerrecht 399.

536  Monica Claes Upon the advice of the Council of State, the bill was withdrawn in 2018.15 Yet, in that same year, a Constitutional Commission advised the introduction of a constitutional court.16 In April 2021, the Lower House adopted a motion urging the Government to delete Article 120 from the Constitution and design a system of constitutional review.17 The Constitution contains no ‘core’ that is labelled as untouchable, and the entire document therefore seems open for amendment. A constitutional amendment is difficult and requires the passing of a bill in both Houses, elections, and the approval by a two-thirds majority in both newly elected Houses. There has been little appetite for constitutional amendment over the past decades, and consecutive governments and parliaments have rejected proposals for constitutional amendment which they considered ‘not indispensable’ or ‘not sufficiently mature’.18 Several amendment proposals are currently pending, including the introduction of a general provision, the strengthening of the right to a fair trial and the modernisation of the right to private communication. There are also no clear constitutional limits as to what powers can be transferred to international organisations. The Constitution is only to a limited extent considered to express the ‘national identity’ of the state, the political community or the people of the Netherlands. The values it protects are, on the whole, values that the Netherlands shares with other nations.

II.  Constitutional Culture The community of constitutional lawyers is rather small, and constitutional lawyers mostly take a rather pragmatic attitude to questions of the origins of authority, or other issues that bother their colleagues in other Member States. There is a low level of sophistication of the Verfassungsrechtslehre and Dogmatik when compared to countries like Germany or Italy. In Dutch constitutional culture, the Constitution and constitutional law play only a limited role in politics and in the public debate.19 In this respect, the Constitution is not ‘a living document’. The Constitution is taken for granted. While people think highly of it, they do not consider it to play a role in their daily or professional lives. This is the case not only for ordinary citizens, but even for those for whom the document would seem to be of immediate relevance, such as members of Parliament and judges.20 Issues of constitutionality of legislation are downplayed, while the compatibility with fundamental principles is often phrased in terms of compatibility with human rights treaties, such as the European Convention on Human Rights (ECHR), rather than of their constitutionality. The legal and political authority of the Constitution is, thus, overshadowed by European and international (human rights) law (mainly ECHR and European Union (EU) law), which operates as a substitute constitution. Even the Council of State, the Government’s prime advisor on 15 Kamerstukken II 2018–2019, 32 334, no 12, 1 [Chamber Documents, www.officielebekendmakingen.nl]. 16 Lage drempels, hoge dijken. Democratie en rechtsstaat in balans. Eindrapport van de staatscommissie parlementair stelsel [Low Thresholds, High Dikes. Democracy and the Rule of Law in Balance. Final report of the State Committee on the Parliamentary System], 2018, ch 6. 17 Kamerstukken II 2020-202, 28 362, no 47. 18 See eg Kamerstukken I 2012/13, 31 570, G, 2; Kamerstukken II, 2011/12, 31 570. 19 See eg Barbara Oomen. ‘Constitutioneel bewustzijn in Nederland: Van burgerschap, burgerzin en de onzichtbare grondwet’ [‘Constitutional Awareness in the Netherlands: on Citizenship, Civic Responsibility and the Invisible Constitution’], (2009) 3 Recht der Werkelijkheid 55; Barbara Oomen and Herman Lelieveldt, ‘Onbekend maar niet onbemind: Wat weten en vinden Nederlanders van de grondwet?’ [‘Unknown but not Unliked: what Do the Dutch Know About the Constitution and what Do They Think of it?’], (2008) Nederlands Juristenblad 577. 20 Janneke Gerards, ‘The Irrelevance of the Netherlands Constitution, and the Impossibility of Changing It’, (2016) 77 Revue interdisciplinaire d’études juridiques 207.

The Netherlands  537 the legality of legislation, does not place the Constitution and constitutional law centrally, but rather looks to the ECHR and other international treaties to draw the legal lines that should not be crossed.21 The limits of government and legislative action are found in treaties, which were paradoxically only meant as a subsidiary European minimum level, below which no state should go, rather than in the Constitution. This silence about the Constitution and its underlying values is generally seen as desirable and good: the Constitution reflects the constitutional settlements of the past,22 political compromises that should not be too easily challenged. Silence on constitutional matters is seen as a sign of constitutional peace and harmony.23 This is reinforced by an insecure attitude relating to constitutional interpretation and contestation. The Constitution says what it says, no more and no less, and constitutional change should be brought about by constitutional amendment, often piecemeal, ‘to keep the Constitution up to date’. Neither the legislature, nor the Council of State, the Government or the courts see it as their role to develop a ‘living constitution’, to take a lead in adapting it to the changes of time through interpretation, or indeed, to present themselves as the ultimate interpreter of the Constitution. The absence of a constitutional court and the political and pragmatic nature of constitutional review by the Advisory Council of State do not contribute to a sophisticated discours. That is, it does not develop a Dogmatik comparable to for instance the German, Italian or Spanish debates, which are heavily affected by the close relationship between the academic community and the constitutional court.24 At the same time, and perhaps unexpectedly, many scholars and practitioners do feel that the Constitution should be revitalised, not only because the Constitution is seen to have little symbolic value, but also because of its perceived lack of legal and normative significance. Several initiatives have been taken to improve the normative value of the Constitution. In 1999, a Constitutional Commission was instituted to investigate the need to reinvigorate the fundamental rights Chapter in light of technological developments.25 In 2005, a National Convention was convened to propose improvements to the constitutional and institutional set-up of the Netherlands.26 The 2009–10 Constitutional Commission was tasked to advise the Government on the feasibility and desirability of a number of fundamental changes, including the need for an 21 Wim Voermans et al, Juridische betekenis en reikwijdte van het begrip ‘rechtsstaat’ in de legisprudentie & jurisprudentie van de Raad van State [Legal Meaning and Scope of the Notion of ‘rechtsstaat’ in the Legisprudence and Jurisprudence of the Council of State] (The Hague, Raad van State, 2011). 22 In his speech at the occasion of the presentation of the report of the Royal Committee on the Constitution, Piet Hein Donner, the former vice president of the Council of State, then Minister for Justice, labelled the constitution as gestold verleden, ‘history solidified’, and expressed the view that the constitution mainly settles political conflicts of the past. It offers a legal framework, rather than an inspiration for the government or even less so, for society, see https://archief06.archiefweb. eu/archives/archiefweb/20180308082951/; www.rijksoverheid.nl/onderwerpen/grondwet-en-statuut/documenten/ toespraken/2010/11/11/toespraak-minister-donner-bij-in-ontvangst-nemen-advies-staatscommissie-grondwet. 23 Tijn Kortmann, a leading scholar in constitutional law, considered the lack of discussion on the Constitution a sign of its strength: ‘a constitution that is not “living” is not necessarily dead as a doornail’, but may well be well accepted in the hearts and minds of politicians, civil servants and the general public. On the latter point: he did not consider it a loss that the constitution was not known among the general public as it was generally of very little avail to them, see Constantinus AJM Kortmann, ‘Wegwerprecht, oude dame of frisse juf?’ [‘Disposable Law, Old Lady or Young Damsel?’], in De Grondwet herzien. 25 jaar later. 1983-2008 (The Hague, Directie Constitutionele Zaken en Wetgeving, 2008) 7, http://deatheensedemocratie.weebly.com/uploads/5/3/1/8/5318482/grondwet.pdf. 24 The contrast with the German debate is striking, see eg Christoph Möllers, ‘Legalität, Legitimität und Legitimation des Bundesverfassungsgerichts’, in C Möllers et al (eds), Das entgrenzte Gericht. Eine kritische Bilanz nach sechzig Jahren Bundesverfassungsgericht (Berlin, Suhrkamp, 2011), 283. 25 Commissie Grondrechten in het digitale tijdperk [Commission for Fundamental Rights in the Digital Era]; report published in Kamerstukken II 2000/01, 27460, no 1, annex 1. 26 Hart voor de publieke zaak. Aanbevelingen van de Nationale Conventie voor de 21e eeuw [Caring for the Public Cause: Recommendations of the National Convention for the 21st Century], September 2006, 42 f.

538  Monica Claes overhaul of the fundamental rights Chapter and the effect of international law in Dutch law. Most recently, the 2015–18 Constitutional Commission was asked to look into the parliamentary and democratic system of the Netherlands.27 Yet, these initiatives have neither led to a major overhaul of the Constitution, nor increased its importance in practice.

III.  Constitutional Foundations of EU Membership and Closely Related Instruments A.  The Dutch Constitution and International Treaties One of the most striking features of the Dutch Constitution is its openness to the international legal order.28 Since 1953 the section on foreign relations opens with the constitutional duty of the Government to promote the development of the international rule of law (now Article 90 GW). Legislative, executive and judicial powers can be attributed to international institutions (Article 92 GW). Provisions of treaties and decisions of international organisations are binding on individuals insofar as they have been duly published (Article 93 GW). National law – including Acts of Parliament and subordinate legislation, and even the Constitution – is inapplicable if its application would be incompatible with provisions of international treaties and decisions of international organisations that are ‘binding on anyone’ (Article 94 GW).29 In light of their traditional place in the constitutional framework and their deference to the legislature, the courts initially shied away from actually conducting such review and setting aside conflicting provisions of national law. Instead, they attempted to interpret national law in accordance with such provisions, in order to make any conflict disappear. In the 1980s, the courts gradually gained confidence, in part as a response to the case law of the European Court of Justice (ECJ/CJEU) and the European Court of Human Rights.30 Nevertheless, even today, the courts remain well aware of the limits of the judicial function, and shy away from judicial law-making. They are reluctant to set primary legislation aside, and when the mere setting aside of an Act of Parliament for infringement of a treaty does not suffice to solve the issue before them, they will defer to the legislature, and wait for legislative intervention. When the legislature, however, persists in inaction, the courts may become more active.31 27 Lage drempels (n 16). 28 See generally among many others Jan Brouwer, Treaty law and practice in The Netherlands, (Groningen, 2002); André Nollkaemper, ‘The Application of Treaties in the Netherlands’, in D Schloss (ed), The Role of Domestic Courts in Treaty Enforcement: A Comparative Study (Cambridge, CUP, 2010) 326; Leonard Besselink, ‘The Kingdom of the Netherlands’, in L Besselink et al (eds), Constitutional law of the EU member states (Alphen aan den Rijn, Kluwer 2014) 1187; Joseph Fleuren, ‘Recent developments regarding the direct and indirect application of treaties by Dutch courts. Fresh approaches to self-executing, non-self-executing and non-binding international law’, (2016) 47 Netherlands Yearbook of International Law 377. 29 From the legal history of the 1956 amendment it is clear that the phrase ‘binding on any one’ is meant to be a description of the concept of a self-executing or directly effective provision. 30 Johanna de Wit, Artikel 94 toegepast. Een onderzoek naar de betekenis, de bedoeling en de toepassing van ‘vinden geen toepassing’ in Artikel 94 van de Grondwet [Applying Article 94 of the Constitution. A Study into the Meaning, the Intention and the Application of the Words ‘Shall not be Applicable’ in Article 94 of the Constitution] (The Hague, Boom Juridische Uitgevers, 2012). 31 An example is the Dutch reaction to the Salduz case law of the Strasbourg Court. Initially, the HRNL considered the case law too imprecise for the Dutch courts to interfere (HRNL 30.06.2009, 08/02411 J [Salduz] ECLI:NL:HR:2009:BH3079). When the legislature had still not acted after eight years, while the EU had adopted the so-called Salduz Directive (Directive 2013/48/EU of the European Parliament and of the Council of 22 October 2013 on the right of access to a lawyer in criminal proceedings and in European arrest warrant proceedings, and on the right to have a third party informed upon

The Netherlands  539 Given the ban on constitutional review and review of the compatibility with directly effective provisions of treaty provisions, international treaties are counted among sources of Netherlands constitutional law, while guidance on constitutional issues is often sought in Strasbourg and Luxembourg.

B.  The Dutch Constitution and Membership in the EU The Constitution has never been amended in light of EU membership specifically. In fact, the Constitution does not even mention Europe or EU membership. Where relevant, if at all,32 the effect of European law in the Dutch legal order is considered to be governed by the provisions of the Constitution that apply to international law in general.33 Thus, Article 90 of the Constitution imposes an obligation on the Government to contribute to the development of the international legal order, an obligation that supposedly applies also for the development of the European legal order. Powers can be transferred to the EU, as they can be transferred to other international institutions and under the same conditions (see below). Conversely, the rules and principles governing the constitutional relations between the domestic institutions continue to apply in the context of EU decision-making. It could seem, thus, on the basis of the current text of the Constitution and its interpretation, that the constitution-maker has simply overlooked Europe. That view would, however, be mistaken. In fact, the Dutch Constitution was prepared for the process of European integration.34 In the 1950s, three Constitutional Commissions35 were asked to recommend reforms of the constitutional provisions concerning foreign affairs, mainly concerning Parliament-Government relations. The work of these committees was heavily influenced by developments on the ground: the conclusion of the ECSC Treaty, the establishment of the Council of Europe, the plans for and failure of the European Defence Community and the European Political Community. The provisions that were inserted in the Constitution in 1953 and 1956 on foreign affairs, the transfer of competences to international institutions and the direct effect and primacy of treaty law were inspired entre autres by the recently signed ECSC Treaty and the plans for further European integration.36 Nevertheless, these provisions do not mention Europe specifically, and concern international organisations in general. Indeed, the qualitative difference between European integration and international cooperation was not as clear then as it is now. In addition to increasing the role of deprivation of liberty and to communicate with third persons and with consular authorities while deprived of liberty, [2013] OJ L 294/1), the HRNL announced that it might become less deferential with time (HRNL 01.04.2014, 11/03714 [Salduz complaint] ECLI:NL:HR:2014:770). When a year later the legislature still had not acted, the HRNL held that from 1 March 2016 onwards, custodial legal assistance had to be available; HRNL 22.12.2015, 14/01680 (post-Salduz) ECLI:NL:HR:2015:3608. 32 As explained below, the tendency in Dutch legal thinking still is that powers can be transferred to the EU even if the Constitution does not explicitly provide for it, and that European law takes effect with precedence over national law in the Dutch legal order, because of the very nature of European law, because the ECJ has so stated it in ECJ 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66; ECJ 07.03.1985 32/84 (Van Gend & Loos v Inspecteur der Invoerrechten en Accijnzen) ECLI:EU:C:1985:104. 33 Arts 90–94 GW. 34 Karin van Leeuwen, ‘On Democratic Concerns and Legal Traditions: The Dutch 1953 and 1956 Constitutional Reforms “Towards” Europe’, (2012) 21 Contemporary European History 357; Jieskje Hollander, Constitutionalising Europe. Dutch Reactions to an Incoming Tide (Groningen, Europa Law Publishing, 2013). 35 The van Eysinga Commission, the van Schaik Commission and the Kranenburg Commission. 36 On the early history, see Monica Claes and Bruno de Witte, ‘Report on the Netherlands’, in AM Slaughter, A Stone Sweet and JHH Weiler (eds), The European Courts and National Courts. Doctrine and Jurisprudence (Oxford, Hart, 1997) 171; and more extensively Hollander, Constitutionalising Europe (n 34).

540  Monica Claes Parliament in foreign affairs, the constitutional amendments explicitly provided for the transfer of competences to international bodies and presented new provisions on the interaction between the Dutch and international legal orders. The post-war constitutions and constitutional amendments prepared in the other founding Member States largely went in the same direction, but only the Dutch attached so much importance and authority to international law. These provisions have not been revised since their 1953/56 revisions, and their starting points have never been reconsidered. On the occasion of the large constitutional overhaul of 1983–87, the provisions on international relations were considered apt for full compliance with the ­obligations resulting from participation in the EU, and no revision was undertaken. Similarly, the Constitution was not adapted on the occasion of the Maastricht Treaty – when the German and French constitutions were modernised – nor more recently, when several Nordic countries chose to update their constitutional provisions on Europe. Asked to look into constitutionalising membership of the EU, the 2009–10 Constitutional Commission did not propose a Europe provision or Europe Chapter. It did, however, propose to add the words ‘and European’ in Article 90 of the Constitution, which would read: ‘the [Government] should promote the development of the European and international legal order’. According to the majority of the Commission, such a change would be more important than a mere confirmation that the country is a member of the EU. In addition, as will be explained in more detail in the following section, the Commission suggested that amending European Treaties should more readily be submitted to the special majority procedure of Article 91(3) of the Constitution, given their substantive impact on the constitutional arrangements and the functioning of the state organs. These proposals were not picked up, and the Dutch Constitution to this day remains one of the very few not to mention EU membership at all in its text.

C.  EU Membership: Procedure for Approval and Ratification Article 92 of the Constitution declares that legislative, executive and judicial powers may be conferred on international institutions by or pursuant to a treaty. The provision is considered to be merely declaratory. Even without this provision in the Constitution, such transfer would not be forbidden under the Constitution. Transfers, thus, require approval by simple majority in both Houses. If, however, a transfer should conflict with the Constitution or would lead to such a conflict, approval of such transfer requires a two-thirds majority in both Houses of Parliament (Article 91(3) of the Constitution).37 So, the fact that a treaty tabled for approval conflicts with the Constitution does not mean that it cannot be approved and ratified. It merely means that approval requires a two-thirds majority in both Houses. This comes close to what is required to amend the Constitution, but it is less demanding, as it does not require approval by two consecutive parliaments and elections in between. Moreover, the approval with a two-thirds majority does not amount to an amendment of the Constitution. Since treaties take precedence over the Constitution anyway, such amendment would not be necessary. The procedure has never been followed for the approval of the EU Treaties, which were, thus, never considered to conflict with the Constitution.38 The provision has been interpreted 37 The decision as to whether the special procedure must be followed is made by a simple majority in both houses. 38 The Council of State did advise the Government to seek approval for the TECE with a two-thirds majority in favour, but the Government did not want to endanger a swift approval and did not want to set a precedent for the approval of future European treaties. See Hollander, Constitutionalising Europe (n 34) ch 2. The only European treaty that was approved with a two-thirds majority in application of (what is now) section 91, para 3 of the Constitution, was the

The Netherlands  541 as requiring a special majority only in cases of direct conflict between a provision of the Treaty and a specific provision of the Constitution. In other words, a tension with the ‘spirit’ of the Constitution or principles and values underlying it is not considered sufficient to require special treatment. There was some discussion when the Treaty of Maastricht was tabled for approval, but the Government insisted that there was no tension between the Treaty and the Constitution and proceeded to ask Parliament for its approval in accordance with the ordinary procedure, ie by way of an ordinary Act of Parliament.39 When the Treaty establishing a Constitution for Europe was up for approval and several Member States decided to hold a referendum on it for legal or political reasons, the Dutch government followed suit, at the request of three members of parliament. The Constitution does not provide for referendums, but it does not forbid them either, as long as they are merely consultative.40 Between 2002 and 2005 there existed a Temporary Referendum Act containing general rules for referenda, but national referendums had not been held in the Netherlands for nearly 200 years. There have been a large number of small-scale local referenda over the years, but a national referendum is a phenomenon that was alien to the Dutch constitutional system throughout the twentieth century and the larger part of the nineteenth century. There was, thus, no experience with the instrument when the referendum on the Constitutional Treaty was held. Asked for advice on such referendum, the Council of State stated –in a fairly sober and balanced opinion – that given the character of the Treaty, its ratification could be compared, to a certain extent, to an amendment of the Constitution and a referendum would offer a realistic means to involve voters. After the Dutch people voted the Constitutional Treaty down and the Lisbon Treaty was up for ratification, that opinion would haunt the Council. The government consulted the Council again, asking whether approval of the Lisbon Treaty would require a referendum, given that it largely resembled the Constitutional Treaty. This time, the Council of State opined that the Lisbon Treaty, in view of its contents, method and ambition, fitted in the ongoing constitutional process and that the changes made after the referendum (abandoning the concept of one written constitutional document, the deletion of the Charter from the body of the Treaty, the clearer demarcation of the competences of the Union and the omission of the symbols of European unification) had to be seen in unison, and justified the choice for a ‘normal’ approval by way of an ordinary Act of Parliament and for not holding a referendum.41 The Lisbon Treaty was accordingly approved with a simple majority in both Houses, and no referendum was held. The idea has been floated to make further steps in the process of European integration dependent on a stricter procedure and enhanced democratic legitimacy. One such proposal unsuccessful Treaty establishing the EDC in 1954. Outside the context of the EU, the provision was applied only twice: concerning the Act approving the Agreement Concerning West New Guinea (Stb 1962, 363), and concerning the Act approving the Statute of the International Criminal Court (Stb 2001, 343). The ESM Treaty and the TSCG were also approved by simple majority. Note, however, that the Act of approval does not indicate which procedure was followed, and that in practice, all European Treaties were passed with very large majorities, usually at least two-thirds. 39 Aalt Heringa, ‘De verdragen van Maastricht in strijd met de Grondwet. Goedkeuring met twee derde meerderheid?’ [‘The Maastricht Treaties are in Conflict with the Constitution. Approval by Two-thirds Majority?’], (1992) Nederlands Juristenblad 749. Heringa pointed to the provisions on EMU, which he considered to be at odds with Art 106 GW which provided (and still does): ‘The monetary system shall be regulated by Act of Parliament and the provisions on visa policy and electoral rights for EU citizens. In all cases, the Constitution endows a task to the Dutch legislature.’ Most scholars did not agree with Heringa’s views, and neither did the Government. More recently, questions have been raised about whether the extension of EU economic governance can be reconciled with Art 105 GW, which allocates budgetary power to the Government and Parliament acting jointly. 40 A binding referendum would be unconstitutional, since the power to create Acts of Parliament – and therefore the power to approve treaties – is vested explicitly in the legislature: Government and Parliament jointly, in accordance the procedure laid down in Arts 81–88 GW. 41 Advisory Opinion of the Council of State, 15 February 2008, W01.08.0004/I/K, Kamerstukken II 2007/08, 31 384 (R 1850), No 4.

542  Monica Claes aims at amending the Constitution to require a two-thirds majority for the approval of all EU Treaties (amending treaties and accession treaties) as well as decisions to proceed to simplified revision.42 It has been approved in a first reading by both Houses and awaits approval with a twothirds majority after the next elections. Both the 2009-10 Constitutional Commission Thomassen and the 2016-18 Constitutional Commission Remkes considered the proposal overly inclusive (applying to all EU Treaties) and advised instead to apply the procedure of Article 91(3) of the Constitution more leniently, and to require a two-thirds majority for the approval of all treaties that have significant constitutional implications. In the wake of the Brexit referendum, Burgerforum EU, a private initiative, managed to collect 300,000 signatures to have a referendum on the Act approving the Association Agreement with Ukraine under the new Consultative Referendum Act 2014.43 An attempt by a citizen to ask the Council of State for an injunction against the referendum was declared inadmissible for lack of personal interest.44 In the referendum, 61 per cent of voters voted against the Act, with a turnout of 32 per cent.45 The result was not binding. However, in accordance with the Consultative Referendum Act, Parliament and Government were under an obligation to ‘reconsider’ the Act ‘at the earliest convenience’. When the Act of Approval was not immediately withdrawn, one of the initiators of the referendum and leader of a new Eurosceptic political party, Thierry Baudet, took legal action seeking the court to order the Government to act instantly and withdraw the Act of Approval. The court rejected that claim, holding that it was for the political branches to decide on the consequences of the negative outcome of the referendum.46 Instead, the Government secured a Decision of the Heads of State or Government of the 28 EU Member States, meeting within the European Council, which allowed the Prime Minister to show to the general public that their concerns had been taken into account, and that the Netherlands had not unconditionally ratified the Agreement. In May 2017, a new law was passed approving the Association Agreement enabling the Netherlands to deposit its instrument of ratification on 15 June 2017. The association agreement entered into force on 1 September 2017. The 2014 Consultative Referendum Act also allowed for referendums to be held on Acts approving EU Treaties. The promoters of the Act acknowledged that it may not be the best target for referendums, but they considered it the only feasible manner to have the European project put before the people. Yet, after the second referendum based on the 2014 Referendum Act also brought the governing coalition difficulties – the majority voted no to the 2017 Intelligence and Security Services Act,47 but the Government and the Parliament decided to maintain it – the Referendum Act was withdrawn.48 It has, thus, never been used to put European Treaties before 42 Voorstel van rijkswet van de leden Herben en Van der Staaij houdende verklaring dat er grond bestaat een voorstel in overweging te nemen tot verandering in de Grondwet, strekkende tot invoering van het vereiste van een meerderheid van twee derden van het aantal uitgebrachte stemmen in de Staten-Generaal voor de goedkeuring van verdragen betreffende de Europese Unie [Draft law of the members Herben and van der Staaij, declaring that there is reason to consider a proposal to change the Constitution, aimed at introducing the requirement of a two-thirds majority of the votes cast in the parliament for the approval of treaties concerning the European Union) 30 874 (R 1818). 43 Law of 30 September 2014, laying down rules concerning the advisory referendum (Advisory Referendum Act). 44 Raad van State (hereinafter RvS) 26.10.2015, 201507896/1/A2 (de Kreek v Kiesraad) ECLI:NL:RVS:2015:3399. 45 Under the Act, the outcome of the referendum was valid if 30% of voters turn out. 46 Rechtbank Den Haag 12.04.2017, C-09-510607-HA ZA 16-553 (Stichting Forum voor Democratie v Staat der Nederlanden) ECLI:NL:RBDHA:2017:3667. 47 Wet van 26 juli 2017, houdende regels met betrekking tot de inlichtingen- en veiligheidsdiensten alsmede wijziging van enkele wetten, Stb 2017, 317. 48 Wet van 10 juli 2018 tot intrekking van de Wet raadgevend referendum (Stb 2018, 214). Attempts to judicially enforce a referendum on the withdrawal of the referendum act failed, see RvS 02.02.2018, 201800830/1/A2 (Stichting Meer Democratie v Minister van Binnenlandse Zaken en Koninkrijksrelaties) ECLI:NL:RVS:2018:364; and RvD 16.01.2019, 201806374/1/A2 (Stichting Meer Democratie v Minister van Binnenlandse Zaken en Koninkrijksrelaties) ECLI:NL:RVS:2019:98.

The Netherlands  543 a popular vote. The 2016-18 Constitutional Commission proposed the introduction of a binding corrective referendum.49 Time will tell whether there is sufficient political support for the proposal.

D.  EU Membership and the Dutch Constitutional Settlement Provisions of the Constitution that seem to be incomplete in the context of membership at the very least are simply read in light of EU law, and are further specified in ordinary law. Thus, for instance, Article 2 of the Constitution, stating that ‘All Dutch nationals shall be equally eligible for appointment to public service’ is read as applying also, where relevant and with the exceptions allowed under EU law, to EU citizens. Article 4 of the Constitution, stating that ‘Every Dutch national shall have an equal right to elect the members of the general representative bodies and to stand for election as a member of those bodies, subject to the limitations and exceptions prescribed by Act of Parliament’, is not considered to infringe EU law, as it does not say that the same does not apply, in particular cases, to EU citizens. The entire Constitution continues to apply when the Government and other state organs act in the context of the EU.50 These bodies cannot, for instance, infringe on their constitutional obligations when participating in the Council of Ministers or in other international organs in which they represent the state. The conventional principles of ministerial responsibility and the duty to inform the parliament and so forth continue to apply in the sphere of European integration. Government–Parliament relations are the same in the context of EU decision-making, as they are in purely national matters. The Constitution also does not contain specific provisions on the domestic effect and application of EU law, and the general provisions relating to treaty law, therefore, would seem to apply. Indeed, the case law of the ECJ on direct effect and primacy of EU law fell on constitutionally fertile ground in the Netherlands. This is because the Constitution provides for domestic effect and primacy of directly effective treaty provisions and decisions of international organisations, as we have seen.51 Paradoxically, while the constitutional provisions seem highly apt to allow for the direct effect and primacy of EU law, the prevailing view is that they have no bearing on the effect of EU law.52 Many argue that the effect of EU law in the national legal order is based on the case law of the ECJ, rather than on the Constitution.53 Most constitutional scholars hold that Dutch participation in the EU and the application with precedence of EU law has very little to do with the Constitution. The Constitution is not considered to be the basis for it, and at most, the Constitution ‘does not impede the application with precedence’ of EU law. It is as if the Constitution evaporates in the presence of the EU. The courts have never dwelled much on the issue, but in a very brief and thinly motivated judgment of 2004, the Hoge Raad merely stated that Articles 93 and 94 of the Constitution do

49 Lage drempels (n 16). 50 See the report of the 2009-10 Constitutional Commission. 51 See Claes/de Witte, ‘Netherlands’ (n 36); Bruno de Witte, ‘Do Not Mention’ (n 4). 52 For the discussion on this issue, see: Evert Alkema, ‘Noot onder HR 2 november 2004’, (2005) Nederlands Juristenblad 80; Leonard Besselink et al (eds), De Nederlandse Grondwet en de Europese Unie (Groningen, Europa Law Publishing, 2002); Michiel van Emmerik, ‘De Nederlandse Grondwet in een veellagige rechtsorde’ [‘The Dutch Constitution in a Multi-layered Legal Order’], (2008) 4 Rechtsgeleerd Magazijn Themis 145, 149 f; Leonard Besselink, ‘Noot onder HR 1 oktober 2003, LJN: AO8913 en HR 2 november 2004, LJN: AR1797’, (2005) 53 SEW, Tijdschrift voor Europees en economisch recht 336. 53 More specifically the ECJ cases 6/64 (Costa v E.N.E.L.) and (Van Gend & Loos) (n 32).

544  Monica Claes not apply to EU law.54 It should be recalled, however, that these provisions were inserted in the Constitution in 1953 precisely with a view to facilitate Dutch participation in the rapidly developing process of European integration. Moreover, in the course of the 1983 revision of the Constitution, the Government confirmed that EU law takes effect with primacy through these provisions of the Constitution.55 Irrespective of whether the constitutional provisions serve as the basis for the direct effect and primacy of EU law, it is achieved. One way or the other, Dutch courts are obliged to give direct effect to EU law in the Netherlands, setting aside Dutch law if necessary, and they do. The issue, thus, remains rather theoretical and the courts adopt a highly pragmatic approach. This issue is also not the subject of much academic debate compared to countries like Germany or Italy. As a consequence, Dutch judges have embraced their European mandate as developed in the case law of the ECJ/CJEU more smoothely than in any of the other founding Member States.56 Dutch judges are keen to make references to the CJEU, and have done so since the very beginning. Ten of the first 13 questions referred to the ECJ came from Dutch courts. Between 1952 and 2008, Dutch courts referred a total of 719 questions.57 This is roughly in the same range as the number of references made by French courts (755) during the same period.58 It is no coincidence that Van Gend en Loos was referred by a Dutch court.59 To this day, Dutch lawyers take pride in the fact that in that case, the ECJ extended Dutch monism to all EU Member States. The courts have never had to develop new doctrines in order to be able to exercise their European mandate, nor have they had to deal with the sometimes very delicate and sensitive issues which some of the (constitutional) courts in the other Member States have had to tackle. The legal issues Dutch courts struggle with, are typically procedural issues dealing with ‘second generation’ questions. The question is not so much whether, but how European law should be applied.

IV.  Constitutional Limits to European Integration? The Constitution has never been considered an obstacle to the accession to the European Communities and the Union, or to the domestic effect of European law. The Constitution never played much of a role. The public debate on those Treaties, including the rather limited debate on the Constitutional Treaty and on the Lisbon Treaty, has hardly, if at all, dealt with issues of their constitutionality.

54 HRNL 02.09.2004, 00156/04 E (Base for Validity) ECLI:NL:HR:2004:AR1797. See also Kabinetsnotitie van 21 september 2007 n.a.v. motie Visser (Kamerstukken II 2006/07, 29 861, no 15). 55 Van Emmerik, ‘Nederlandse Grondwet’ (n 52) 149; Besselink, ‘Kingdom’ (n 28) 1205. 56 For a comparative analysis, see Monica Claes, The National Courts’ Mandate in the European Constitution (Oxford, Hart, 2006). 57 Between 1952 and 2008, the Administrative Jurisdiction Division of the RvS made 59 references, the HRNL 177, and the Administrative High Court referred 46 times, the Trade and Industry Appeals Tribunal 137 times. The Tariff Commission, which no longer exists as a separate tribunal since 2000, asked for a preliminary ruling 34 times. Other courts or tribunals made 266 references. 58 Annual report of the ECJ and the Court of First Instance, http://curia.europa.eu/jcms/jcms/Jo2_7000/ annual-report, Statistics, table 20: General trend in the work of the Court (1952-2008) – New references for a preliminary ruling (by Member State and by court or tribunal). 59 On Van Gend en Loos and its context see Monica Claes, ‘Het arrest Van Gend en Loos’ [‘The Van Gend and Loos Judgment’], in R Schutgens et al (ed), Canon van het Recht (Nijmegen, Ars Aequi Dutch Law, 2010) 145; see also Bruno De Witte, ‘The Continuous Significance of Van Gend en Loos’, in M Maduro and L Azoulai (eds), Past and Future of EU Law: The Classics of EU Law Revisited on the 50th Anniversary of the Rome Treaty (Oxford, Hart, 2010) 9; Daniel Halberstam, ‘Pluralism in Marbury and Van Gend’, in ibid, 26; and Leeuwen, ‘Democratic Concerns’ (n 34).

The Netherlands  545 The Constitution is completely open to European integration, to an extent even that it would seem that it would be possible for the state to commit constitutional hara-kiri.60 The Dutch Constitution remains agnostic towards fundamental questions of European integration. As a consequence of the openness of the Dutch Constitution, the latter has never developed into a legal obstacle to foreign influence, or as a harness for national identity.61 There is no list of essential state functions that cannot be transferred, and the Constitution mentions no principles that cannot be amended or restricted in the light of European integration. Instead, the constitutional doctrine is rather pragmatic. An example is the absence of a legal sovereignty debate and the absence of a fully-fledged theory of an unamendable constitutional core. This is even more the case, if the Constitution is considered not to even govern the relationship between the European and the national legal orders. The issue of limits to participation in European integration is accordingly a political question rather than a legal constitutional one. In fact, the concept of sovereignty is largely absent from the legal doctrinal debate, both in the context of European integration and more generally, with reference to the question of ultimate authority and the source of all state power. There is no theory of national or popular sovereignty. Constitutional thinking seems to have a different starting point: the Constitution deals mainly with limiting the powers of the previously sovereign King. In the past, this has made for a very smooth adaptation to membership, since no core principles, no inalienable principles or values require protection against the majority of the day or against the outside world. The controlimiti developed in other Member States are not developed in Netherlands constitutional doctrine. There is no Solange case law, if only because constitutional rights are less enforceable than EU law and fundamental rights. Thus, EU law was never considered as a threat to national constitutional rights. No Dutch court has ever claimed jurisdiction to review the validity or applicability of EU law, or developed a theory of ultra vires acts; no court has ever formulated a view on ‘national identity’ or ‘constitutional identity review’. This approach sits well with Dutch consociationalism and corporatism and the patterns of democracy that have prevailed in the Netherlands for a long time. This constitutional culture has proven to be a fertile soil for EU law. Also, the theory of constitutional pluralism sits well with Dutch constitutional theory. European constitutional pluralism could be only a new, transnational guise of a broader pluralism, which is known as soevereiniteit in eigen kring (‘sovereignty within one’s own sphere’).62 Nevertheless, the question has now been raised in political academic and in government circles whether the Dutch legal order should continue to be so open towards the EU, and whether the Constitution could not or should not provide a bulwark against the incoming tide of EU law. Moreover, the Netherlands happened to be one of the Member States which led to the defeat of the Constitutional Treaty, and is no longer so Europhile.

60 Polak, referred to in Bruno de Witte, ‘Do Not Mention’ (n 4). 61 V Rossem, Soevereiniteit (n 4) 123. 62 This idea was developed in Calvinist circles, initially by Kuyper in the late 19th century and in a more profound philosophical sense by Herman Dooyeweerd as a counterpart of Catholic doctrines of ‘subsidiarity’ and the social-democratic theory of ‘functional decentralisation’. See more extensively Leonard Besselink and Monica Claes, ‘The Netherlands: The Pragmatics of a Flexible, Europeanised Constitution’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law – National Reports (The Hague, TMC Asser Press, 2019) 177 section 1.3.3.

546  Monica Claes

V.  Constitutional Rules and/or Practice on Implementing EU Law The Constitution contains no special provisions on the role of Parliament in EU decisionmaking. The normal rules concerning government-parliament relations – the negative rule of confidence and the principles of political accountability – continue to apply.63 The practice of the Netherlands States General has been varied. For a long time, the Upper House was relatively more alert and systematic in its approach to European decision-making. The Lower House has become more active since the approval of the Lisbon Treaty. Hitherto, formal legally binding procedures existed only for certain EU decisions. These were mainly the legally binding decisions taken in the field of justice and home affairs (the so-called pre-Lisbon third pillar decisions). These decisions required the prior consent of both Houses before the Dutch representative could vote in the Council. Thus the Houses had effective powers to enforce the duty of the government to provide information to Parliament. In the Dutch approach, the European Parliament’s lack of formal co-decision powers was the major justification for a role of the Dutch Parliament, and the increase of co-decision powers of the European Parliament was used as a justification to tone down the role of the national Parliament.64 The requirement of prior consent was accordingly abolished with the approval of the Lisbon Treaty, and the Government no longer provides as much information to Parliament as previously. Instead, the Act on Approval of the Lisbon Treaty contains a new provision, allowing either House to stipulate that the Government makes a parliamentary scrutiny reservation in the Council to any legislative EU proposal which the House deems of special political importance, in order for it to scrutinise the proposal. This reservation is in principle valid for four weeks. The Constitution does not provide for any special rules on the implementation of EU law: the normal rules of the legislative game continue to apply. Moreover, there is a broad awareness that adopting legislation with a view to implementing EU law comes with the obligation to comply with EU law requirements and that EU law takes precedence over national law and national policy preferences.65

VI.  Resulting Relationship between EU Law and National Law A.  Explaining the Smooth Constitutional Engagement with Europe in a ‘Europhile Country’ As already explained, the Dutch constitutional system is particularly well placed to be receptive to EU law and its doctrines of direct effect and primacy. There is no controlimiti or constitutional identity case law, and no ultra vires doctrine. 63 Eg Art 68 GW (the information duty) and Art 42 GW (active duty to provide information). On the Dutch Parliament and EU decision-making, see Brecht van Mourik, Parlementaire controle op Europese besluitvorming [Parliamentary Control of European Decision-making] (Tilburg, Wolf Legal Publishers, 2012); Leonard Besselink and Brecht van Mourik, ‘The Roles of the National Parliament and the European Parliament in EU Decision-Making: The Approval of the Lisbon Treaty in the Netherlands’, (2009) 15 European Public Law 307. 64 It should be stressed that the reasoning is flawed: increased powers of the EP to co-decide do not replace control powers of national parliaments of national members of the Council. 65 Ellen Mastenbroek, ‘Guardians of EU law? Analysing roles and behaviour of Dutch legislative drafters involved in EU compliance’, (2017) 24 Journal of European Public Policy 1289.

The Netherlands  547 Why did the approval of the European Treaties, the application of EU law and the case law of the CJEU never give rise to any constitutional problems in the Netherlands? A large part of the explanation lies in the sober nature of the Dutch Constitution, the pragmatic nature of the Dutch constitutional debate, and the absence of a constitutional court. Of course, it is impossible to prove that the absence of a constitutional court contributes to the absence of constitutional problems. However, comparison with other Member States may serve to show that where constitutional courts are established as guardians of the constitution, as agents of the people or as trustees of the political system, these courts will be eager to protect and defend the Constitution and the values it enshrines. This is so, even where the text of the Constitution itself seems to be more open,66 or where the Constitution makes EU law immune to constitutional review.67 Another part of the explanation can be found in the Dutch version of monism, which comes closer to the monist extreme than any other EU Member State. Probably no other national system in the EU is as unconditionally open and receptive to treaties as the Netherlands.68 The prime role of treaties, especially human rights treaties, in the domestic legal order has further decreased the status of the Constitution as a limiting and guiding document. Dutch constitutional law and its history can help to understand why European law found such easy acceptance in the Netherlands. However, they cannot explain and justify everything. The openness of the Dutch Constitution is usually further explained by the need for good international relations of a small trading nation. There may even be a straight line from Grotius’ Mare Liberum to the radical version of pacta sunt servanda espoused in Dutch internationalist doctrine. The receptivity of international and EU law may be the combined result of national interest and internationalist ideology. Even today, Dutch jurists and politicians bring a missionary spirit to international relations and pretend to contribute to the world order by setting the example. Lofty principles are combined by a rather pragmatic application of those principles, without causing any real disruptions of the domestic legal order.69 Moreover, the legal and political élite were in the founding period strongly in favour of European integration. The Dutch academic community and the association of European law, the Dutch association of FIDE, took up the question of the application of EU law in the domestic order, and made sure that the question was referred to the ECJ in Van Gend en Loos.70 Popular support for Dutch participation in European integration has always been high.

B.  A Sudden Drop of Temperature? Nevertheless, over recent decades, public support for European integration has dropped, probably due to the same anxieties that have caused a drop in popular support elsewhere. Such anxieties include globalisation, the economic crisis, the feeling that the EU is overly intrusive, the return to nationalism and the idealisation of Dutch values and choices, immigration, the enlargement process, the fact that the Netherlands has become a net contributor to the EU, and fear of the unknown. 66 BVerfG 12.10.1993, 2 BvR 2134 (Maastricht) BVerfGE 89, 155; and Art 23 GG. 67 As is the case in the Constitution of the Republic of Ireland. 68 This is interesting in itself, since it would be more correct to say that these treaties are supra-constitutional. This is further developed below. 69 Claes/de Witte, ‘Netherlands’ (n 36) 189. 70 On the fascinating story of the case of Van Gend en Loos on the making of the doctrines of direct effect and primacy, Antoine Vauchez, ‘The transnational politics of judicialization. Van Gend en Loos and the making of EU polity’, (2010) 16 European Law Journal 1.

548  Monica Claes And yet, even at this time of Euroscepticism, Dutch public actors tend to play by the rules agreed with the other EU Member States. Where EU rules are challenged, the Government is usually called upon to seek amendment of these rules in accordance with the procedures provided in the Treaties, rather than to break the rules. Wilders’ PVV, which takes the most sceptical view towards European integration, regularly warns that if the EU continues to create obstacles for developing policies, mainly with a view to blocking further immigration and expelling foreigners, the Netherlands will have to withdraw from the EU. Yet, there is overall no strong call for leaving the EU, and membership is still considered beneficial.

C.  Amending the Constitution? For a number of years now, there has been a growing awareness that the absence of the EU in the Constitution may be an anomaly.71 Several proposals for constitutional amendment have been made over the past decades. They proposed constitutionalising membership,72 adding the words ‘and European’ to Article 90 of the Constitution,73 which would read: ‘the government should promote the development of the European and international legal order’, or requiring a special majority for the approval of EU Treaties. None of the proposed amendments have gained enough support to see the light of day. The Constitution is difficult to amend, and moreover, there seems to be no legal urgency to introduce the EU in the Constitution. The Constitution as it stands has not hindered Dutch membership and it does not prevent EU law from having its full force and effect in the Dutch legal order. In addition, while the view that Europe at times harms Dutch interests is currently widespread, Europe is not considered to threaten the principles and values of the Constitution. While the absence of the EU from the Constitution may be an anomaly, it does not complicate Dutch EU membership, and the added value of one or more provisions concerning the EU and EU membership is not evident. Currently, however, a private member’s bill74 is still pending to change the procedure for the approval of EU Treaties, requiring a two-thirds majority for their endorsement. By strengthening the conditions for the approval of such treaties, the constitutional implications of the EU Treaties would be acknowledged. At the same time, the procedure would not be as complex as for the amendment of the Constitution, which requires votes in two consecutive Parliaments and is considered less intrusive than a referendum. A two-thirds majority would be required for all Treaty amendments, including those following the simplified procedure and accession Treaties. Furthermore, the procedure would apply irrespective of whether or not the Treaty would or would not contain provisions which could be considered contrary to the letter or spirit of the Constitution. The 2009–10 Constitutional Commission was invited to give its view on the proposal. The Commission did not see a need for the introduction of a Europe provision, and instead proposed

71 See eg the motion of Senator Jurgens, Parliamentary proceedings (Senate) 2000/01, 26 200 VI, No 65, No 37b; Besselink et al, Nederlandse Grondwet (n 52); van Emmerik, ‘Nederlandse Grondwet’ (n 52) 155 ff. In advice to the Act approving the TECE the RvS mentioned the issue, but it did not draw any conclusions. 72 The Scientific Council for Government Policy (Wetenschappelijke Raad voor het regeringsbeleid – WRR) recommended that two provisions be added to the Constitution: a provision that explicitly states the value and objectives of Dutch EU membership, and a provision that aims to ensure the autonomy of the Netherlands within the framework of the EU. The Government rejected the proposals as vague. 73 Staatscommissie Grondwet, Rapport Staatscommissie Grondwet (November 2010), https://zoek.officielebekendmakingen.nl/blg-86969.pdf. 74 See n 42.

The Netherlands  549 a less restrictive reading of the existing Article 91(3) of the Constitution, so that certain European Treaties that affect the spirit of the Constitution would be covered and hence require a two-thirds majority. In addition, the Committee advised the Government to consider introducing the words ‘and European’ in the text of Article 90 of the Constitution: ‘The [government]75 contributes to the development of the international and European legal order’. The addition may seem rather meagre at first sight. However, the Committee preferred this to a mere declaratory statement that ‘The Netherlands is a member of the European Union’. Indeed, the inclusion in Article 90 of the Constitution would impose a constitutional obligation to participate in the EU and to do so with a view to the development of the European legal order. The provision furthermore would have to be read in the context of the entire Constitution, which binds the Government and other public officials who act in the context of European integration: the Constitution continues to apply to their participation in the European context. The EU would not be mentioned in so many words but would feature under the broader concept ‘European’, which also covers the ECHR. Furthermore, with respect to the question of the core principles of the Constitution, the Commission proposed the introduction of a general provision in the Constitution stating the fundamental principles of the Constitution. The Commission was unable to agree on the effect of the provision, and on what should happen if a treaty or a decision of an international organisation would infringe the general provision. Five members were of the opinion that if such exceptional situation should occur, the treaty or decision should not be given effect in the domestic legal order. In practical effect, the absolute priority of treaties over the Constitution would be qualified. The remaining members of the Commission were of the opinion, that in case it would breach fundamental principles, a treaty or decision should not be applied, but that the grounds for such disapplication should be found in international law, for instance in human rights treaties. The reactions to the proposal were lukewarm at best. A few years later, in 2014, the Government finally proposed the introduction of a general provision in the Constitution, reading: ‘The Constitution guarantees democracy, the rule of law and fundamental rights’.76 The proposal has been slightly amended and approved in the first reading and now awaits a two-thirds majority approval after the next elections.77 The impact of the provision – if carried – is not yet clear. It has not been put forward as a bulwark against European integration, even though some authors have argued that it could be used as the confirmation that EU law – just as much as national law and even constitutional amendments – cannot and should not infringe the basic principles of the Dutch Constitution.78

75 A majority also advised that the obligation should be addressed to all public bodies, rather than to the Government alone as the current provision does, in order to reflect that international relations and international law are no longer the responsibility of the Government alone. 76 ‘De Grondwet waarborgt de grondrechten en de democratische rechtsstaat’ [‘the Constitution ensures fundamental rights and democracy under the rule of law’], Wet van 9 maart 2018, houdende verklaring dat er grond bestaat een voorstel in overweging te nemen tot verandering van de Grondwet, strekkende tot het opnemen van een algemene bepaling [Law of 9 March 2018, stating that there is reason to consider a proposal to amend the Constitution, to include a general provision], Stb 2018, 86. 77 Maarten Stremler, ‘De voorgestelde algemene bepaling als fundamentele constitutionele norm’ [‘The Proposed General Provision as a Fundamental Constitutional Standard’], (2018) 9 Tijdschrift voor Constitutioneel Recht 204. 78 Ibid, 216; Janneke Gerards, ‘Een algemene bepaling in de Grondwet?’ [‘A General Provision in the Constitution?’], in De grondwetsherziening van 1983: 30 jaar oud of 30 jaar jong? (The Hague, Ministerie van Binnenlandse Zaken en Koninkrijksrelaties, 2013) 14, 24 f.

550  Monica Claes

VII. Conclusion In the current political landscape and in the face of a critical public opinion and growing resistance to European integration, constitutional law has very little to offer in terms of a theoretical framework. While it may be true that salvation has never come from legal doctrine and constitutional concepts, the complete absence of any form of conceptualisation and constitutional theory on the transformation of the state and public authority as a consequence of European integration, has left the state dumbstruck when faced with questions concerning the European Constitution. The recent referendums demonstrate how the pragmatic approach towards European integration has backfired. Indeed, the concerns occupying the minds of people, politicians, lawyers and judges in the Netherlands, are the same as elsewhere. Indeed, the concept of sovereignty does feature in the public debate on European integration, but it not as a concept of constitutional theory. ‘Sovereignty’ is shorthand for ‘national decision making power’. Leonard Besselink warned, on the occasion of the 150th anniversary of the Constitution in 1998, that the constitutional culture of the Netherlands, characterised by its openness and pragmatism, its aversion of grand principles and doctrines, and the gap between the Constitution and politics, which together have made the insertion in a broader European constitutional project smoother, would, one time or another, bounce back: In the Netherlands, the issue of what is or should be unchangeable in the Constitution has never really been debated. But the question will arise one time or another, when the Constitution will be included in another constitutional system which is not decided nationally and which is not put in perspective by a political culture like the Dutch, but which is embedded in a culture governed by hierarchical, legal and absolute claims of validity. The fundamental question of why we have a Constitution and what for, will then be asked.79

References M Adams and G van der Schyff, ‘Constitutional Review by the Judiciary in the Netherlands. A Matter of Politics, Democracy or Compensating Strategy?’, (2006) 66 Zeitschrift für ausländisches öffentliches Recht und Völkerrecht 399. E Alkema, ‘Noot onder HR 2 november 2004’, (2005) Nederlands Juristenblad 80. L Besselink, ‘An Open Constitution and European Integration: The Kingdom of the Netherlands’, (1996) 44 SEW, Tijdschrift voor Europees en economisch recht 192. L Besselink, ‘Constitutionele klimatologie’, (1998) Nederlands Juristenblad 212. L Besselink, Constitutional Law of the Netherlands (Nijmegen, Ars Aequi Dutch Law, 2004). L Besselink, ‘Noot onder HR 1 oktober 2003, LJN: AO8913 en HR 2 november 2004, LJN: AR1797’, (2005) 53 SEW, Tijdschrift voor Europees en economisch recht 336. L Besselink, ‘Constitutional Adjudication in the Era of Globalization: The Netherlands in Comparative Perspective’, (2012) 18 European Public Law 231.

79 In the original: ‘In Nederland is er nooit uitdrukkelijk nagedacht over wat er onveranderbaar zou zijn (of zou moeten zijn) aan de Grondwet. Maar deze kwestie zal zich vroeg of laat wel voordoen als men de Grondwet ondergeschikt maakt aan een andere constitutie die niet in overwegende mate op nationale schaal bepaald is, en die niet gerelativeerd wordt door een politieke cultuur als de Nederlandse, maar juist ingebed is in een cultuur die door meer hiërarchische, juridische en absolute geldigheidsclaims gekenmerkt wordt. Ook de kritieke vraag waartoe wij een eigen Grondwet hebben, zal dan in volle omvang worden gesteld’, Leonard Besselink, ‘Constitutionele klimatologie’, (1998) Nederlands Juristenblad 212.

The Netherlands  551 L Besselink, ‘The Kingdom of the Netherlands’, in L Besselink et al (eds), Constitutional law of the EU member states (Alphen aan den Rijn, Kluwer 2014) 1187. L Besselink and M Claes, ‘The Netherlands: The Pragmatics of a Flexible, Europeanised Constitution’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law – National Reports (The Hague, TMC Asser Press, 2019) 177. L Besselink and B van Mourik, ‘The Roles of the National Parliament and the European Parliament in EU Decision-Making: The Approval of the Lisbon Treaty in the Netherlands’, (2009) 15 European Public Law 307. L Besselink et al (eds), De Nederlandse Grondwet en de Europese Unie (Groningen, Europa Law Publishing, 2002). G Boogaard, Het Wetgevingsbevel. Over constitutionele verhoudingen en manieren om de wetgever tot regelgeving aan te zetten [Court Orders to Legislate. On Constitutional Relations and Ways to Entice the Legislature to Act] (Tilburg, Wolf Legal Publishers, 2013). J Brouwer, Treaty law and practice in The Netherlands (Groningen, 2002). M Claes, The National Courts’ Mandate in the European Constitution (Oxford, Hart, 2006). M Claes, ‘Het arrest Van Gend en Loos’ [‘The Van Gend and Loos Judgment’], in R Schutgens et al (eds), Canon van het Recht (Nijmegen, Ars Aequi Dutch Law, 2010) 145. M Claes and B de Witte, ‘Report on the Netherlands’, in AM Slaughter, A Stone Sweet and JHH Weiler (eds), The European Courts and National Courts. Doctrine and Jurisprudence (Oxford, Hart, 1997) 171. J de Wit, Artikel 94 toegepast. Een onderzoek naar de betekenis, de bedoeling en de toepassing van ‘vinden geen toepassing’ in Artikel 94 van de Grondwet [Applying Article 94 of the Constitution. A Study into the Meaning, the Intention and the Application of the Words ‘Shall not be Applicable’ in Article 94 of the Constitution] (The Hague, Boom Juridische Uitgevers, 2012). B de Witte, ‘Do Not Mention the Word: Sovereignty in Two Europhile Countries: Belgium and The Netherlands’, in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 359. B de Witte, ‘The Continuous Significance of Van Gend en Loos’, in M Maduro and L Azoulai (eds), Past and Future of EU Law: The Classics of EU Law Revisited on the 50th Anniversary of the Rome Treaty (Oxford, Hart, 2010) 9. J Fleuren, ‘Recent developments regarding the direct and indirect application of treaties by Dutch courts. Fresh approaches to self-executing, non-self-executing and non-binding international law’, (2016) 47 Netherlands Yearbook of International Law 377. J Gerards, ‘Een algemene bepaling in de Grondwet?’ [‘A General Provision in the Constitution?’], in De grondwetsherziening van 1983: 30 jaar oud of 30 jaar jong? (The Hague, Ministerie van Binnenlandse Zaken en Koninkrijksrelaties, 2013) 14. J Gerards, ‘The Irrelevance of the Netherlands Constitution, and the Impossibility of Changing It’, (2016) 77 Revue interdisciplinaire d’études juridiques 207. D Halberstam, ‘Pluralism in Marbury and Van Gend’, in M Maduro and L Azoulai (eds), Past and Future of EU Law: The Classics of EU Law Revisited on the 50th Anniversary of the Rome Treaty (Oxford, Hart, 2010) 26. A Heringa, ‘De verdragen van Maastricht in strijd met de Grondwet. Goedkeuring met twee derde meerderheid?’ [‘The Maastricht Treaties are in Conflict with the Constitution. Approval by Two-thirds Majority?’], (1992) Nederlands Juristenblad 749. J Hollander, Constitutionalising Europe. Dutch Reactions to an Incoming Tide (Groningen, Europa Law Publishing, 2013). CAJM Kortmann, ‘Wegwerprecht, oude dame of frisse juf?’ [‘Disposable Law, Old Lady or Young Damsel?’], in De Grondwet herzien. 25 jaar later. 1983-2008 (The Hague, Directie Constitutionele Zaken en Wetgeving, 2008) 7. A Lijphart, The Politics of Accommodation: Pluralism and Democracy in the Netherlands (Berkeley CA, University of California Press, 1968). E Mastenbroek, ‘Guardians of EU law? Analysing roles and behaviour of Dutch legislative drafters involved in EU compliance’, (2017) 24 Journal of European Public Policy 1289. C Möllers, ‘Legalität, Legitimität und Legitimation des Bundesverfassungsgerichts’, in C Möllers et al (eds), Das entgrenzte Gericht. Eine kritische Bilanz nach sechzig Jahren Bundesverfassungsgericht (Berlin, Suhrkamp, 2011) 283.

552  Monica Claes A Nollkaemper, ‘The Application of Treaties in the Netherlands’, in D Schloss (ed), The Role of Domestic Courts in Treaty Enforcement: A Comparative Study (Cambridge, CUP, 2010) 326. B Oomen. ‘Constitutioneel bewustzijn in Nederland: Van burgerschap, burgerzin en de onzichtbare grondwet’ [‘Constitutional Awareness in the Netherlands: on Citizenship, Civic Responsibility and the Invisible Constitution’], (2009) 3 Recht der Werkelijkheid 55. B Oomen and H Lelieveldt, ‘Onbekend maar niet onbemind: Wat weten en vinden Nederlanders van de grondwet?’ [‘Unknown but not Unliked: what Do the Dutch Know About the Constitution and what Do They Think of it?’], (2008) Nederlands Juristenblad 577. M Stremler, ‘De voorgestelde algemene bepaling als fundamentele constitutionele norm’ [‘The Proposed General Provision as a Fundamental Constitutional Standard’], (2018) 9 Tijdschrift voor Constitutioneel Recht 204. J Uzman, Constitutionele remedies bij schending van grondrechten – Over effectieve rechtsbescherming, rechterlijk abstineren en de dialoog tussen rechter en wetgever [Constitutional Remedies for Fundamental Rights Violations – on Effective Judicial Protection, Judicial Inaction and the Dialogue Between Judge and Legislature], (Alphen aan den Rijn, Kluwer, 2013). M van Emmerik, ‘De Nederlandse Grondwet in een veellagige rechtsorde’ [‘The Dutch Constitution in a Multi-layered Legal Order’], (2008) 4 Rechtsgeleerd Magazijn Themis, 145. K van Leeuwen, ‘On Democratic Concerns and Legal Traditions: The Dutch 1953 and 1956 Constitutional Reforms “Towards” Europe’, (2012) 21 Contemporary European History 357. B van Mourik, Parlementaire controle op Europese besluitvorming [Parliamentary Control of European Decision-making] (Tilburg, Wolf Legal Publishers, 2012). JW van Rossem, Soevereiniteit en pluralisme [Sovereignty and Pluralism] (Alphen aan den Rijn, Kluwer, 2014). A Vauchez, ‘The transnational politics of judicialization. Van Gend en Loos and the making of EU polity’, (2010) 16 European Law Journal 1. W Voermans et al, Juridische betekenis en reikwijdte van het begrip ‘rechtsstaat’ in de legisprudentie & jurisprudentie van de Raad van State [Legal Meaning and Scope of the Notion of ‘rechtsstaat’ in the Legisprudence and Jurisprudence of the Council of State] (The Hague, Raad van State, 2011).

20 Austria Membership without Reservation – Or with Hidden Reservations? STEFAN GRILLER

I.  Main Characteristics of the National Constitutional System A.  Fragmentation and Flexibility The Austrian Constitution is, generally speaking, of a fragmented and flexible nature.1 Its main instrument, the Bundes-Verfassungsgesetz (‘Federal Constitutional Act’) (B-VG)2 dates back to 1920, and has been in force ever since, interrupted by the authoritarian period 1934–45. The B-VG has been amended more than 100 times by Constitutional Acts. This figure relates only to those amendments which were incorporated into the text of the B-VG. However, this is not obligatory. Amendments to the Constitution can instead also be enacted by leges fugitivae, meaning both Constitutional Acts other than the B-VG and constitutional provisions embedded within acts of ordinary legislation. More than 50 such Acts and 500 such provisions, all of them having the same legal status as the B-VG itself, are still in force. However, the Austrian Constitution is not only fragmented, it is also highly flexible, the former resulting from the latter. There is no ‘eternity clause’, meaning that, at least in principle, everything can be changed, provided that the respective procedure for constitutional amendments is observed.3 The respective hurdles, compared to most other EU Member States, are low. ‘Normal amendments’ require a two-thirds majority vote in Parliament (the National Council/

1 For a general introduction in English language compare eg Herbert Hausmaninger, The Austrian Legal System, 4th edn (Vienna, Manz, 2011); Manfred Stelzer, The Constitution of the Republic of Austria (Oxford, Hart, 2011). For some of the most common textbooks in German compare: Walter Berka, Verfassungsrecht, 8th edn (Vienna, Verlag Österreich, 2021); Christoph Grabenwarter and Michael Holoubek, Verfassungsrecht – Allgemeines Verwaltungsrecht, 4th edn (Vienna, Facultas, 2019); Heinz Mayer, Gabriele Kucsko-Stadlmayer and Karl Stöger, Grundriss des österreichischen Bundesverfassungsrechts, 11th edn (Vienna, Manz, 2015); Theo Öhlinger and Harald Eberhard, Verfassungsrecht, 12th edn (Vienna, Facultas, 2019). 2 The English version (as amended until 2017) is to be found at www.ris.bka.gv.at/Dokument.wxe?Abfrage=Erv&Dok umentnummer=ERV_1930_1. A list of Austrian laws in English is at www.ris.bka.gv.at/RisInfo/LawList.pdf. 3 Art 44 B-VG. The same is true at the level of the Austrian provinces (Länder), which are entitled to enact constitutions of their own, albeit within the limits of the federal constitution (Art 99 B-VG). The Länder level is not explored in the following.

554  Stefan Griller Nationalrat), with at least half of the MPs being present, and the explicit designation of the provision as a Constitutional Act. A referendum for such a ‘normal constitutional act’ is not obligatory. It is, however, to be performed if a third of the members of Parliament should so wish, and its result is then binding. A referendum is mandatory (and its result of course also binding) only in cases of a total revision of the Constitution.4 This requirement comes in addition to the two-thirds majority and the explicit designation as a Constitutional Act. ‘Total revision’ (Gesamtänderung) in this context is a terminus technicus: It has to take place whenever one of the fundamental principles of the Constitution is to be amended or abandoned. These principles are, according to the prevailing view, democracy, the rule of law, separation of powers, the liberal principle enshrined in the guarantee of fundamental rights and freedoms, and the federal and the republican organisation of the state. Only one such total revision has been adopted: in preparing for Austria’s EU membership in 1994.

B.  A Strong and Threefold Judiciary The Austrian judiciary is modelled after the traditional continental differentiation between private and public law. Private and penal law are primarily matters for the ordinary courts. The final instance in civil and criminal proceedings is the Supreme Court (Oberster Gerichtshof, OGH).5 At least in principle, civil and penal jurisdiction is separate from the public law courts, which consist of the Administrative Courts, and the Constitutional Court (Verfassungsgerichtshof, VfGH). As a result, there are three highest courts, competent for their respective area of competence. There is no direct remedy between these courts. In this sense, they are independent from each other, even if interlinkages do exist.6 The Constitutional Court is empowered, inter alia, to scrutinise alleged violations of the constitution both by general norms (eg of Acts of Parliament or ordinances issued by the­ executive)7 as well as by individual – and by origin: administrative – decisions.8 Two review bases are of specific importance, both in terms of quality and quantity. One of these bases is the protection of the Fundamental Rights standards, which form part of the Constitution. The other base is the enforcement of the (formal) rule of law principle,9 requiring the democratically elected legislator to determine as precisely as possible the powers of the administration and the judiciary. The administrative court system was reformed in 2014. That reform was to a considerable extent driven by the guarantees enshrined in Article 6 of the European Convention on Human Rights (ECHR) and Article 47 of the EU Charter of Fundamental Rights (CFR).10 The new system 4 Art 44 para 3 B-VG (Bundes-Verfassungsgesetz = Federal-Constitutional Law). For the minority position – not developed in the above text – that the core of the constitution cannot be amended legally, also not under Art 44 para 3: Peter Pernthaler, Der Verfassungskern (Vienna, Manz, 1998); the prevailing view contends the opposite: see, eg Markus Vašek, Unabänderliches Verfassungsrecht und Revisionsschranken in der österreichischen Bundesverfassung (Vienna, Verlag Österreich, 2013). 5 Art 92 B-VG. 6 Compare Michael Holoubek, ‘Auswirkungen des Unionsrechts auf das Verhältnis der österreichischen Höchstgerichte zueinander’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 625. 7 Arts 139 and 140 B-VG. 8 Art 144 B-VG. This is now ‘mediated’ by the Administrative Courts which review administrative decisions, meaning that what can be scrutinised by the VfGH are no longer administrative decisions as such, but Administrative Court decisions reviewing individual administrative measures. 9 Compare below in the text near n 95. 10 Compare Stefan Griller, ‘Die Neuordnung der Gerichtsbarkeit des öffentlichen Rechts’, (2018) 20(1) ÖJT, 7 ff.

Austria  555 consists of nine Administrative Courts (Verwaltungsgerichte), one for each province, and two federal courts: the Federal Administrative Court (Bundesverwaltungsgericht), and the Federal Financial Court (Bundesfinanzgericht).11 All of these administrative courts decide on complaints against administrative authorities, mainly administrative decisions addressed to individuals (Bescheide). All of these courts are, albeit to a limited extent,12 subject to appeals, which might be filed with the High Administrative Court (Verwaltungsgerichtshof, VwGH). The latter then acts as a court of last instance and a guarantor for uniform jurisprudence. With the exception of the administrative courts of the provinces, all courts are established by the Federation. As mentioned, there is, in principle, no legal remedy which would introduce a hierarchy between ordinary and public law courts. There is an important modification of this principle regarding fundamental rights protection. It is possible to file a complaint with the Constitutional Court against judgments of the lower courts, claiming that a violation of fundamental rights occurred. But such a remedy is not available against judgments of the Supreme Court or the High Administrative Court.13 Consequently, even here, there is no strict hierarchy which would provide for an appeal against a judgment of the Supreme Court or the High Administrative Court to be filed with the Constitutional Court. As a result, diverging jurisdiction on similar issues including the interpretation and application of EU law, between ordinary courts and the Constitutional Court or the Administrative Courts cannot entirely be ruled out. However, regarding EU law, specifically with respect to the core features of the relationship between EU law and national law, that has never been of major importance.14

II.  Constitutional Culture The Austrian Constitution structures the polity as a federal system. The respective legislators, one for the Federation, nine for the provinces, enjoy a high level of political leeway. These legislators are the ‘problem solvers’, the federal legislator being the most powerful among them. It is the most powerful because of the clear preponderance the Constitution guarantees to the Federation in terms of competencies and the weak position of the second legislative chamber at federal level, the Federal Council (Bundesrat). The latter consists of deputies from the level of the provinces but disposes only of a suspensive veto power, in general. A strong federal administration is not only implementing legislation, but also preparing legislative amendments, while control of both the legislator and the administration lies with the courts.

11 Art 129 B-VG. 12 Art 133 para 4 B-VG. 13 The VfGH, called upon by individuals, at the occasion of appealing against the decision of an ordinary court of first instance, alleging the infringement of their rights through applying an unconstitutional Act of Parliament, pronounces on the unconstitutionality of that act (Art 140 para 1 lit d B-VG, so called Gesetzesbeschwerde). Note that a similar situation exists regarding the 11 Administrative Courts (Art 144 B-VG). By contrast, such an appeal exists neither regarding judgments of the High Administrative Court, nor the Supreme Court. Consequently, these two Highest Courts are, strictly speaking, not subordinated to the Constitutional Court. Generally, however, any court, including the last-mentioned highest courts, suspecting the unconstitutionality of an Act of Parliament, has to file an application with the VfGH aiming at the nullification of that Act (Art 89 para 2 B-VG). But there is no formalised remedy if they should not address the VfGH. 14 There has been, by contrast, some controversy on who is the guardian of EU fundamental rights protection in Austria. Compare below in the text following n 111.

556  Stefan Griller The attitude over the decades has rightly been characterised as legalistic in the sense that each and every problem tends to be referred to and then addressed by an Act of Parliament.15 This is not withstanding that many of these ‘solutions’ are being negotiated and settled outside the parliamentary fora, not the least in negotiations between the social partners, or between the provinces’ governors (Landeshauptleute), or at least including them. The respective Act of Parliament sometimes comes near a mere ratification of these extra-parliamentary processes. Unresolved differences happen to disappear in ambiguous wording, which is later instrumentalised by both sides. It is not an exception that, under these circumstances, conflict resolution is sometimes ‘delegated’ to the courts, which decide one way or the other. This in turn may lead to another legislative reaction, thus upholding the momentum in the ‘language game’ between the constitutional actors. Given the relatively low hurdle for the production of ‘ordinary’ constitutional Acts of Parliament, compromises often find their way into the Constitution, irrespective of the constitutional significance of the matter. This explains the notorious fragmentation and high number of constitutional Acts. EU membership has rightly been described as the main factor transforming the Austrian Constitution, first by the deliberate and explicit amendment of the fundamental principles, but then by affecting the entire legal system. Direct applicability and primacy of EU law inevitably diminish the weight of national legislation, including the Constitution. This leads to a heightened importance of the courts which are competent to decide which provisions would apply and clarify the matter in preliminary reference proceedings before the European Court of Justice (CJEU). At the same time, these phenomena undermine the traditional reviewing and setting aside of general norms, formerly a monopoly of the Constitutional Court. Altogether, EU membership renders the Austrian legal order less hierarchical, and decisions of the courts less predictable.

III.  Constitutional Foundations of EU Membership and Closely Related Instruments A.  Total Revision of the Austrian Constitution As a constitutional prerequisite of EU membership, the Austrian Constitution had, as already mentioned, to be totally amended. According to widespread consent, EU membership would bring about a fundamental change to several of the fundamental principles of the Constitution, namely democracy, separation of powers, the rule of law, and the federal state principle.16 15 Compare for the entire subject, and with further references Theo Öhlinger, ‘Die Europäisierung des österreichischen Rechts’, in P Luif (ed), Österreich, Schweden, Finnland. Zehn Jahre Mitgliedschaft in der Europäischen Union (Vienna/ Cologne/Weimar, Böhlau, 2007) 113. 16 Compare the following eg Stefan Griller, ‘Verfassungsfragen der österreichischen EU-Mitgliedschaft’, (1995) Zeitschrift für Europarecht, Internationales Privatrecht und Rechtsvergleichung 89; Stefan Griller, ‘Individueller Rechtsschutz und Gemeinschaftsrecht’, in J Aicher, M Holoubek and K Korinek (eds), Gemeinschaftsrecht und Wirtschaftsrecht. Zentrale Probleme der Einwirkung des Gemeinschaftsrechts auf das österreichische Wirtschaftsrecht (Vienna, Orac, 2000) 27; K Lachmayer, ‘The Constitution of Austria in International Constitutional Networks: Pluralism, Dialogues and Diversity’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (The Hague, T.M.C. Asser Press, 2019) 1271.Theo Öhlinger, Verfassungsfragen einer Mitgliedschaft zur Europäischen Union (Vienna, Springer, 1999); Theo Öhlinger and Michael Potacs, EU-Recht und nationales Recht, 7th edn (Vienna, LexisNexis, 2020); Heinz Schäffer, ‘Österreichischer Landesbericht’, in J Schwarze (ed), Die Entstehung einer europäischen Verfassungsordnung (Baden-Baden, Nomos, 2000) 339; Roland Winkler, Integrationsverfassungsrecht. Das österreichische Verfassungsrecht und das Recht der Europäischen Union – Koordination, Kooperation, Konflikt (Vienna/New York, Springer, 2003). For contributions in English, compare Stefan Griller, ‘Introduction to the Problems in the Austrian, Finnish and

Austria  557 This was because of the widely discussed fundamental changes in the production of binding law. Far-reaching powers of law-making are being transferred from the national legislator to the organs of the EU (at the time, the European Communities), which may enact acts that are directly applicable within the Austrian legal order, both at federal and provincial level and take precedence over conflicting national law. Thus, the Act of Parliament is dismantled of its dominant position as the means of general law-making and its highest rank within the legal system. Thereby, also the roles of the Executive and the judiciary are fundamentally altered, amongst others, the Constitutional Court losing its monopoly of scrutinising ordinary legislation against the yardstick of higher-ranking law. Against this background, the Austrian Parliament passed a bill which was the subject of a referendum before entering into force and the signing of the Treaty of Accession. The referendum took place in June 1994 and produced a majority of more than 66 per cent of positive votes. The result of this procedure, the Federal Constitutional Act on Austria’s Accession to the EU17 (the so called EU Accession Act), reads as follows:18 Based on the results of the referendum it is hereby promulgated: • Article I: With the approval of the Austrian people to this Federal Constitutional Act the competent organs under the federal constitution are authorised to conclude the Treaty on Austria’s Accession to the EU according to the results of the accession conference agreed upon on April 12, 1994. • Article II: The Treaty on Austria’s accession to the EU may only be concluded upon approval of the Nationalrat and the Bundesrat. These decisions require the presence of at least half of the respective members and a two-thirds majority of the votes cast. • Article III: The Federal Government is in charge of implementing this Federal Constitutional Act.

It has been contended that this text hides more than it reveals. This is certainly true insofar as it does not make clear that the referendum (the ‘approval of the Austrian people’) was mandatory and why this was the case. Swedish Constitutional Order’, in A Kellerman, J de Zwaan and J Czuczai (eds), EU Enlargement: The Constitutional Impact at EU and National Level (The Hague, Asser Press, 2001) 147; Klaus Kröll and Georg Lienbacher, ‘Austria’, in European Parliament, DG for Internal Policies (ed), National Constitutional Law and European Integration (Brussels, European Parliament, 2011) 141; Alina Lengauer, ‘First Experiences with EU-membership – the Case of Austria’, in V Kandžija et al (eds), Economic System of European Union and Adjustment of the Republic of Croatia (Rijeka/Trieste, Faculty of Economics, 1997) 539; Rainer Palmstorfer, ‘Austria’, in S Griller and E Lentsch (eds), EMU Integration and Member States’ Constitutions (Oxford, Hart, 2021) 523; Ignaz Seidl-Hohenveldern, ‘Constitutional Problems involved in Austria’s Accession to the EU’, (1995) 32 CML Rev 727. 17 Bundesverfassungsgesetz über den Beitritt Österreichs zur Europäischen Union, BGBl (Federal Law Gazette) 1994/744. It is one of the mentioned leges fugitivae outside the text of the B-VG, which might be seen as characteristic: the most important legal change to the Constitution since its creation was enacted outside the text of the B-VG. 18 Translation by the author. The original version of this brief Act is as follows: ‘Bundesverfassungsgesetz über den Beitritt Österreichs zur Europäischen Union Auf Grund des Ergebnisses der Volksabstimmung wird kundgemacht: Artikel I Mit Zustimmung des Bundesvolkes zu diesem Bundesverfassungsgesetz werden die bundesverfassungsgesetzlich zuständigen Organe ermächtigt, den Staatsvertrag über den Beitritt Österreichs zur Europäischen Union entsprechend dem am 12. April 1994 von der Beitrittskonferenz festgelegten Verhandlungsergebnis abzuschließen. Artikel II Der Staatsvertrag über den Beitritt Österreichs zur Europäischen Union darf nur mit Genehmigung des Nationalrates und der Zustimmung des Bundesrates hiezu abgeschlossen werden. Diese Beschlüsse bedürfen jeweils der Anwesenheit von mindestens der Hälfte der Mitglieder und einer Mehrheit von zwei Dritteln der abgegebenen Stimmen. Artikel III Mit der Vollziehung dieses Bundesverfassungsgesetzes ist die Bundesregierung betraut.’ For an official, yet not binding translation into English compare https://www.ris.bka.gv.at/Dokument.wxe?Abfrage= Erv&Dokumentnummer=ERV_1994_744.

558  Stefan Griller Either way, the legal effects of this authorisation of the total revision of the Austrian Constitution by concluding the Treaty of Accession19 may be summarised as follows: • Austria’s accession to the EU was enacted on the basis of a specific constitutional Act of the highest possible rank within the Austrian legal order. • The authorisation for this step was deliberately narrow: it covered only the conclusion of the Treaty of accession according to the results of the accession conference in April 1994. Any later alterations of the drafted text would have been excluded from approval and ratification. Furthermore, the authorisation does not cover later amendments of the EU legal order, ­especially not those agreed upon during intergovernmental conferences.20 • The EU Accession Act only implicitly tackled the crucial issue of the relation between EC (and EU) law and Austrian law, by referring to the Draft Treaty of Accession and the Act of Accession which included, inter alia, the obligation to accept the acquis communautaire.21 As will be discussed further, this leaves several core aspects open to controversy. It might be concluded that the Austrian way into the EU on the grounds of a Constitutional Act of the highest possible rank, appears to be unique among all EU members.22 As we shall see, this prevented several obstacles to proper implementation of Community law in Austria which occurred in other countries. Austria’s accession to the EU entailed the most fundamental change to the constitutional order since 1918, which was the starting point of the continuous development of the Austrian constitutional order.23 The Austrian Parliament also passed flanking constitutional measures, mainly providing for the participation of Austrian organs in EU decision-making (including the Common Foreign and Security Policy (CFSP)), for the election of Austrian members to the European Parliament, and for the appointment of members to the European Commission, the ECJ and the Court of First Instance, the European Court of Auditors, the board of the European Investment Bank, the Economic and Social Committee, and the European Committee of the Regions.24

B.  Subsequent Changes by ‘Simple’ Constitutional Acts of Parliament An important and direct consequence of this way into the EU – specifically tailored for and limited to the accession treaty, but not addressing any subsequent developments – was that every

19 It is submitted that the changes were not brought about directly by the EU Accession Act, but by the conclusion of the Treaty of Accession on the grounds of the EU Accession Act. Even if, politically, it would have been out of the question not to approve or ratify the accession treaty, that would nevertheless have been legal. 20 Art 46 (ex Art N) TEU Amsterdam; today Art 48 TEU. 21 On these issues, nothing can be found either in the flanking constitutional measures adopted to ease the transition to EU membership (see below n 24). 22 The Austrian Chancellery could rely on a comparative study on the constitutional foundations of membership in the then 12 EC Member States, which was undertaken as a part of the preparations for membership – see Stefan Griller, Franz Maislinger and Andreas Reindl, Fundamentale Rechtsgrundlagen einer EG-Mitgliedschaft. Verfassungsfragen der Übernahme von EG-Recht in den bisherigen Mitgliedstaaten in vergleichender Sicht (Vienna, Österreichische Staatsdruckerei, 1991). It might be concluded that at least in parts the specific steps taken were a deliberate departure from the previous examples. 23 With the exception of the Nazi regime from 1938–45. After 1945, the constitutional order was ‘reset’ to the state of the law in 1933 (before the constitutional amendments creating the so called Ständestaat, the authoritarian period preceding the Nazi regime). 24 Arts 23a–23f B-VG. Compare eg Christoph Grabenwarter, ‘Änderungen der österreichischen Bundesverfassung aus Anlaß des Beitritts zur Europäischen Union’, (1995) 55 Zeitschrift für ausländisches öffentliches Recht und Völkerrecht 166;

Austria  559 future amendment of the Treaties was difficult to judge in terms of its constitutional ratification requirements. In particular, every approval and subsequent ratification of such amendments without constitutional majorities would have risked the allegation of violating the Constitution. Consequently, in the cases of the Amsterdam Treaty (1997) and the Nice Treaty (2001), but also regarding the 2003 and 2006 Enlargement Treaties,25 the Government again decided to propose the approval and the ratification of the Treaty on the basis of special Constitutional Acts.26 By contrast to the Accession Treaty, these Constitutional Acts (rightly) were not treated as total revisions of the Constitution, meaning that no referendum was convened. But like the EU Accession Act, they necessitated a qualified approval of the respective Treaties (Amsterdam, Nice and enlargement) by the Austrian Parliament. The approval by the Nationalrat and the Bundesrat required, as in the EU Accession Act, the presence of at least half of the respective members and a two-thirds majority of the votes, which is the majority usually required for amendments of the Constitution.27 Unlike for accession itself, an explicit derogation from the labelling duty for of all constitutional amendments was granted to these later changes.

C.  Creating A Permanent Basis for Treaty Changes: The Constitutional Reform of 2008 In 2008, the constitutional legislator undertook an effort to reduce the enormous number of constitutional provisions outside the text of the B-VG (by dismantling them of their constitutional rank), and at the same time the constitutional reasons for creating similar provisions in the future. This was done by a Constitutional Act amending the B-VG and enacting the First Federal Constitutional Simplification Act.28 As a part of this exercise, Article 50 B-VG was amended in order to allow for future primary EU law amendments without inserting fresh individual authorisations (as leges fugitivae) into the Austrian constitution each time. The result was a general provision modelled after the mentioned practice since EU accession. Article 50 B-VG is the provision traditionally incorporating international treaties as a part of domestic law equalling Acts of Parliament. Hence, it allows for the approval of primary law Gerhart Holzinger, ‘Die Auswirkungen der österreichischen EU-Mitgliedschaft auf das österreichische Verfassungsrecht’, (1996) 4 Journal für Rechtspolitik 160. 25 Art I Bundesverfassungsgesetz über den Abschluß des Vertrages von Amsterdam, BGBl I 76/1998, Art I Bundesverfassungsgesetz über den Abschluss des Vertrages von Nizza, BGBl. I 120/2001, Art I des Bundesverfassungsgesetzes des Vertrages über den Beitritt der Tschechischen Republik, der Republik Estland, der Republik Zypern, der Republik Lettland, der Republik Litauen, der Republik Ungarn, der Republik Malta, der Republik Polen, der Republik Slowenien und der Slowakischen Republik zur Europäischen Union, BGBl I 53/2003, Art 1 Bundesverfassungsgesetz über den Abschluss des Vertrages über den Beitritt der Republik Bulgarien und Rumäniens zur Europäischen Union, BGBl I 25/2006. The same applied to the Treaty Establishing a Constitution for Europe which, due to negative referenda in the Netherlands and France, was never ratified: BGBl I 12/2005. 26 A certain similarity seems to result regarding the traditional Irish way, even if the rank of the authorising provision is of course different. Compare the contribution of Gavin Barret in this volume. 27 At least for the Nationalrat. The consent of the Bundesrat by two-thirds majority is, according to Art 44 para 2 B-VG, only required in cases of diminishing the competences of the provinces. Otherwise the Bundesrat disposes only of a suspensive veto, as in general. 28 Bundesverfassungsgesetz, mit dem das Bundes-Verfassungsgesetz geändert und ein Erstes Bundesverfassungsrechtsbereinigungsgesetz erlassen wird, BGBl I 2/2008. In the meantime, this was complemented by a second Federal Simplification Act (an ordinary, not a constitutional Act) that, in principle, lists all Acts (also ordinances) that were promulgated before 1 January 2000 and continue to be valid after 31 December 2018: Bundesgesetz betreffend die Bereinigung von vor dem 1. Jänner 2000 kundgemachten Bundesgesetzen und Verordnungen (Zweites Bundesrechtsbereinigungsgesetz – 2. BRBG), BGBl I Nr 61/2018.

560  Stefan Griller amendments including EU accession by a two-thirds qualified majority in both chambers of Parliament: (1) The conclusion of

1. Political state treaties and state treaties the contents of which modify or complement existing laws … , as well as 2. State treaties by which the Treaty bases of the European Union are modified, requires the approval of the National Council. … (4) Notwithstanding Article 44 para 329 state treaties according to para 1 subpara 2 may only be concluded with the approval of the National Council and the consent of the Federal Council. These resolutions each require the presence of at least half of its members and the majority of two thirds of the votes cast.30

The new provision was for the first time applied for approving the Treaty of Lisbon,31 followed by the EU accession of Croatia.32 The revised Article 50 B-VG in principle rules out the incorporation of treaties into the domestic legal order ranking them as constitutional law.33 Any such amendment of the Constitution would have to be prepared by a Constitutional Act of Parliament. Importantly, there is an exception for primary EU law. Approving EU treaty changes requires the ‘normal’ two-thirds constitutional majority, barring the labelling requirement, and applies irrespective of the contents of such changes. Thus, they might even encompass implicit amendments to the Austrian Constitution. This is different for total revisions of the Constitution. Such revisions would have, even in the case of the EU, to be prepared by a Constitutional Act of Parliament subject to a referendum. It was controversial whether the Treaty of Lisbon would require another total revision. The prevailing view answered the question in the negative.34

D.  EU Integration Outside the EU Legal Order: ESM, Fiscal Compact Both the EU Accession Act and Article 50(1) section (2) of the Constitution cover only primary EU law, the former as it stood at the time of Austria’s accession, the latter in the form of ‘treaties

29 This

is the provision on the total revision of the Constitution (requiring a referendum). 50: ‘(1) Der Abschluss von 1. politischen Staatsverträgen und Staatsverträgen, die gesetzändernden oder gesetzesergänzenden Inhalt haben und nicht unter Art. 16 Abs. 1 fallen, sowie 2. Staatsverträgen, durch die die vertraglichen Grundlagen der Europäischen Union geändert werden, bedarf der Genehmigung des Nationalrates. … (4) Staatsverträge gemäß Abs. 1 Z 2 dürfen unbeschadet des Art. 44 Abs. 3 nur mit Genehmigung des Nationalrates und mit Zustimmung des Bundesrates abgeschlossen werden. Diese Beschlüsse bedürfen jeweils der Anwesenheit von mindestens der Hälfte der Mitglieder und einer Mehrheit von zwei Dritteln der abgegebenen Stimmen. (5) …’. 31 BGBl III 132/2009. 32 BGBl III 171/2013. 33 Compare for the following Kröll and Lienbacher, ‘Austria’ (n 16) 147 ff, with further references to the controversial debate. It is also being argued that it would be possible to submit Treaties amending the EU legal bases to a referendum in case of a total revision of the constitution. 34 Compare the debate and the references in the explanations of the government attached to the submission of the Lisbon Treaty for approval to the Austrian Parliament, 417 Beilagen zu den Stenographischen Protokollen des Nationalrats, XXIII. GP, 43 ff. 30 Art

Austria  561 by which the Treaty bases of the European Union are modified’. This narrow wording35 does not capture integration instruments outside the EU legal order such as the European Stability Mechanism (ESM) or the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union (TSCG; aka Fiscal Compact). As a consequence, treaties of that kind come under the first Section of the mentioned Article 50(1) of the Constitution: ‘1. Political state treaties and state treaties the contents of which modify or complement existing laws’. Such treaties fall under the provisions governing the enactment of simple legislation, requiring simple majority in the National Council and being subject to the suspensive veto of the Federal Council.36 Should such treaties entail amendments of the Austrian Constitution, they cannot simply be adopted by a constitutional majority. In such cases a Constitutional Act of Parliament would have to be passed before approving and ratifying the treaty, which could then be done by a simple majority. In both of the above-mentioned cases, the federal government decided to go for a simple majority, claiming that no constitutional changes were involved. Parliament followed this stance, as did the Constitutional Court, which was seized to review them. As a result, both treaties were approved by Parliament by a simple majority and ratified accordingly.37

IV.  Constitutional Limits to EU Integration A.  Conceptualising ‘Unconditional Membership’: The Relation between EU Law and Austrian Law 1.  The EU Accession Act and Subsequent Transfer of Powers Unlike other EU members, during the accession procedure, Austria did not adopt any constitutional reservations regarding EU membership. By contrast, the reasons attached to the draft EU Accession Act make it abundantly clear that the intention was to accept unconditionally the principles developed by the ECJ governing the relation between EC law (today: EU law) and national law. It was explicitly stated that the EU Accession Act should guarantee the opening up of the Austrian legal order to the EU legal order according to its specific requirements, especially including primacy and direct effect.38 Thus, the prevailing view is that the EU Accession Act does not, unlike eg Article 23 of the German Basic Law, include any specific constitutional reservations regarding the implementation and enforcement of primary or secondary EC law at the time of accession.39 As long as EC (and EU) law is within the boundaries of the ‘results of the accession conference’ of 1994, every alteration of the Austrian legal system was accepted in advance.

35 Being deliberately narrower than that of Art 23 para 2 German Basic Law (Grundgesetz) calling for the two Chambers of Parliament to ‘participate in matters of the European Union’. See for this Matthias Jestaedt in this volume. 36 Art 50 para 3 B-VG. 37 ESM: BGBl III 138/2012; TSCG: BGBl III 17/2013. For the respective controversy, see below in the text near n 66. 38 See the explanatory memorandum of the government when it submitted the draft EU-Accession-Act to Parliament for approval: 1546 BlgNR (attachments to the minutes of the National Council) XVIII. GP (legislative period) 4. 39 Eg Griller, ‘Verfassungsfragen’ (n 16) 99 f; Öhlinger, Verfassungsfragen (n 16) 180 ff; Theo Öhlinger, ‘Staatlichkeit zwischen Integration und Souveränität’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 111, 121 ff. For a different reading compare eg Andreas Janko, Gesamtänderung der Bundesverfassung (Vienna, Verlag Österreich, 2004) 412 ff.

562  Stefan Griller Clearly, this is different when it comes to subsequent treaty amendments like those of Amsterdam, Nice and Lisbon.40 The EU Accession Act is silent in this respect. Article 50(4) of the Constitution makes it crystal clear that another total revision would need another referendum. However, it is far from being clear which changes would necessitate such a step. In any event, the yardstick must be the fundamental principles of the constitution in their already fundamentally changed version after EU accession. Direct effect or primacy as such cannot be such a reason for the second time. Consequently, the quest has to be for substantive changes like the additional transfer of competences, blows to the level of fundamental rights protection, democratic decision-making, separation of powers, or undermining statehood. However, neither the academic debate nor court decisions have so far clarified this issue.41 Specifically, no effort has been made similar to that by the German Constitutional Court to identify in abstracto core subjects of national competences, the transfer of which would render the respective Treaty unconstitutional.

2. Theory The ‘Austrian way’ provoked a debate on underlying concepts that might be of general interest and arguably influenced case law at every possible level, even if the courts are cautious not to express themselves on the matter. A first possible interpretation of the reported constitutional changes is that the Austrian Constitution was open to the incorporation of EU law, including amendments of the traditional rules governing the relation of municipal law and international law. These alterations, however, would be limited by the need to fulfil the relevant membership duties of abiding by EU law. The alterations would, therefore, not go beyond what is necessary to reach that goal.42 Consequently, the Accession Treaty would have to be considered as a treaty transformed into the Austrian legal order, the legal effects of which are not determined by the traditional constitutional rules for international Treaties, but by the jurisprudence of the CJEU. Secondary EU law would form a part of Austrian law, its legal effects to a large extent also being determined by EU law. An alternative view is that Austria’s accession to the EU is, in terms of the consequences for the legal system, to be treated as an accession to a federal state.43 Consequently, the Accession Treaty is to be seen as both a treaty and similar as the founding Act of a federal state. Consequently, there cannot be any ‘transformation’ of the Accession Treaty – and, consequently, of the acquis communautaire and future EU legislation – into the Austrian legal order. Moreover, the legal effects of Union law in Austria are in no way influenced by (modified) principles on the relation between municipal and international law, but solely by the ‘autonomous’ legal order of the EU. Even if

40 Ewald Wiederin, ‘Der Vorrang des Unionsrechts vor nationalem Recht’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 179, 190 ff. 41 Most tellingly, the substantive additional competence transfers the Lisbon Treaty brought about – namely integrating the former third pillar, the so-called Area of Freedom, Security and Justice (Arts 67 ff TFEU), into the realm of supranational law-making, and the extension of the Common Commercial Policy (Art 207 TFEU), an exclusive EU competence, to agreements relating also to trade in services, the commercial aspects of intellectual property, and foreign direct investment, in a rather comprehensive manner – did not raise any concern in Austria regarding an alleged second total revision. 42 Argued in Stefan Griller, ‘Der Stufenbau der österreichischen Rechtsordnung nach dem EU-Beitritt’, (2000) 8 Journal für Rechtspolitik 273, 278 ff. Traces of this position can be found in the early jurisprudence of the VfGH: VfGH 16.06.1998, VfSlg 15.189/1998 ECLI:AT:VFGH:1998:V6.1998; VfGH 24.06.1998, VfSlg 15.215/1998 ECLI:AT:VFGH:1998:G2.1997 (selected post-2012 VfGH decisions are available in English at www.vfgh.gv.at/rechtsprechung/Ausgewaehlte_ Entscheidungen.en.html). 43 Argued in Öhlinger, Verfassungsfragen (n 16) 194, 201 ff. Later, however, he modified this position: compare now Öhlinger, ‘Staatlichkeit’ (n 39) 120 ff.

Austria  563 today this alternative view certainly is not the prevailing one, the fact that it was reflected upon by one of the most prominent scholars gives evidence that even the transformation of the EU into a Federal State could be dealt with under the Austrian Constitution by performing another total revision. A third reading put forward in the debate is that EU law and national law are two distinct legal systems, to be treated in the sense of a dualist understanding,44 sometimes labelled as a ‘double constitution’.45 EU law and national law are founded in two completely separated legal grounds (Geltungsgründe), and not by one common legal fundament. This position arrives at a complete alteration of the legal situation. The conclusion to be drawn is that EU law is not ‘part of the law of the land’ in Austria. Due to the dualist construction, EU law forms a legal order of its own. Also, the Treaty of Accession has not been transformed into the Austrian legal order, which would be the same for the acquis communautaire and presumably also for future secondary legislation. However, it is hard to see how to resolve possible conflicts between EU law and national law on the ground of this ‘dualist approach’, except if the EU Accession Act could be read as incorporating the ‘second’, the EU legal order, into the Austrian legal system. At first sight, all this might appear a purely academic debate. However, it can be argued that the consequences of EU membership especially for the functions of the system of judicial protection enshrined in the Austrian Constitution – to be discussed below – are dependent on the underlying theoretical approach.

B.  Scrutiny of Secondary Legislation, Especially Ultra Vires Doctrine In academic writing it is widely accepted that scrutiny of secondary EU legislation by Austrian Courts cannot, on the grounds of the EU Accession Act or Article 50 of the Constitution, be entirely ruled out.46 However, given the responsibility of the CJEU in enforcing the respect of primary law, when it comes to secondary legislation this is seen as a rather theoretical debate. It has never – unlike in other EU Member States – gained momentum in the jurisprudence of the courts. The reason for this ‘constitutional restraint’ lies in the mentioned fact that Austria’s EU accession was without reservation and performed by a constitutional Act of the highest possible rank. The remaining limits for secondary legislation consequently flow from what had been approved through accession, and later in the treaties of Amsterdam, Nice and Lisbon. The powers of the EU are specified and at the same time limited in these instruments. The waiver of any control of those limits would render them irrelevant constitutionally. Thus, the yardstick for such a residual control competence would not be some core contents or principles of the Austrian Constitution, but the limits agreed upon in the EU founding treaties. In this vein, the remaining, very limited option is to question the legality of EU law in cases of apparent and severe ultra vires acts. Such measures might then be considered null and void for the purpose of the Austrian legal order.

44 Argued in Bernd-Christian Funk, ‘Rechtserzeugung heute – Rechtserzeugung im 21. Jahrhundert’, in M Holoubek and G Lienbacher (eds), Rechtspolitik der Zukunft – Zukunft der Rechtspolitik (Vienna/New York, Springer, 1999) 45, 49 ff; compare also Ludwig Adamovich, Bernd-Christian Funk, Gerhart Holzinger and Stefan Leo Frank, Österreichisches Staatsrecht I, 2nd edn (Vienna/New York, Springer, 2011) paras 01.045, 17.060. 45 Doppelverfassung: eg Öhlinger, ‘Staatlichkeit’ (n 39) 119f. 46 See for the following eg Griller, ‘Verfassungsfragen’ (n 16) 99 f; Wiederin, ‘Vorrang’ (n 40) 194 f; a slightly modified stance – allowing for ‘fundamental principles scrutiny’ – is now expressed in Öhlinger, ‘Staatlichkeit’ (n 39) 123 ff.

564  Stefan Griller It goes without saying that thereby Austrian constitutional principles would at the same time be ‘protected’ indirectly, simply because of the EU lacking the competence to modify them beyond what had been agreed upon in the founding instruments.47

C.  A Specific Feature: Permanent Neutrality 1.  Neutrality under the TEU and the TFEU? It is still disputed whether or not neutrality is compatible with EU membership.48 The dispute concerns Austria, which prior to EU accession was considered a permanently neutral country. Finland, Sweden and Ireland were deemed to conduct a consistent policy of neutrality.49 It shall only be summarised briefly here why good reasons suggest that neutrality in both variants is incompatible with EU membership.50 The legal rules of neutrality as enshrined in international law include several duties for a neutral state: duties of abstention, especially the prohibition on participating in a war between other states; duties of prevention, eg concerning the passage of foreign troops and territorial overflights; and duties of equal treatment, especially concerning possible restrictions on trade of war materials, and – a controversial issue – possible restrictions on trade in goods other than war material. The latter are subject to scrutiny not under a strict equal treatment requirement, but under the prohibition to intervene in a war (eg by unilateral embargo measures). Permanently neutral states,51 such as Austria, have to respect, in addition, the so-called pre-effects (Vorwirkungen) of neutrality, ie they are required, even in times of peace, to work towards preventing their potential involvement in future wars. For example, according to its Neutrality Act, Austria may not join military alliances, nor may it allow the establishment of foreign military bases on its territory. Membership in the EU and the EC has, for some time, been regarded as being inconsistent with neutrality by Europe’s neutral states. In the 1950s and 1960s, all neutral EFTA countries (four of which are now EU members, plus Switzerland) saw neutrality as a major obstacle to EEC membership. However, in the academic debate, it has always been controversial whether neutrality and EC, and later, EU membership, are reconcilable. The crucial issue is whether the provisions on the CFSP, and the corresponding provisions (today) in the TFEU, authorise Member States to remain neutral in the event of a conflict between third states or between a third state and a Member State of the Union, or the Union itself and a third state. As a matter of principle, it appears reasonable, at first sight, to answer this question in the affirmative, for the following three reasons. First, there is the principle of unanimity,52 which renders possible the

47 Eg dissolving the Austrian Parliament through secondary legislation would be a clear ultra vires act. Disrespect would enforce the limits of the powers conferred, and at the same time the fundamental principle of democracy in Austria against changes which had not been agreed upon. 48 For a recent account and further references compare Theo Öhlinger, ‘Österreichs Neutralität in der Europäischen Union’, (2018) 73 Zeitschrift für Öffentliches Recht, 621. 49 That could also be characterised as a policy of non-alliance aiming at neutrality in the case of war between other states. 50 For a more detailed discussion compare Stefan Griller, Dimitri Droutsas, Gerda Falkner, Katrin Forgó and Michael Nentwich, The Treaty of Amsterdam (Vienna/New York, Springer, 2000) 431–40; recently Öhlinger ‘Neutralität’ (n 48). For a Swedish perspective compare Göran Lysén, ‘Some Views on Neutrality and Membership of the European Communities: The Case of Sweden’, (1992) 29 CML Rev 229. 51 These are states that are bound, under international law, to remain neutral in future wars between subjects of international law. 52 Art 31 TEU.

Austria  565 blocking of decisions that would be inconsistent with neutrality. Second, according to the TEU, ‘The policy of the Union … shall not prejudice the specific character of the security and defence policy of certain Member States.’53 This so-called ‘Irish clause’ aims to create a shield for neutrality. Third, one could have recourse to the safeguard clauses embodied in the TFEU,54 eg in a case where embargo measures are implemented by secondary EU law. However, while political evaluation of each case, including a possible veto, is certainly in line with the Treaties, the same does not seem to be true with regard to a general reservation on grounds of neutrality (or on other grounds). Mutual solidarity and the absence of reservations, as foreseen by the TEU,55 appear to render impossible a general reservation, based on whatever national policy principles, in the field of the CFSP. With regard to Finland, Sweden and Austria, a Joint Declaration in the Final Act to the Accession Treaty56 makes this principle abundantly clear. Declaration No 1 on Common Foreign and Security Policy, which – according to Article 31(2)(b) of the Vienna Convention on the Law of Treaties – is relevant for the interpretation of the TEU, states inter alia that the Union and the EFTA states agree that the new Member States will, from the time of their accession, be ready and able to participate fully and actively in the Common Foreign and Security Policy as defined in the Treaty on European Union; the new Member States will, on accession, take on in their entirety and without reservation all the objectives of the Treaty, the provisions of Title V thereof, and the declarations attached to it.

Systematic blocking of all measures inconsistent with neutrality (especially of unilateral embargo measures against belligerent states) as well as insisting, without exception, on being exempted from such measures or not having to participate, certainly cannot be regarded as ‘participat[ing] fully’ in the CFSP, nor as ‘tak[ing] on in their entirety and without reservation’ the provisions of the TEU. Certainly, there is no obligation to participate in every, or not to block any, action that is irreconcilable with neutrality. It is only a systematic obstruction of the adoption of measures, for the sole reason that they are inconsistent with neutrality, which has to be regarded as prohibited. At the time of EU accession, the resulting effects on neutrality could be seen as affecting economic duties of neutrality only.57 The new status of the European neutrals after Maastricht could be described, therefore, as one of mere military or differential neutrality. The Amsterdam Treaty, by paving the way for the so-called Petersberg tasks,58 introduced a major military dimension into the CFSP. Thus, the obligation relating to solidarity discussed above now covers a wider range of measures. These issues go even more to the substance of the law of neutrality than the modification of economic obligations. It is the general obligation to stay neutral in a situation of war that is the essential element of neutrality. It is this specific aspect that has been discarded, not only in the context of economic sanctions, but also, quite generally, within CFSP and especially within the framework of the Petersberg tasks. Systematic blockage of such measures for the sole reason of neutrality – which would be the duty of a neutral state – would be, as noted, illegal.

53 Art 42 para.2(2) TEU. The text continues as follows: ‘… and shall respect the obligations of certain Member States, which see their common defence realised in the North Atlantic Treaty Organisation (NATO), under the North Atlantic Treaty and be compatible with the common security and defence policy established within that framework’. 54 Arts 346 and 347 TFEU. 55 Arts 3, 21, 24, 31, 32 TEU; see also Art 222 TFEU. 56 [1994] OJ C 241/21 as amended by [1995] OJ L 1/1. 57 And it should be remembered that such duties were disputed as a part of the law on neutrality, at least as regards goods other than war material. That made the issue less spectacular. 58 Art 43 TEU.

566  Stefan Griller As a consequence, and summing up, it is submitted that both the concept of permanent neutrality and a constant neutrality policy are inconsistent with EU membership.

2.  Constitutional Amendments in Austria Austria’s permanent neutrality is (or better, as argued here: had been) an obligation by virtue of a special constitutional law enacted in 1955.59 Austria explicitly amended its Constitution – but not its Neutrality Act(!) – several times with regard to CFSP matters: first, in the course of EU accession, and, second and third, during the ratification procedure relating to the Amsterdam Treaty, and after the Lisbon Treaty. The Constitution now explicitly allows for Austria’s participation not only in economic sanctions against third countries, but also in Petersberg actions. After the third amendment, the crucial passage of the relevant Article of the Constitution60 now reads as follows: Austria participates in the Common Foreign and Security Policy of the European Union enshrined in title V chapter 1 and 2 of the Treaty on European Union as amended by the Treaty of Lisbon … This includes participating in tasks according to Article 43 para 1 of this Treaty and in measures which interrupt or reduce, in part or completely, economic relations with one or more third countries …61

These sentences have to be regarded as derogation from the constitutional commitment to permanent neutrality. However, and mainly for political reasons, explicitly amending the Constitutional Act on Austria’s neutrality was avoided.62 Consequently, the legal situation lacks clarity. Furthermore, there are still some remainders of neutrality to be found.63 Above all, the unchanged parts of the Neutrality Act in connection with the amendments mentioned do not authorise departures from neutrality commitments except in the CFSP framework, ie not in other cases. Moreover, the establishment of a common defence, and consequently membership of a military alliance, remain prohibited. Yet, one decisive aspect has been abandoned, ie the principle of non-intervention in a war, which hitherto has been a core aspect of neutrality. The ‘rest’ is not enough, therefore, to be considered as permanent neutrality.

59 Bundesverfassungsgesetz vom 26. Oktober 1955 über die Neutralität Österreichs, BGBl 1955/211. The following remarks concentrate on the constitutional law aspects and leave aside the equally delicate issue of an eventual modification of Austria’s obligation under international law to stay neutral. 60 Art 23 j B-VG. 61 Translation by the author. In German it reads as follows: ‘Österreich wirkt an der Gemeinsamen Außen- und Sicherheitspolitik der Europäischen Union auf Grund des Titels V Kapitel 1 und 2 des Vertrags über die Europäische Union in der Fassung des Vertrags von Lissabon mit … Dies schließt die Mitwirkung an Aufgaben gemäß Art. 43 Abs. 1 dieses Vertrags sowie an Maßnahmen ein, mit denen die Wirtschafts- und Finanzbeziehungen zu einem oder mehreren Drittländern ausgesetzt, eingeschränkt oder vollständig eingestellt werden.’ 62 Permanent neutrality had been a hotly debated issue before accession. However, in 1994 the Government and the competent Parliamentary Committee in the debate on the Draft EU Accession Act claimed that Austria could accede to the EU without amending or deleting the Neutrality Act (1600 BlgNR XVIII. GP, 12). Nevertheless, the same document (at 13) provides proof that the repercussions on permanent neutrality were clearly detected, when the Committee stated that possible participation in economic sanctions could be incompatible with Austrian law and should therefore be authorised by a specific constitutional foundation which would not touch upon the ‘core features’ (Kernbestand) of Austrian neutrality. That could only be interpreted as the intention to enact a material derogation of the Neutrality Act without changing the wording of that Act, which was later realised by the insertion of Art 23 f B-VG, which today is the cited Art 23j B-VG. A similarly ambiguous attitude was taken when that article was changed to accommodate the changes to come by the Amsterdam Treaty (to be inferred from the Parliamentary debate: StenProt der 130. Sitzung des NR, 20. GP (Minutes of the 130th Nationalrat-session, 20th legislative period)). Until today there is no clear statement of the Austrian Government acknowledging that development, although it is obvious from the mentioned parliamentary debates that the legal implications for Austria’s neutrality were clearly recognised. 63 See also, similarly, Öhlinger ‘Neutralität’ (n 48) 633 f.

Austria  567 The summarising conclusion is that EU membership and permanent neutrality, as well as a constant policy of neutrality, are incompatible. The ‘European neutrals’, including Austria, are well aware of this fact,64 and have been prepared to participate in the CFSP in the spirit of mutual solidarity and without general reservation. In Austria, this is at the same time the next example for changing an important feature of constitutional law as a sort of ‘side-effect’.

D.  Limits to European Integration Outside the EU Legal Order Article 50 of the Constitution privileges and hampers the transfer of powers to the EU at the same time. On the one hand, amendments to primary EU law need not be expressly designated as constitutional amendments, even if they entail such changes. On the other hand, the constitutional majority of two-thirds of the votes cast in both Chambers of Parliament is required even where no constitutional changes are at stake. Outside the EU legal order, neither of the two features applies. Treaties like the ESM or the TSCG need to pass the constitutionality test. Should they pass it, a simple majority as for ordinary legislation is sufficient for approval. Should they miss it, the Constitution has to be amended by a Constitutional Act of Parliament before the approval of the treaty. The latter is also true in all cases of total revisions, with a mandatory referendum as a complement. This means that any conflict with existing constitutional law, or provisions complementary to that law, trigger the necessity of amending the Constitution before the approval of the treaty. With regard to both the ESM and the TSCG, the controversy was mainly on two points.65 First, would these treaties diminish the ‘budgetary autonomy’ of the Austrian Parliament, assuming that such autonomy is protected by the Constitution? Second, can the transfer of powers to EU organs, as foreseen in these treaties, be covered by Article 9(2) of the Constitution,66 limiting transfers of competences to intergovernmental bodies with ‘single’ competences? Finally, the matter was decided by the Constitutional Court, which upheld the decision of Government and Parliament to approve both Treaties by a simple majority claiming that no constitutional changes were at stake.67 Of major systemic importance is the answer given to the relation of Article 50(1) and (4) of the Constitution on the constitutional requirements for approving primary EU law amendments, and Article 9(2) of the Constitution restricting the transfer of powers to intergovernmental bodies. The matter is specifically salient for the TSCG,68 which foresees the lending of EU organs (the 64 Even if their political reasoning is not always consistent, and public statements far from being precise. 65 Compare for the following, on the ESM: Andreas Hauer, ‘Der ESM-Vertrag auf dem Prüfstand’, in D Ennöckl, N Raschauer, E Schulev-Steindl and W Wessely (eds), Festschrift für Bernhard Raschauer (Vienna, Jan Sramek, 2013) 155; Claudia Mayer, ‘Europäischer Stabilitätsmechanismus nicht verfassungswidrig’, (2013) Österreichische Zeitschrift für Wirtschaftsrecht 53, 70 ff; on the TSCG: Stefan Griller, ‘Zur verfassungsrechtlichen Beurteilung des Vertrags über Stabilität, Koordinierung und Steuerung in der Wirtschafts- und Währungsunion („Fiskalpakt“)’, (2012) 20 Journal für Rechtspolitik 177; Michael Potacs and Claudia Mayer, ‘Fiskalpakt verfassungswidrig’? (2013) 21 Journal für Rechtspolitik 140. For a summary of the controversy, and including the constitutional dimension under EU law, compare Stefan Griller, ‘Die Wirtschafts- und Währungsunion vor, in und nach der Krise’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 791, 838 ff, 849 ff. 66 Art 9 para 2 B-VG: ‘Durch Gesetz oder durch einen gemäß Art. 50 Abs. 1 genehmigten Staatsvertrag können einzelne Hoheitsrechte auf zwischenstaatliche Einrichtungen übertragen werden …’ 67 ESM: VfGH 16.03.2013, VfSlg 19.750/2013 ECLI:AT:VFGH:2013:SV2.2012 (available in English, see fn 42); Fiscal Compact: VfGH 03.10.2013, VfSlg 19.809/2013 ECLI:AT:VFGH:2013:SV1.2013 (available in English, see fn 42). 68 It is less decisive for the ESM, because here the Austrian constitutional legislator had adopted specific constitutional provisions on the rights of the Austrian Parliament to participate in ESM decision taking. These provisions can at the same time be qualified as specific constitutional authorisations to conclude the ESM.

568  Stefan Griller European Commission and the CJEU) for decision-taking. In essence, the Constitutional Court, relying on one strain of the academic debate, held that Article 9(2) of the Constitution allows for isolating these competences from the ones transferred by primary EU law, and consequently qualifying them as ‘single’ competences within the meaning of this provision, thus paving the way for approval by simple majority. This author, by contrast, had argued that such a position includes the circumvention of Article 50 of the Constitution in a double sense. First, it makes Article 9(2) of the Constitution available for the transfer of powers to EU organs although, because of its limits to ‘single’ competences, it had never been available for transfers to the EU. Second, the principled decision of the Constitutional legislator from the EU Accession Act to Article 50(4) of the Constitution is set aside, that the qualified majority of both chambers of Parliament is needed for transferring competences to the EU.

V.  Constitutional Rules and Practice on Implementing EU Law A.  Participatory Rights of Parliament In an effort to mitigate the repercussions of EU membership to the constitutional principle of democracy, an effort was made to foster parliamentary influence as well as regional and local influence onto the conduct of Austria’s representatives in the EU. The primary targets of these efforts were the ministers as members of the Council of the EU. Without going into the details it might be said that the most important rules concern participatory rights of the Austrian provinces and of the Nationalrat, the people’s chamber of the Austrian Parliament. Included is not only a right to be informed on developments at the EU level, but also a right to issue opinions which are binding on the Government.69 The Federal Government is obliged to inform the provinces (Länder) on all EU projects affecting the autonomous sphere of the Länder. The Länder (and also representative bodies of local entities) have the right to express opinions on these matters. In the case of the Länder, such opinions are binding on the federal government provided that they reach a ‘uniform’ opinion.70 This is only the case if a majority of the nine Länder votes for such an opinion and no Land makes use

69 For this (in Austria at the time of accession) politically very sensitive issue (the solution being said to have been the price of the then oppositions parties for their consent to EU membership, which was needed to achieve the required two-thirds majority in Parliament) – see eg Peter Bußjäger, ‘Mitwirkung der Länder an der Rechtsetzung in der Europäischen Union’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 359; Heinz Fischer, ‘Die Demokratisierung der EU aus der Sicht des österreichischen Parlaments’, in A Leicht (ed), Regierungskonferenz 1996. Wohin steuert die EU? (Vienna, Signum, 1996) 167; Anna Gamper, ‘Mitwirkung des nationalen Parlaments an der Subsidiaritätskontrolle’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 339; Hans Hegeland and Christine Neuhold, ‘Parliamentary participation in EU affairs in Austria, Finland and Sweden: Newcomers with a different approach’, (2002) 10 European Integration online Papers 2 ff, 8 f, 11 f; Michael Morass, ‘Österreich im Entscheidungsprozess der Europäischen Union’, in E Tálos and G Falkner (eds), EU-Mitglied Österreich (Vienna, Manz, 1996) 32; Theo Öhlinger, ‘Die Mitwirkung des Nationalrates, des Bundesrates und der Länder an der Entstehung von sekundärem Gemeinschaftsrecht’, in S Griller and HP Rill (eds), Verfassungsrechtliche Grundfragen der EU-Mitgliedschaft (Vienna, Österreichische Staatsdruckerei, 1997) 1; Heinz Schäffer, ‘Österreichs Beteiligung an der Willensbildung in der Europäischen Union, insbesondere an der europäischen Rechtsetzung’, (1996) 51 Zeitschrift für Öffentliches Recht 3; Stefan Storr, ‘Mitwirkung des nationalen Parlaments an der Gesetzgebung der Union’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 315. 70 Art 23d para 2 B-VG.

Austria  569 of its veto right. The Federal Government may only depart from a uniform opinion for compelling reasons of foreign and integration policy. In practice this procedure is of minor importance. The second and in practice more important possibility is that the Nationalrat adopts an opinion on EU projects. Such opinions are binding on the Austrian representative in the Council if it concerns matters to be transformed by federal law or aiming at the enactment of directly effective measures, which would otherwise be a matter of federal legislation.71 Again a departure from such an opinion is only allowed for compelling reasons of foreign and integration policy. Furthermore, the minister has to consult the Nationalrat before such departure takes place. Should the EU measure entail an alteration of federal constitutional law, a departure from the opinion is only allowed if the Nationalrat does not object within due time.72 In practice and after some difficulties at the beginning, this procedure leads to regularly informing of the Länder and the Federal Parliament on issues of European integration. The flood of information and envisaged projects that these bodies have to digest is a major problem. Deliberations on EU projects in Parliament take place on a regular basis. However, binding opinions are rather the exception.73 Also, more recent instructions tend – and rightly so – to be flexible in order to avoid the marginalisation of the ministers in the Council. Furthermore, parliamentary participation according to the Protocol on subsidiarity is addressed by constitutional provisions. The National Council and the Federal Council may issue reasoned opinions on draft EU legislation whenever they opine that it violates the principle of subsidiarity.74 Since the entering into force of the Lisbon Treaty allowing for the subsidiarity complaint, the National Council and the Federal Council issued some 20 such complaints, with the Federal Council being much more active than the National Council.75 Equally, the two chambers of Parliament may request that the Republic shall initiate proceedings before the CJEU in cases of violations of that principle.76 Specific and restrictive provisions exist for instructing the Austrian representative in the European Council when it comes to certain matters. These include the use of a Simplified Treaty Revision under Article 48(7) TEU,77 the creation of new types of own funds for the EU, and generally, where EU law calls for the ratification by the Member States according to their constitutional provisions.78 It is not easy to evaluate the effect of these many participatory rights. It might be fair to conclude that they induce active participation of the Austrian Federal Parliament as well as that of the provinces in the developments at EU level. However, it would be exaggerated to claim that Austria’s EU policy is, on the basis of these provisions, Parliament-driven to an extent that would outweigh or balance the reduction of its influence caused by the transfer of powers to the EU organs.

71 Art 23e para 3 B-VG. It shall only be mentioned here that a related right exists under Art 23e para 4 B-VG for the Bundesrat, the chamber consisting of representatives of the parliaments of the Austrian Länder, providing for their participation in the law making procedure at the federal level, which is an essential element of the federal principle of the Austrian Constitution. However, the right to issue a binding opinion is restricted to cases where a project of the EU would have to be transposed by a constitutional act requiring the consent of the Bundesrat; the practical relevance of that constellation is dubious. 72 Art 23e para 3 B-VG. 73 Compare the statistics in Storr, ‘Mitwirkung’ (n 69) 333 ff. 74 Art 23g B-VG. 75 See for a discussion Gamper, ‘Mitwirkung’ (n 69) 355 ff. 76 Art 23h B-VG. In practice, this did not yet happen. 77 Art 23e para 2 and Art 23i paras 1 and 2 B-VG. 78 Art 23i paras 3 and 4 B-VG.

570  Stefan Griller It should be added that provisions to strengthen parliamentary influence, namely that of the Nationalrat, even beyond what has been described above, were enacted with regard to the ESM. Also here, the Austrian representative in the ESM has to respect possible instructions by the Parliament and has to give reasons whenever she or he deviates from such instructions.79 Moreover, any ESM decisions granting stability aid to another Member State, amendments to the share capital, the maximum lending capacity, the call for approved but not paid in share capital, or amendments to the financial instruments, may only be approved (or abstained from) if the Nationalrat has authorised the Austrian representative to that end.80

B.  Direct Applicability and Primacy: Opening up the Austrian Legal Order Through the Lens of the Courts Primacy of (at the time) Community law was quickly and undisputedly accepted by the Highest Courts.81 Moreover, the attitude of all three of them: the Constitutional Court, the High Administrative Court, and the Supreme Court can be characterised as ‘EU-friendly’ in the sense that application, implementation and clarification (through initiating preliminary rulings with the CJEU) are actively performed without the reticence found in other Member States.82 As regards the Constitutional Court, this attitude is based on the position that joining the EU opened up Austria’s legal order. However, this openness went no further than the extent necessary to abide by EU law.83 This attitude implies the readiness, and not just of the Constitutional Court, to set aside without much ado even constitutional law provisions, if they would otherwise impede the implementation and enforcement of EU law.84 One of the first and most spectacular cases shall serve as an illustration.85 It concerned indirect conflicts between EU law and Austrian law in the field of legal protection. The Constitutional Court set aside a constitutional law provision which restricted access to the High Administrative Court in Austria. The Constitutional Court considered whether the Austrian law on telecommunications was compatible with applicable Community law, in particular the so-called Open Network Provision Directive (ONP Directive).86 The directive required a right of appeal against 79 Art 50c B-VG. 80 Art 50b B-VG. 81 Compare only the first judgments in this respect: VfGH 10.12.1997, B2955/95; B3553/95; B570/96; B1741/96 ECLI:AT:VFGH:1997:B2955.1995; VfGH 12.04.1997, VfSlg 14.805/1997 ECLI:AT:VFGH:1997:G400.1996; VwGH 23.10.1995, 95/10/0081; VwGH 25.06.1996, 96/09/0088; OGH 14.09.1995, 8 ObS 37/95 ECLI:AT:OGH0002:1995:008 OBS00037.95.0914.000. 82 To illustrate this point, see the contributions of the presidents of the three Courts summarising the respective jurisprudence of ‘their’ Court: Gerhart Holzinger and Stefan Leo Frank, ‘Auswirkungen des Unionsrechts auf die Verfassungsgerichtsbarkeit’, Rudolf Thienel, ‘Auswirkungen des Unionsrechts auf Zuständigkeit und Verfahren des VwGH’; Eckart Ratz, ‘Auswirkungen des Unionsrechts auf Zuständigkeit und Verfahren des OGH’, all of them in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 545; 567; 603. 83 VfGH VfSlg 15.215/1998 (n 42). In VfGH VfSlg 15.189/1998 (n 42), the VfGH stressed, in the same vein: there is no compelling reason for an interpretation which would assume a modification of the Constitution which goes beyond what is necessary in order to abide by the requirements of Community law. 84 Eg VfGH 24.02.1999, VfSlg 15.427/1999 ECLI:AT:VFGH:1999:B1625.1998; VfGH 28.11.2003 VfSlg 17.065/2003 ECLI:AT:VFGH:2003:KR1.2000, VfGH 14.03.2012, VfSlg 19.632/2012 ECLI:AT:VFGH:2012:U466.2011 (available in English, see n 42). 85 VfGH VfSlg 15.427/1999 (n 84). After the entering into force of the major reform of the administrative court system in 2014, the constitutional situation is now changed. However, the case is still illustrative. 86 Directive 97/51/EC of the European Parliament and of the Council of 6 October 1997 amending Council Directives 90/387/EEC and 92/44/EEC for the purpose of adaptation to a competitive environment in telecommunications, [1997] OJ L 295/23.

Austria  571 a decision of the national regulatory authority to a body independent of the parties involved.87 In Austrian law, this body used to be the Telekom-Control-Kommission. In terms of the Austrian legal system, this was not a court (Gericht) but a board (Kollegialbehörde), that is, an executive body entrusted with judicial functions under the Federal Constitution. It was, thus, (very likely) a tribunal within the meaning of Article 6 ECHR and a court or tribunal within the meaning of Article 267 TFEU. It was also not clear whether it met the requirements of independence set by the ONP Directive. Furthermore, a specific provision of the Constitution (Article 133 Section 4 B-VG at the time) excluded – for all such boards, and thus also for the Telekom-Control-Kommission – appeals to the High Administrative Court. Recognising that Article 5a of the ONP Directive had direct effect, the Constitutional Court held that primacy of Community law required that Article 133 section 4 B-VG, which was the only obstacle to such an appeal, should be set aside. As a result, an appeal to the High Administrative Court (VwGH) appeared possible. However, the latter quite obviously was not convinced by the interpretation given by the Constitutional Court.88 When the matter was brought before the VwGH,89 it sought a preliminary ruling by the ECJ. The VwGH asked the ECJ to clarify, first, whether Article 5a ONP Directive was to be interpreted in such a way as to exclude the Telekom-Control-Kommission as the required independent appellate body. Second, if this was the case, whether the directive was, on this point, unconditional and precise enough to supersede the mentioned constitutional provision excluding an appeal to the VwGH. The ECJ qualified the Telecom-Control-Kommission as not meeting the requirements of Article 5a ONP Directive. It also concluded that access to a national court or tribunal that satisfies the relevant requirements (the VwGH in the given case) remains mandatory and that this court was obliged to disapply any national provision explicitly excluding its competence.90 Although there has been an obvious controversy between the two courts of public law on the interpretation of the directive at issue, the case shows the far-reaching openness of Austrian courts to give way to the efficient application of EU law. It is at the same time underpinning the fact that Austria since its EU accession constantly has been among the front-runners when it comes to the sheer numbers of initiating preliminary ruling procedures. This might be caused by a combination of three things. First, the unconditional constitutional basis of EU membership. Second, a sound preparation of EU membership including the introduction of specific procedural rules for initiating preliminary ruling procedures for all courts. Third, a remarkable effort in the field of post-graduate education aiming at lawyers and civil servants to make them familiar with issues of EC/EU law. There is another simpler, but not less spectacular, case that can serve as an example. The Constitutional Court was faced with one of the many Constitutional Acts of Parliament outside the text of the B-VG. In our case it required disclosure, for purposes of limiting excessive remuneration at the expense of the taxpayer, and for economic control, of not only the salary of certain civil servants and functionaries in public enterprises, but also their names, to the Court of Auditors. The constitutional requirement to do so was clear. Even if its conformity with the constitutional guarantee for data protection appeared to be questionable, the disclosure requirement

87 Art 5a para 3 Dir 90/387/EEC. 88 In fact, it can hardly be said that the interpretation given by the VfGH was beyond reasonable doubt. Therefore one could have assumed an obligation of the VfGH itself under Art 234 TEC to bring the matter before the ECJ. The VwGH in its reference rightly pointed to ECJ 17.09.1997 C-54/96 (Dorsch Consult Ingenieursgesellschaft v Bundesbaugesellschaft Berlin) ECLI:EU:C:1997:413, paras 40 ff, which could raise doubts regarding the position of the VfGH. 89 VwGH 24.11.1999, 99/03/0071. 90 ECJ 22.05.2003 C-462/99 (Connect Austria) ECLI:EU:C:2003:297.

572  Stefan Griller would have clearly taken precedence over the constitutional guarantee, given that it was both: lex posterior and lex specialis. In this situation the Constitutional Court asked the ECJ whether the obligation to disclose would be in conformity with the EU Data Protection Directive. The ECJ provided ‘guidance’ suggesting that this might not be the case.91 That paved the way for the Constitutional Court to disapply the constitutional law obligation.92 It held that the Data Protection Directive was directly applicable and prohibited the disclosure of the names as disproportionate. Such disclosure was, according to the Court, not necessary in order to guarantee an orderly and efficient use of public funds.

C.  Double Obligation According to standing jurisprudence of the Constitutional Court, the ordinary legislator is under a double obligation to respect both EU law and Austrian constitutional law, also when incorporating (eg by transposing directives), implementing or enforcing EU law.93 This entails several important consequences, which result at least partly from the tension of this principle vis-à-vis the opening up of the Austrian constitutional system towards EU law, and the readiness of the courts to disapply even constitutional law, as demonstrated above. Details of how to solve that tension are far from being undisputed or clarified.94 Of fundamental importance, in this respect, is Article 18 of the B-VG, spelling out what is normally dubbed the principle of legality (Legalitätsprinzip) which forms the cornerstone of the fundamental rule of law principle of the Austrian Constitution. Article 18(1) reads: ‘The entire public administration shall act upon the basis of Acts of Parliament only.’ The legislator is required to determine as precisely as possible the substance and the limits of powers delegated to the administration, which in turn would act unlawfully in the absence of such authorisation. Article 18(2) authorises every administrative authority to issue ordinances within its sphere of competence on the basis of Acts of Parliament only. Whenever an Act of Parliament is not sufficiently precise in determining the powers of the administration, it is declared null and void by the Constitutional Court.95 EU membership changed this without any constitutional amendments to the effect that directly applicable Union law is now also a valid basis for administrative action. However, that would not do away with Article 18 B-VG as a still viable cornerstone of the rule of law principle. The Constitutional Court, in its standing jurisprudence, affirms that Austria’s EU accession neither created a general authorisation to incorporate Community law (today: Union law) into municipal law by administrative ordinances, nor to delegate to the administration the power to concretise Union law by ignoring the power of the legislator. 91 ECJ 20.05.2003 Joined cases C-465/00, C-138/01 and C-139/01 (Österreichischer Rundfunk and Others) ECLI:EU:C:2003:294, esp paras 76–101. 92 VfGH VfSlg 17.065/2003 (n 84). 93 Starting with VfGH 17.06.1997, VfSlg 14.863/1997 ECLI:AT:VFGH:1997:B592.1996, and many subsequent judgments, among which of specific importance: 11.10.2006, VfSlg 17.967/2006 ECLI:AT:VFGH:2006:G138.2005; 05.12.2008, VfSlg 18.642/2008 ECLI:AT:VFGH:2008:G113.2008; 15.06.2016, VfSlg 20.070/2016 ECLI:AT:VFGH:2016:G25.2016; and 12.10.2017 VfSlg 20.209/2017 ECLI:AT:VFGH:2017:G52.2016. The Constitutional Court normally calls this doppelte Bindung (‘double binding’). Sometimes, this is also dubbed as double legal conditionality. 94 As for the doctrinal debate, compare only Stefan Griller, ‘Direktwirkung und richtlinienkonforme Auslegung’, in T Eilmansberger and G Herzig (eds), 10 Jahre Anwendung des Gemeinschaftsrechts in Österreich (Vienna/Graz, NWV, 2006) 91, 104 ff; Michael Holoubek, ‘Doppelte Bindung und Richtlinienumsetzung’, (2018) 73 Zeitschrift für Öffentliches Recht 603; Öhlinger and Eberhard, Verfassungsrecht (n 1) paras 200 ff; Öhlinger and Potacs, EU Recht (n 16) 124 ff. 95 See Öhlinger and Eberhard, Verfassungsrecht (n 1) paras 601 ff, with further references.

Austria  573 [B]y contrast, the transposition of Union law is, according to Article 18 para 2 B-VG a task for the legislator, not for organs empowered to issue ordinances … It is for the same reasons … that Articles 18 para 1 and para 2 B-VG have not been modified, insofar as these provisions entail the obligation for the legislator to determine in a sufficiently precise manner the entire activity of the administration, specifically issuing administrative ordinances.96

Consequently, EU directives are regularly transposed by ordinary legislation. Administrative ordinances are considered sufficient only where such legislation is sufficiently determined as required by Article 18 B-VG, and at the same time allowing for ordinances to transpose EU directives.97 The significance of the mentioned reading of the double-obligation principle goes beyond Article 18 B-VG. It touches upon all sorts of incompatibilities between Union and constitutional law. This is specifically challenging in all instances where dis-applying national law, even if combined with directly applying EU law would not create an EU conform solution. Of specific importance in this regard are the division of competences, organisational requirements, and the protection of fundamental rights. As regards the division of competences between the federal state and the Länder to transpose a directive that included the obligation to create uniform procedures throughout the country, the Constitutional Court found that this division had not been altered by acceding the EU. In case of inability to transpose a directive uniformly by simple legislation because of the competence division, an amendment of the Constitution would be mandatory.98 Similarly, constitutional requirements for the organisational framework of review bodies are to be respected and not set aside by the obligation of the respective directive to provide for such procedures.99 This ruling was given in a case dealing with reviewing an award of public contracts. Later, however, the Constitutional Court confirmed a dispensation from constitutional requirements in cases where Union law did not include any discretion, specifically for the transposition of directives. It would, due to the supremacy of Community Law [today: Union law] even over constitutional law of the Member States be unlawful to nullify national legislation that transforms Union law into national legislation in a case where Union law does not include any margin of discretion for the national legislator, which would render it impossible for the legislator to enact a replacing provision that would abide both by Union law and national constitutional law.100

Subsequently, the Court even extended this legal effect beyond the concept of supremacy to cases where fully harmonising Union law would entail limitations on fundamental rights without any 96 VfGH 13.12.2005, VfSlg 17.735/2005 ECLI:AT:VFGH:2005:G104.2005: ‘Der Verfassungsgerichtshof vertritt in ständiger Rechtsprechung … den Standpunkt, dass durch den Beitritt Österreichs zur Europäischen Union der Verwaltung keine generelle Ermächtigung zur Umsetzung von Gemeinschaftsrecht durch Rechtsverordnungen erteilt und Art 18 Abs 2 B-VG nicht so weit verändert wurde, dass den Verwaltungsorganen die Befugnis übertragen worden wäre, Regelungen des Gemeinschaftsrechts unter Ausschaltung des Gesetzgebers zu konkretisieren; vielmehr ist zur Umsetzung gemeinschaftsrechtlicher Vorschriften nach dem Konzept des Art 18 Abs 2 B-VG nicht der Verordnungsgeber, sondern der Gesetzgeber berufen. Aus den selben Erwägungen, die dieser Rechtsprechung zu Grunde liegen … , ist der Verfassungsgerichtshof weiters der Auffassung, dass Art 18 Abs1 und 2 B-VG durch den Beitritt Österreichs zur Europäischen Union auch insoweit nicht modifiziert wurde, als aus dieser Bestimmung der österreichischen Bundesverfassung das an den Gesetzgeber gerichtete Gebot abgeleitet wird, das gesamte Verwaltungshandeln, und im Besonderen auch die Erlassung verwaltungsbehördlicher Verordnungen, inhaltlich hinreichend vorherzubestimmen’. Compare already VfGH VfSlg 15.189/1998 (n 42). 97 The area of foodstuff regulation is one of the rather rare examples for such practice. 98 VfGH 10.10.2003, VfSlg 17.022/2003 ECLI:AT:VFGH:2003:G212.2002. 99 VfGH 04.10.2003, VfSlg 17.001/2003 ECLI:AT:VFGH:2003:G53.2003. 100 VfGH VfSlg 18.642/2008 (n 93).

574  Stefan Griller margin of discretion for the national legislator.101 As a consequence, such legislation remains constitutional even if deviating from requirements of the Constitution. This might be characterised as opening up the Austrian legal order beyond supremacy (including direct effect and blocking effect) vis-à-vis national law. It would at the same time considerably reduce the necessity of amendments to the Constitution. Most importantly, the Constitutional Court seems, as indicated, to accept that limitations on the exercise of the rights and freedoms recognised by the CFR102 would have to be judged upon more or less exclusively by the CJEU. Simple legislation that would be entirely determined by Union law, eg by a directive could, thus, only be nullified after the CJEU – which would, if the question arises before the Constitutional Court, have to be asked under Article 267 TFEU – would have annulled the directive. Otherwise, those limitations by simple legislation would remain valid even if an isolated scrutiny against Austrian fundamental rights standards would suggest a violation.103 Striking the balance between opening up the Austrian legal order to EU law and the continuing respect for the Constitution (double obligation) could as a consequence, mean that constitutional requirements have to remain unapplied where Union law – irrespective of direct applicability – does not include any margin of discretion for the national legislator. In cases, however, where Union law allows for discretion of the national legislator, both Union law and national constitutional law must be observed.104 This includes cases where EU law does not provide for any margin of discretion but could arguably go beyond the Union’s powers – eg affects the national division of competences between legislators, organisational requirements of legal protection, separation of powers between the legislative, the executive, and the judiciary.

D.  Constitutional Review as a Means to Enforce EU Law? 1.  General Approach Whether an EU law violation could simultaneously also be a violation of the Austrian Constitution has been a controversial issue. If it could, the competences of the Constitutional Court would be expanded to review Austrian law not only against the yardstick of the Austrian Constitution but also against EU law, and thus serve for enforcing the latter. There are several arguments in favour of such a position.105 First, on the grounds of the EU Accession Act, there is a constitutional duty to abide by EU law (at the time: EC law). EU law 101 VfGH VfSlg 20.070/2016 (n 93) paras 60 ff, and 20.209/2017 (n 93) para 50. Compare the discussion in Holoubek, ‘Doppelte Bindung’ (n 94) 609 f, 613 ff. 102 Arts 51 and 52 CFR. 103 At least, this is a sound reading of VfGH VfSlg 20.070/2016 (n 93) and 20.209/2017 (n 93), as suggested by Holoubek, ‘Doppelte Bindung’ (n 94) 609 ff, especially 615 ff. This reading would also be in conformity with the position that the Austrian Constitution does not include (Solange-type) reservations against Union law. 104 Similarly Holoubek, ‘Doppelte Bindung’ (n 94) 614 ff. 105 For a discussion of the issue see Griller, ‘Individueller Rechtsschutz’ (n 16) 127 ff; Mayer, Kucsko-Stadlmayer and Stöger, Grundriss (n 1) para 246/26; Peter Vcelouch, ‘Auswirkungen der österreichischen Unionsmitgliedschaft auf den Rechtsschutz vor dem VwGH und dem VfGH’, Österreichische Juristen-Zeitung 1997, 721, 724. The prevailing view in literature has been critical to such suggestions and supported – until the judgment of 2012 – the contrasting position of the Constitutional Court – eg Öhlinger and Potacs, EU-Recht (n 16) 185 ff; compare also two literary contributions of (former) sitting judges of the VfGH: Gerhart Holzinger, ‘Zu den Auswirkungen der österreichischen EU-Mitgliedschaft auf das Rechtsschutzsystem der Bundesverfassung’, in H Haller et al (eds), Staat und Recht. Festschrift für Günther Winkler (Vienna, Springer, 1997) 351, 357; Karl Korinek, ‘Die Bedeutung des Gemeinschaftsrechts für das verfassungsgerichtliche Verfahren’, in M Holoubek and M Lang (eds), Das verfassungsgerichtliche Verfahren in Steuersachen (Vienna, Linde, 1998) 31.

Austria  575 is, to put it differently, to be regarded as ‘material’ constitutional law insofar as it must always be taken into consideration when a new Act is to be adopted. Second, and closely related, if it is true that, on the grounds of the EU Accession Act, EU law forms part of the Austrian legal order106 not according to its legal form, but according to its contents, enforcement could very well follow the mechanisms foreseen for equivalent domestic sources of law. In other words, administrative courts would, as demonstrated, without any difficulties arrive at qualifying violations of EU law as ‘unlawful’ in the sense of Article 130 B-VG107 – even if prior to Austria’s EU accession such a reading was certainly excluded. If this is legally sustainable, as in the prevailing view, it is hard to find what would prevent the Constitutional Court from also qualifying violations of EU law as unconstitutional, in the sense of Articles 140 and 144 B-VG. This should at least hold true for those norms of EU law which are equivalent to constitutional norms. Third, at least partly, also the principle of equivalent judicial protection under EU law requires the application of a remedy that is open for complaints on alleged violations of the Constitution, if no other effective remedy against EU law violations would exist.108 However, the Constitutional Court in the early years rejected that view in toto.109 It held that a violation of EU law was, under Austrian law, equal to a violation not of constitutional, but of ‘ordinary’ legislation.110 Consequently, the protection against violations of EU law was, in principle, according to the VfGH, not a matter for that court, but a matter for the VwGH. The VwGH followed this view. The Supreme Court acted similarly, within the limits of its competence.

2.  The Exception: Charter Rights as Constitutional Rights a) A Fundamental Change in the Jurisprudence of the Constitutional Court In 2012111 the Constitutional Court changed its stance in one very important respect. Henceforth,

rights guaranteed by the CFR were deemed as having the same legal status as Fundamental Right guaranteed by Austrian Constitutional Law. Consequently, they can be invoked before the Constitutional Court in every possible respect, especially in order to challenge: • administrative court decisions violating fundamental rights,112 or violating rights through applying general norms (mainly ordinary legislation) which are violating fundamental rights (Article 144 B-VG);

106 Compare for this position in the above text near n 42. 107 And similarly, the ordinary courts on the grounds of the respective domestic law provisions. 108 Which could mainly be argued for cases where the Constitution excluded the possibility of filing a complaint with the VwGH (Art 133 section 4 B-VG old version before the 2014 reform). Today, this would apply to the (allegedly rare) cases where an EU matter would not be subject to revision before the VwGH (Art 133 para 1 section 3 B-VG). However, problems of that kind could also be overcome by not applying the respective restrictions: compare in the above text near n 91. 109 VfGH 26.06.1997, VfSlg 14.886/1997 ECLI:AT:VFGH:1997:B877.1996: ‘ein Verstoß gegen Gemeinschaftsrecht [ist] aus der Sicht des Verfassungsgerichtshofes der Verletzung einfachgesetzlicher Vorschriften gleichzuhalten … die wahrzunehmen der Verwaltungsgerichtshof berufen ist’. The court gave no reasons why this result should be easier to achieve than the extension of its own competencies, given that EU law certainly comes in different legal forms than both constitutional and ordinary Austrian legislation. 110 Obviously there is a strong affinity of this position to that of some other Constitutional Courts, eg: a) TCE 14.02.1991 Sentencia 28/1991 ECLI:ES:TC:1991:28; and TCE 22.03.1991 Sentencia 64/1991 ECLI:ES:TC:1991:64, and Gregorio Garzón Clariana, ‘The Spanish Constitutional Order’, in A Kellerman, J de Zwaan and J Czuczai (eds), EU Enlargement: The Constitutional Impact at EU and National Level (The Hague, Asser Press, 2001) 117 (compare also the contribution of Torres Pérez and González Pascual in this volume); or b) the Portuguese Constitutional Court; starting with TCPT 05.05.1998, Acordão 326/98, in Diário da República, II Série, n.E 160, 14 July 1998, 9745-47, and Rui Manuel Moura Ramos, ‘The Adaptation of the Portuguese Constitutional Order to Community Law,’ in the same volume, at 131. 111 VfGH VfSlg 19.632/2012 (n 84).

576  Stefan Griller • general norms, mainly ordinary legislation and administrative ordinances because of ­violating fundamental rights (Articles 139, 140, and 140a B-VG). As a consequence, the CFR, within the Austrian legal order, is enforceable as if it were national constitutional law. The Constitutional Court113 arrived at this conclusion in claiming that its previous jurisprudence could not be transferred to the EU Charter of fundamental rights. Based on the domestic law situation, it follows,114 according to the Court, from the EU principle of equivalence115 that the rights guaranteed by the EU Charter must be enforceable before the Constitutional Court as constitutionally guaranteed rights. Within the scope of the CFR those rights also constitute a standard of review for general norms by the Constitutional Court. This is the case whenever the respective guarantee of the EU Charter, in its wording and precision equals individual rights guaranteed by the Austrian Constitution. At the same time, the Constitutional Court is limiting the resulting consequences: • as mentioned, the change applies not to all guarantees of the EU Charter, but only to those which equal constitutional rights in terms of wording and precision; moreover, it does not apply to any other EU rights (the four freedoms, additional EU Citizens rights, etc); • the Court referred to the jurisprudence of the CJEU limiting the duty to introduce preliminary ruling procedures, especially if the CJEU ruling could not have any impact on the decision of the case.116 The Constitutional Court claimed that this would always be the case if a constitutionally guaranteed right, especially one guaranteed by the ECHR117 has the same scope of application as an EU Charter right. In such a case, the Constitutional Court would base its decision on the Austrian Constitution without any need to refer the matter to the CJEU. Even if, undisputedly, this is an ‘EU-friendly turn’ in the Constitutional Court’s jurisprudence, going beyond what already before had been an EU friendly approach, it raises several concerns.118 First, the starting point of the Constitutional Court is dubious.119 It notes that under EU law, the Charter ‘is an area that is markedly distinct from the “Treaties”’ and refers to Article 6(1) TEU as talking of ‘the Charter of fundamental rights and the Treaties’. However, it suppresses the passage that the CFR ‘shall have the same legal value as the Treaties’. If this equivalence were to be taken seriously, it would call for treating all EU rights, if not EU law in general, as equal, instead of creating a special regime for the Charter. 112 Given that those administrative courts are controlling the executive, mainly through scrutinising individual acts of the administration (Bescheide), this means that, indirectly, the VfGH can be addressed for alleged violations of the constitution by the administration. 113 VfGH VfSlg 19.632/2012 (n 84) paras 25 ff, esp 35. 114 Two years later, in VfGH 12.03.2014, VfSlg 19.865/2014 ECLI:AT:VFGH:2014:B166.2013 the Court slightly modified its reasoning. Now, it ‘follows’ from the principle of equivalence, but is not mandated per se by the principle, that Rights flowing from the Charter may be invoked as constitutional rights. 115 The Constitutional Court mainly referred to ECJ 01.12.1998 C‐326/96 (Levez v Jennings Ltd) ECLI:EU:C:1998:577 para 18; ECJ 19.06.2003 C‐34/02 (Pasquini) ECLI:EU:C:2003:366 para 59; ECJ 29.10.2009 C-63/08 (Pontin) ECLI:EU:C:2009:666 para 45. 116 ECJ 06.10.1982 Case 283/81 (CILFIT v Ministero della Sanità) ECLI:EU:C:1982:335. 117 The ECHR forms part of Austria’s constitutional legal order, on an equal basis to constitutional legislation. 118 Compare esp Magdalena Pöschl, ‘Verfassungsgerichtsbarkeit nach Lissabon. Anmerkungen zum Charta-Erkenntnis des VfGH’, (2012) 67 Zeitschrift für Öffentliches Recht, 587; but see also Michael Potacs, ‘Das Erkenntnis des VfGH zur Grundrechte-Charta und seine Konsequenzen’, Jahrbuch Öffentliches Recht 13 (Vienna, NWV, 2013) 11. 119 VfGH VfSlg 19.632/2012 (n 84) para 25.

Austria  577 Second, the cited CJEU jurisprudence does not seem to support the conclusions. The CJEU had ruled that the principle of equivalence is not requiring a Member State to extend the most favourable domestic law provision on legal protection to EU law. By contrast, the national court must establish objectively, in the abstract, whether the rules at issue are similar taking into account the role played by those rules in the procedure as a whole, the conduct of that procedure and any special features of those rules …120

That would mean that the new jurisprudence of the Constitutional Court is in line with EU law, but it is not mandated by the principle of equivalence. As long as legal protection is equivalently guaranteed by the High Administrative Court or the Supreme Court, as in the past, that would be fine, too. However, the modification of the reasoning in a later judgment of the Constitutional Court121 transforms the turn into a modified reading of the principle of equivalence in combination with Austrian constitutional law. That argument seems to provide a better basis for the change, given that it is drawing on the Court’s discretion to interpret Austrian constitutional law. However, it reinforces the doubts on the remaining differentiation regarding EU rights other than those flowing from the Charter. Third, the restriction of the obligation to introduce a preliminary ruling procedure before the CJEU in ECHR-related cases is unconvincing.122 It cannot be ruled out in advance that within the context of EU law an ECHR guarantee could lead to different results than those within the context of Austrian constitutional law. That alone would trigger the duty to refer the case to the CJEU. The turn of the Constitutional Court has, amongst others, to be seen against the consequences of its original reticence to open constitutional procedures for the enforcement of EU law. This led to the result that the two other highest courts, the High Administrative Court, and the Supreme Court, became the ‘EU fundamental rights courts’ in Austria, thereby to a certain extent diminishing the former dominance of the Constitutional Court in the protection of fundamental rights. Thus, the question arises which consequences flow from that turn of the Constitutional Court. b) The Resulting Relation between the Three Highest Courts As mentioned, the original position of the Constitutional Court, that violations of EU law equal violations of ordinary Austrian legislation, made the administrative courts with the High Administrative Court as the last instance, and also the ordinary courts with the Supreme Court as the last instance, the main judicial organs for enforcing EU law. The turn of the Constitutional Court put that development in doubt in two respects. First, Article 133(5) B-VG123 excludes matters under the competence of the Constitutional Court from the competence of the High Administrative Court. Consequently, it was questionable whether the High Administrative Court could continue with its jurisprudence to protect EU law and specifically rights flowing from the CFR. Second, Article 89(2) B-VG obliges any ordinary court suspecting a violation of the Constitution by any Act of Parliament, to refer the case to the 120 ECJ C‐63/08 (Pontin) (n 115), paras 45 and 46 (quotation). Compare also ECJ 27.06.1991 C-348/89 (MecanarteMetalurgica da Lagoa v Alfandega do Porto), ECLI:EU:C:1991:278. 121 VfGH VfSlg 19.865/2014 (n 114). 122 This materialises in VfGH VfSlg 19.865/2014 (n 114), where the court should have referred the matter to the CJEU not the least in order to clarify the scope of application of both the freedom of establishment, and of Art 21 para 2 CFR, which the VfGH denied without asking the CJEU. 123 Until 2014 (the year when the reform of the administrative court system in Austria entered into force) that had been Art 133 section 1 B-VG.

578  Stefan Griller Constitutional Court for the annulment of that act. Consequently, one could suspect that courts would have to bring the matter of a possible violation of the EU Charter before the Constitutional Court and are no longer entitled to enforce the Charter as directly applicable law. Regarding the first question, the High Administrative Court gave its answer implicitly.124 It simply continued applying the EU Charter of fundamental rights also in cases where it would be directly applicable and lead to the annulment of an individual administrative decision that was challenged before it. The second issue provoked a more spectacular reaction and simultaneously clarified not only the position of the ordinary courts, but also that of the High Administrative Court. The Supreme Court requested a preliminary ruling from the CJEU on the matter.125 The latter in essence strengthened the position of the Supreme Court. It confirmed its previous jurisprudence by stating, first, that national courts have the widest discretion in referring matters to the CJEU and that the jurisdiction of the CJEU on the validity of an Act of EU law may not be undermined.126 Second, the national court does not have to wait for the decision of the Constitutional Court and might refuse on its own motion to apply any conflicting provision of national legislation, even if the impediment to full effectiveness of EU law were only temporary.127 Third, a mandatory reference to the Constitutional Court might not curtail the right of the court to refer the matter to the CJEU. The national court must be free to refer for a preliminary ruling at whatever stage of the proceedings it considers appropriate, even at the end of an interlocutory procedure for the review of constitutionality.128 Fourth, the national court is free to adopt any measure necessary to ensure the judicial protection of EU law and to disapply, at the end of an interlocutory procedure, the national legislative provision if that court holds it to be contrary to EU law.129 As a consequence, the turn in the jurisprudence of the Constitutional Court does not curtail the right of the other courts, specifically that of the High Administrative Court and of the Supreme Court, to apply the CFR and to set aside conflicting national legislation.

3.  Primacy and Constitutional Scrutiny of General Norms The effect of the primacy of EU law on the control functions of the Austrian Verfassungsgerichtshof is a very peculiar feature. This is especially so with respect to the constitutional review of Acts of Parliament – apart from the mentioned violations of Charter Rights, which is a real exception to the general approach of the Court.130 The issue is whether a conflict of national law with EU law could be an obstacle for the VfGH to declare that law null and void for the reason of violating the Constitution. The problem occurs in cases of concrete norm control131 where the admissibility of a respective complaint is dependent on an alleged violation of the Constitution by possible ‘application’ of the respective law. However, if primacy of Union law prevents the application of the law,

124 VwGH 23.01.2013, 2010/15/0196. 125 OGH 17.12.2012, 9 Ob 15/12i ECLI:AT:OGH0002:2012:0090OB00015.12I.1217.000. The detailed, and partly flawed questions shall not be addressed in detail here. 126 CJEU 11.09.2014 C-112/13 (A) ECLI:EU:C:2014:2195. 127 Ibid, paras 36 f. 128 Ibid, paras 38 f. 129 Ibid, para 40. 130 If the Court would apply the approach presented in the above text Austrian legislation obviously conflicting with directly applicable charter rights would be inapplicable, which would, in principle, exclude declaring such legislation null and void by the VfGH. 131 Arts 89 paras 2, 139 and 140 B-VG.

Austria  579 would that not mean that any unconstitutionality procedure is to be declared inadmissible as long as the compatibility of the law with EU law is not established? The core of the VfGH132 findings is that such constitutional scrutiny is only inadmissible if the application of the law in question is completely inconceivable (denkunmöglich). As long as the contention that there is no superseding EU law is fairly reasonable, there is no hindrance for the Court to fulfil its duty of constitutional review.133 However, it has to be added, that the mentioned turn in the Constitutional Court’s jurisprudence on EU Charter rights necessarily calls for a corresponding change of this admissibility issue, which the Court so far has not yet explained. One of the first cases concerned the 1992 Burgenländische Tourismus-Förderungsgesetz (Promotion of Tourism Act in the province of Burgenland), providing for a tax to promote tourism in Burgenland on the basis of the net yield ‘within the meaning of the VAT Act [Umsatzsteuergesetz]’. In similar cases in Styria, the Tyrol and Carinthia, the VwGH had sought preliminary rulings from the CJEU on the question whether Article 33(1) of the Sixth VAT Directive precluded such taxes, ie whether this was a form of VAT or not.134 The VfGH did not refer the question to the CJEU but reviewed the statute for constitutionality in relation to the domestic legal order. Its justification for proceeding in this fashion was that there was no possible conflict between its decision whether or not to annul the statute and the primacy of Community law. The VfGH held that in cases where it would not annul a statute for unconstitutionality, it remained possible for Austrian courts exercising public law jurisdiction to clarify the question of compatibility with Community law (seeking a preliminary ruling in appropriate cases).135 In the case at hand, the VfGH held that it had to annul a phrase of the scrutinised law as inconsistent with the constitutional principle of equality, since applying the phrase would render net yields not only in Burgenland but throughout the country subject to the tax to promote tourism. As a consequence of the position that constitutional scrutiny is only admissible as long as the applicability of the law is not completely inconceivable (denkunmöglich), the VfGH, in a later case, terminated proceedings aiming at the annulment of a statutory law for the reason that its inapplicability was obvious.136 The Court noted that after the complaint against a decision of an administrative body had been filed with the VfGH, a judgment of the ECJ had made it sufficiently clear that the law at issue – concerning a differentiation between full-time and part-time employees in the pharmaceutical sector regarding financial claims – violated the equal pay principle under EC law.137 Primacy of EC law thus required the law in the present case to be set aside. Consequently, the precondition for the constitutional scrutiny was missing, which led to the termination of the procedure. The starting point of the procedure had been a complaint against an administrative decision that had applied Austrian law. Since that law was, according to the VfGH, inapplicable, this rendered the administrative decision illegal, simply for the reason that the required legal 132 VfSlg 15.215/1998 (n 42). 133 For a discussion of the judgment see Griller, ‘Individueller Rechtsschutz’ (n 16) 138 ff. 134 In ECJ 08.06.1999 Joined cases C-338/97, C-344/97 and C-390/97 (Pelzl and Others) ECLI:EU:C:1999:285 the Court could not find a violation of Community law: ‘The Sixth Council Directive (77/388/EEC) of 17 May 1977 on the harmonisation of the laws of the Member States relating to turnover taxes – Common system of value added tax: uniform basis of assessment, and in particular Article 33 thereof, does not preclude a charge of the kind introduced by the Steiermärkische Tourismusgesetz, the Tiroler Tourismusgesetz and the Kärntner Fremdenverkehrsabgabegesetz, which is payable by traders in a federal Land who have an economic interest in tourism, which is calculated, in principle, on the basis of annual turnover and from which input tax is not deductible.’ 135 At that stage, possible primacy and direct effect would lead to an obligation to disapply the Austrian law. 136 VfGH 11.12.1998, VfSlg 15.368/1998 ECLI:AT:VFGH:1998:G57.1998. 137 At the time: Art 119 ECT (today Art 157 TFEU). The VfGH referred to ECJ 17.06.1998 C-243/95 (Hill and Stapleton v The Revenue Commissioners and Department of Finance) ECLI:EU:C:1998:298.

580  Stefan Griller fundament in an Act of Parliament was missing. The Court stated – after having terminated the constitutional scrutiny of the inapplicable law and coming back to the original complaint against the administrative decision – that the decision violated the fundamental right to private property, since it severely curtailed this right without a proper legal basis.138 It shall be noted, however, that under the new Charter jurisprudence of the Constitutional Court the result would possibly be very different. If we agree that equal pay is now also protected by Article 23 CFR, the Court would have to annul the statutory law. The consequence for the administrative decision based on that law would, however, be the same.

4.  State Liability Another issue of enforcing EU law by the courts relates to state liability for violations of EU law. Normally, this is a matter for the ordinary courts which are also responsible for dealing with state liability issues apart from EU law. However, what about liability resulting from court decisions? In this respect the CJEU holds that individuals ‘must have the possibility of obtaining redress in the national courts for the damage caused by the infringement of those rights owing to a decision of a court adjudicating at last instance’.139 This prompted the VfGH, although formally being of equal ranking with the other ‘Highest Courts’, the OGH and VwGH, to examine a decision issued by the VwGH, containing an alleged infringement of Community law.140 The VfGH took recourse to a kind of lacunae clause (Reserve-Kompetenz) within the Constitution on financial claims (Article 137 B-VG) to legally justify its step and qualified it as securing the prerogative of the VfGH to decide on possible damages to be granted as a result of a violation of Community law by Austria’s highest courts. Earlier, the VfGH had decided that it feels competent to decide on cases of state liability attributable to an infringement of EU law by the legislator. However, the Court narrowed that down considerably by saying that such a competence would only exist in cases where breaches are ‘directly attributable’ to the legislator.141 In this context, the Court also explicitly rejected a possible competence to review alleged infringements of EU law set by the ‘Republic of Austria’ as an actor under private law, referring such claims to the ordinary courts.

VI.  Concluding Remarks The Austrian Constitution may be considered a straightforward one. The EU accession was enacted on the basis of the first total revision of the Constitution reflecting the most substantial changes to the Austrian legal system since 1920. As a principle flowing from the constitutional system, ‘core elements’ of constitutional identity that would be entirely excluded from European integration are unknown. The Austrian Constitution is a ‘constitution of procedures’ which would even allow for creating a European Federation – through another total revision of the Constitution.

138 VfGH 05.03.1999, VfSlg 15.448/1999 ECLI:AT:VFGH:1999:B3073.1996. 139 ECJ 30.09.2003 C-224/01 (Köbler) ECLI:EU:C:2003:513 para 36. 140 VfGH 10.10.2003, VfSlg 17.019/2003 ECLI:AT:VFGH:2003:A36.2000. In this case the claimant referred to a decision of the VwGH refusing the requested restitution of her costs for a preliminary ruling during the proceedings pending before the Austrian court. The VfGH finally upheld the decision issued by the VwGH. 141 VfGH 06.03.2001, VfSlg 16.107/2001 ECLI:AT:VFGH:2001:A23.2000. This means that whenever directly applicable EU law has been violated, which could have been directly applied by the courts or by the administration, it is for the ordinary courts to decide on eventual damages.

Austria  581 It is certainly a consequence of the path chosen to enter the EU in 1995 that the specific features of the EU legal order, especially direct effect and primacy, were quickly accepted without dispute in Austrian practice. Constitutional reservations both against subsequent treaty changes and against secondary EU legislation have not been playing a major role in the Austrian legal scene. ‘Ultra vires protection’ against secondary legislation might still be available in principle, but on rather restrictive grounds, and is so far just a theoretical possibility. Participatory rights of the Austrian Parliament were created as a counterbalance to the transfer of powers to the EU and were further reinforced after the Lisbon Treaty. However, these rights of Parliament are created and at the same time limited by the constitutional legislator. They are not specifically mandated by the constitutional foundations of EU membership. A very specific, but generally fundamental issue, is the turn in the jurisprudence of the Constitutional Court leading to the use of its competences for constitutional review as a means to enforce EU law. In this respect, the 2012 decision of the Constitutional Court to treat the EU Charter of fundamental rights as if it were a piece of constitutional legislation, and consequently enforceable by the Constitutional Court, is path-breaking and controversial at the same time. It causes a number of fresh problems and not all of them have been solved.

References L Adamovich, B-C Funk, G Holzinger and S L Frank, Österreichisches Staatsrecht I, 2nd edn (Vienna/New York, Springer, 2011). W Berka, Verfassungsrecht, 8th edn (Vienna, Verlag Österreich, 2021). P Bußjäger, ‘Mitwirkung der Länder an der Rechtsetzung in der Europäischen Union’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 359. H Fischer, ‘Die Demokratisierung der EU aus der Sicht des österreichischen Parlaments’, in A Leicht (ed), Regierungskonferenz 1996. Wohin steuert die EU? (Vienna, Signum, 1996) 167. B-C Funk, ‘Rechtserzeugung heute – Rechtserzeugung im 21. Jahrhundert’, in M Holoubek and G Lienbacher (eds), Rechtspolitik der Zukunft – Zukunft der Rechtspolitik (Vienna/New York, Springer, 1999) 45. A Gamper, ‘Mitwirkung des nationalen Parlaments an der Subsidiaritätskontrolle’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 339. G Garzón Clariana, ‘The Spanish Constitutional Order’, in A Kellerman, J de Zwaan and J Czuczai (eds), EU Enlargement: The Constitutional Impact at EU and National Level (The Hague, Asser Press, 2001) 117. C Grabenwarter, ‘Änderungen der österreichischen Bundesverfassung aus Anlaß des Beitritts zur Europäischen Union’, (1995) 55 Zeitschrift für ausländisches öffentliches Recht und Völkerrecht 166. C Grabenwarter and M Holoubek, Verfassungsrecht – Allgemeines Verwaltungsrecht, 4th edn (Vienna, Facultas, 2019). S Griller, ‘Verfassungsfragen der österreichischen EU-Mitgliedschaft’, Zeitschrift für Europarecht, Internationales Privatrecht und Rechtsvergleichung (1995) 89. S Griller, ‘Individueller Rechtsschutz und Gemeinschaftsrecht’, in J Aicher, M Holoubek and K Korinek (eds), Gemeinschaftsrecht und Wirtschaftsrecht. Zentrale Probleme der Einwirkung des Gemeinschaftsrechts auf das österreichische Wirtschaftsrecht (Vienna, Orac, 2000) 27. S Griller, ‘Der Stufenbau der österreichischen Rechtsordnung nach dem EU-Beitritt’, (2000) 8 Journal für Rechtspolitik 273. S Griller, ‘Introduction to the Problems in the Austrian, Finnish and Swedish Constitutional Order’, in A Kellerman, J de Zwaan and J Czuczai (eds), EU Enlargement: The Constitutional Impact at EU and National Level (The Hague, Asser Press, 2001) 147.

582  Stefan Griller S Griller, ‘Direktwirkung und richtlinienkonforme Auslegung’, in T Eilmansberger and G Herzig (eds), 10 Jahre Anwendung des Gemeinschaftsrechts in Österreich (Vienna/Graz, NWV, 2006) 91. S Griller, ‘Zur verfassungsrechtlichen Beurteilung des Vertrags über Stabilität, Koordinierung und Steuerung in der Wirtschafts- und Währungsunion („Fiskalpakt“)’, (2012) 20 Journal für Rechtspolitik 177. S Griller, ‘Die Wirtschafts- und Währungsunion vor, in und nach der Krise’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 791. S Griller, Die Neuordnung der Gerichtsbarkeit des öffentlichen Rechts, (2018) 20(1) ÖJT. S Griller, D Droutsas, G Falkner, K Forgó and M Nentwich, The Treaty of Amsterdam (Vienna/New York, Springer, 2000). S Griller, F Maislinger and A Reindl, Fundamentale Rechtsgrundlagen einer EG-Mitgliedschaft. Verfassungsfragen der Übernahme von EG-Recht in den bisherigen Mitgliedstaaten in vergleichender Sicht (Vienna, Österreichische Staatsdruckerei, 1991). S Harden (ed), Neutral States and the European Communities (London, Brassey’s, 1994). A Hauer, ‘Der ESM-Vertrag auf dem Prüfstand’, in D Ennöckl, N Raschauer, E Schulev-Steindl and W Wessely (eds), Festschrift für Bernhard Raschauer (Vienna, Jan Sramek, 2013) 155. H Hausmaninger, The Austrian Legal System, 4th edn (Vienna, Manz, 2011). H Hegeland and C Neuhold, ‘Parliamentary participation in EU affairs in Austria, Finland and Sweden: Newcomers with a different approach’, (2002) 10 European Integration online Papers. M Holoubek, ‘Auswirkungen des Unionsrechts auf das Verhältnis der österreichischen Höchstgerichte zueinander’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 625. M Holoubek, ‘Doppelte Bindung und Richtlinienumsetzung’, (2018) 73 Zeitschrift für Öffentliches Recht 603. G Holzinger, ‘Die Auswirkungen der österreichischen EU-Mitgliedschaft auf das österreichische Verfassungsrecht’, (1996) 4 Journal für Rechtspolitik 160. G Holzinger, ‘Zu den Auswirkungen der österreichischen EU-Mitgliedschaft auf das Rechtsschutzsystem der Bundesverfassung’, in H Haller et al (eds), Staat und Recht. Festschrift für Günther Winkler (Vienna, Springer, 1997) 351. G Holzinger and S Frank, ‘Auswirkungen des Unionsrechts auf die Verfassungsgerichtsbarkeit’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 545. A Janko, Gesamtänderung der Bundesverfassung (Vienna, Verlag Österreich, 2004). K Korinek, ‘Die Bedeutung des Gemeinschaftsrechts für das verfassungsgerichtliche Verfahren’, in M Holoubek and M Lang (eds), Das verfassungsgerichtliche Verfahren in Steuersachen (Vienna, Linde, 1998) 31. K Kröll and G Lienbacher, ‘Austria’, in European Parliament, DG for Internal Policies (ed), National Constitutional Law and European Integration (Brussels, European Parliament, 2011) 141. K Lachmayer, ‘The Constitution of Austria in International Constitutional Networks: Pluralism, Dialogues and Diversity’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (The Hague, T.M.C. Asser Press, 2019) 1271. A Lengauer, ‘First Experiences with EU-membership – the Case of Austria’, in V Kandžija et al (eds), Economic System of European Union and Adjustment of the Republic of Croatia (Rijeka/Trieste, Faculty of Economics, 1997) 539. G Lysén, ‘Some Views on Neutrality and Membership of the European Communities: The Case of Sweden’, (1992) 29 CML Rev 229. C Mayer, ‘Europäischer Stabilitätsmechanismus nicht verfassungswidrig’, (2013) Österreichische Zeitschrift für Wirtschaftsrecht 53. H Mayer, G Kucsko-Stadlmayer and K Stöger, Grundriss des österreichischen Bundesverfassungsrechts, 11th edn, (Vienna, Manz, 2015). M Morass, ‘Österreich im Entscheidungsprozess der Europäischen Union’, in E Tálos and G Falkner (ed), EU-Mitglied Österreich (Vienna, Manz, 1996) 32. RM Moura Ramos, ‘The Adaptation of the Portuguese Constitutional Order to Community Law’, in A Kellerman, J de Zwaan and J Czuczai (eds), EU Enlargement: The Constitutional Impact at EU and National Level (The Hague, Asser Press, 2001) 131.

Austria  583 T Öhlinger, ‘Die Mitwirkung des Nationalrates, des Bundesrates und der Länder an der Entstehung von sekundärem Gemeinschaftsrecht’, in S Griller and HP Rill (eds), Verfassungsrechtliche Grundfragen der EU-Mitgliedschaft (Vienna, Österreichische Staatsdruckerei, 1997) 1. T Öhlinger, Verfassungsfragen einer Mitgliedschaft zur Europäischen Union (Vienna, Springer, 1999). T Öhlinger, ‘Die Europäisierung des österreichischen Rechts’, in P Luif (ed), Österreich, Schweden, Finnland. Zehn Jahre Mitgliedschaft in der Europäischen Union (Vienna/Cologne/Weimar, Böhlau, 2007) 113. T Öhlinger, ‘Staatlichkeit zwischen Integration und Souveränität’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 111. T Öhlinger, ‘Österreichs Neutralität in der Europäischen Union’, (2018) 73 Zeitschrift für Öffentliches Recht, 621. T Öhlinger and H Eberhard, Verfassungsrecht, 12th edn (Vienna, Facultas, 2019). T Öhlinger and M Potacs, EU-Recht und nationales Recht, 7th edn (Vienna, LexisNexis, 2020). R Palmstorfer, ‘Austria’, in S Griller and E Lentsch (eds), EMU Integration and Member States’ Constitutions (Oxford, Hart, 2021) 523. P Pernthaler, Der Verfassungskern (Vienna, Manz, 1998). M Potacs, ‘Das Erkenntnis des VfGH zur Grundrechte-Charta und seine Konsequenzen’, Jahrbuch Öffentliches Recht 13 (Vienna, NWV, 2013) 11. M Potacs and C Mayer, ‘Fiskalpakt verfassungswidrig’? (2013) 21 Journal für Rechtspolitik 140. E Ratz, ‘Auswirkungen des Unionsrechts auf Zuständigkeit und Verfahren des OGH’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 603. H Schäffer, ‘Österreichs Beteiligung an der Willensbildung in der Europäischen Union, insbesondere an der europäischen Rechtsetzung’, (1996) 51 Zeitschrift für Öffentliches Recht 3. H Schäffer, ‘Österreichischer Landesbericht’, in J Schwarze (ed), Die Entstehung einer europäischen Verfassungsordnung (Baden-Baden, Nomos, 2000) 339. I Seidl-Hohenveldern, ‘Constitutional Problems involved in Austria’s Accession to the EU’, (1995) 32 CML Rev 727. M Stelzer, The Constitution of the Republic of Austria (Oxford, Hart, 2011). S Storr, ‘Mitwirkung des nationalen Parlaments an der Gesetzgebung der Union’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 315. R Thienel, ‘Auswirkungen des Unionsrechts auf Zuständigkeit und Verfahren des VwGH’; in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 567. M Vašek, Unabänderliches Verfassungsrecht und Revisionsschranken in der österreichischen Bundesverfassung (Vienna, Verlag Österreich, 2013). E Wiederin, ‘Der Vorrang des Unionsrechts vor nationalem Recht’, in S Griller, A Kahl, B Kneihs and W Obwexer (eds), 20 Jahre EU-Mitgliedschaft Österreichs (Vienna, Verlag Österreich, 2015) 179. R Winkler, Integrationsverfassungsrecht. Das österreichische Verfassungsrecht und das Recht der Europäischen Union – Koordination, Kooperation, Konflikt (Vienna/New York, Springer, 2003).

584

21 Poland MIROSŁAW WYRZYKOWSKI AND MICHAŁ ZIÓŁKOWSKI

I.  Main Characteristics of the National Constitutional System A.  The Constitution of 1997 as a Revolutionary Born and Reasonable Compromise The Constitution of the Republic of Poland (Konstytucja Rzeczypospolitej Polskiej, hereinafter KRP or Constitution) was enacted in 1997 after long public debate and discussion within the Constitutional Committee.1 It is important to note that the Constitutional Committee was inspired by the transformations of the government system that commenced as early as April 1989. The Round Table Agreement2 spelt out the terms of political change negotiated between the Government of the socialist state and the democratic opposition. Democratic opposition represented primarily the constituency of the ‘Solidarity’ trade union that grew into a social movement after the trade union itself had been deemed illegal during the martial law period of 1981–83. The Round Table provided, inter alia, for a partial constitutional reform that would create two new institutions within the system of government, ie the Presidency and the Senate as the second chamber of the Parliament. A negotiated new electoral law provided for opposition representatives to stand as candidates. In autumn 1989, the opposition was to dominate the Parliament to the extent the Members of Parliament were freely elected; 35 per cent of seats in the Sejm and all seats at the newly created Senate were subject to free elections and were taken by the opposition.3 Even though the Communist Party and its satellites formally retained a majority in the Sejm and one seat in the Senate, the unanticipated landslide victory of the democratic opposition ­undermined the political legitimacy of the communist MPs.

1 Ryszard Chruściak, ‘The Constitution of the Republic of Poland of 2 April 1997 – The Course of Parliamentary Work from 1995–1997’, (1997) 113–16 Droit Polonais Contemporain 163; Marek Mazurkiewicz, ‘The Draft of the Constitution of the Republic of Poland (1996) – Its Political Conditions and Expectations’, in M Wyrzykowski (ed), Constitution-making Process (Warsaw, Instytut Spraw Publicznych 1998) 115; Wiktor Osiatyński, ‘Rights in New Constitutions of East Central Europe’, (1994) 26 Columbia Human Rights Law Review 119. 2 Stanisław Gebethner, ‘Political Reform in the Process of Round Table Negotiations’, in G Sanford (ed), Democratization in Poland. 1988–90 (London, Palgrave Macmillan, 1992) 50. 3 Krzysztof Jasiewicz and Tomasz Żukowski, ‘The Elections of 1984–89 as a Factor in the Transformation of the Social Order in Poland’, in G Sanford (ed), Democratization in Poland. 1988–90 (London, Palgrave Macmillan, 1992) 98.

586  Mirosław Wyrzykowski and Michał Ziółkowski Parliament proceeded with democratic amendments to the Constitution of 1952.4 The first ground-breaking5 enactment came on 29 December 1989,6 when the Polish state ceased to be a socialist state in so far as the system of government was concerned. The nation was proclaimed as the source of sovereignty and the constitutional setup was now based on the notion of a democratic state ruled by law. The 1989 Amendment also entrenched the principle of political pluralism, the principle of territorial self-government, the right to private property and the right to pursue economic activity.7 The clause proved attractive to other constitution-makers in Central and Eastern Europe undergoing transition in the 1990s. The 1989 Amendment, as well as amendments adopted from 1990–92, rejected the political philosophy8 as well as all fundamental principles of undemocratic Constitution of 1952.9 This was the first step on the way to constitutional change. Another step was the constitutional Act of 17 October 1992,10 called Small Constitution of 1992, providing a temporally new regulation of legislative and executive powers for the transformation period until a new constitution entered into force. It also guaranteed a constitutional status and powers for local self-government. The Small Constitution of 1992 derogated the Constitution of 1952 while at the same time maintaining a few of the previously binding constitutional provisions in order to secure smoothness of the constitutional transformation. Its ‘main idea was to eliminate ambiguities in the April Amendment and to limit the powers of the President of the Republic’.11 The next step towards a new constitution was made by the 1989–97 case law of the Constitutional Tribunal (Trybunał Konstytucyjny (hereinafter TK), established in 1986).12 That period is considered as having been very active in the Tribunal’s judicial history.13 The Tribunal applied a progressive interpretation of the principle of the democratic state ruled by law.14 The

4 It was important at that time not undermine legality of constitutional transformation. Therefore, all changes of the Constitution of 1952 had been carefully made in accordance with the amendment procedure. Even changes axiologically revolutionary in their nature were introduced without the violation of the former constitutional order. The ‘chain of authority’ in a narrow (normative, Kelsenian) sense was unbroken. See Wojciech Sokolewicz, ‘The Legal-Constitutional Bases of Democratisation in Poland: Systemic and Constitutional Change’, in G Sanford (ed), Democratization in Poland. 1988–90 (London, Palgrave Macmillan, 1992) 97. 5 Miroslaw Wyrzykowski, ‘Constitutional Changes in Poland 1989–1991’, (1992) 17 Bulletin of the Australian Society of Legal Philosophy 25. 6 ‘Ustawa z dnia 29 grudnia 1989 r. o zmianie Konstytucji Polskiej Rzeczypospolitej Ludowej’ [‘Law of 29 December 1989 Amending the Constitution of the Polish People’s Republic’], Dziennik Ustaw [Journal of Laws, hereinafter DU] 1989, No 75, Item 444. 7 On the Rechtsstaat principle and its relation with the rule of law in the Polish context see Mark Brzezinski and Leszek Garlicki, ‘Judicial Review in Post-Communist Poland: The Emergence of a Rechtsstaat?’, (1995) 31 Stanford Journal of International Law 13, passim. 8 Mirosław Wyrzykowski, ‘Reflections on Some Recent Constitutional Developments in Eastern Europe’, (1992) 2 Tilburg Foreign Law Review, 164. 9 Leszek Garlicki, in Norske videnskaps-akademi (ed), The Role of the Constitution in a Changing Society. Joint PolishNorwegian Conference. Oslo, 14–16 May 1991 (Norwegian Academy of Science and Letters, Oslo, 1991) 18. 10 DU 1992, No 84, item 426. 11 Lech Garlicki and Zofia Garlicka, ‘Constitution Making, Peace Building, and National Reconciliation. The Experience of Poland’, in L Miller (ed), Framing the State in Times of Transition: Case Studies in Constitution Making (Washington/DC, US Institute of Peace Press, 2000) 394. 12 See Mark Brzezinski, ‘The Emergence of Judicial Review in Eastern Europe: The Case of Poland’, 41 American Journal of Comparative Law (1993) no 2, 173. 13 See Zdzislaw Czeszejko-Sochacki, ‘The Origins of Constitutional Review in Poland’, Saint Louis-Warsaw Transatlantic Law Journal (1996) 24; Leszek Garlicki, ‘Das Polnische Gesetz über den Verfassungsgerichtshof vom 29. April 1985’, (1987) 26 Der Staat 279. 14 Lech Garlicki, ‘Constitutional Court of Poland: 1986–2009’, in P Pasquino/F Billi (eds), The Political Origins of Constitutional Courts (Rome, Fondazione Adriano Olivetti, 2009) 13.

Poland  587 principle allowed the TK to construe a number of principles and rules that might not have been specified in positive constitutional law but became the foundation of the constitutional order in 1990–97 and beyond. Having regard to the fact that the constitutional changes between 1989–92 left a new human rights guarantee for the further regulation in a new Constitutional Act, the Rechtsstaat principle served as a universal axiological and normative basis for a broad human rights protection. It became a bridge for the standards of European Convention on Human Rights (ratified by Poland in 1993). The Tribunal used the Rechtsstaat principle and the ECHR to provide a well-developed standard of human rights protection before its full expression in the Constitution. The Constitution was preceded by a special Constitutional Act in 1992,15 providing the legal basis for the Constitutional Committee to prepare the draft law, for the National Assembly (ie Sejm and Senate acting as one body) to adopt the text of the Constitution and for the Sovereign to approve it in a referendum. The higher (constitutional) status of those provisions guaranteed the reliable framework for the constitution-making process. It stabilised and rationalised the debate and protected it against rapid and violent political changes. By a 52 per cent majority vote in the constitutional confirmatory referendum,16 the Constitution entered into force on 17 October 1997.17 The history and nature of the ‘Solidarity’ movement, Round Table Negotiations in 1989 as well as their constitutional achievements made from 1989–97 may suggest that Polish modern constitutionalism is revolutionary in its origins.18 However, Poland had no single constitutionmaking moment,19 but rather the new constitution was in the making from 1991–97.20 The Constitutional Committee of the National Assembly had to consider not only contemporary constitutional standards but also Polish constitutional antecedents as well as the institutional reality of democratic transition. The Constitution integrated the political compromise reached by the political parties and other important extra-parliamentary actors of political life (ie extraparliamentary opposition, Polish Catholic Church).21 As was later observed, the Constitution of 1997 ‘was drafted as a codification (confirmation) of concepts and practices that have developed in Poland since 1989 and reflected the European context of the Polish transformation. To a considerable extent, it was a constitution of continuation’.22

15 Act of 23 April 1992 on proceeding of preparing and adopting constitution of the Republic of Poland, DU 1992, No 67, item 336 as amended. 16 Michał Staszewski (ed), Referendum Konstytucyjne w Polsce [The Constitutional Referndum in Poland] (Warszawa, Instytut Studiów Politycznych PAN, 1997) passim. 17 Stanisław Gebethner, ‘The 1997 Referendum on the Constitution in Poland. The Controversies and the Compromise’, (1997) 113–16 Droit Polonais Contemporain 135. 18 Bruce Ackerman, The Rise of World Constitutionalism (Yale Occasional Papers No 4, 1996) and Bruce Ackerman, Revolutionary Constitutions Charismatic Leadership and the Rule of Law (Cambridge MA, Harvard University Press, 2019), pt II, ‘Solidarity’s Triumph in Poland’. As was later observed, ‘it seems that in Ackerman’s … typology, the current Polish Constitution cannot be classified as a typical revolutionary constitution but rather than as a revolutionary born constitution. However such a classification does not include a whole exceptionality of the Polish transformation’ (Michał Ziółkowski, ‘Constitutional Moment and the Polish Constitutional Crisis 2015–2018 (a few Critical Remarks)’, (2018) 4 Przegląd Konstytucyjny 76). 19 Compare András Sajó, ‘Constitution without the constitutional moment: a view from the new member states’, (2005) 3 International Journal of Constitutional Law 243. 20 Wiktor Osiatyński, ‘The Constitution-Making Process in Poland’, (1991) 13(2) Law & Policy, 129. 21 Wiktor Osiatyński, ‘Poland’s Constitutional Ordeal’, (1994) 3(2) East European Constitutional Review, 29; Wiktor Osiatyński, ‘A Brief History of the Constitution’, (1997) 6 (2/3) East European Constitutional Review, 67. Compare Konstanty Gebert, ‘Poland since 1989: Muddling through, Wall to Wall’, in S Ramet/C Hassenstab (eds), Central and Southeast European Politics since 1989 (Cambridge, CUP, 2019) 137. 22 Lech Garlicki, ‘Constitutional Court and Politics. The Polish Crisis’, in C Landfried (ed), Judicial Power How Constitutional Courts Affect Political Transformations (Cambridge, CUP, 2019) 144.

588  Mirosław Wyrzykowski and Michał Ziółkowski The Polish Constitution is thus a constitution of compromise. A single document had to be forged from several competing drafts, including those from the Constitutional Committee of the pre-1993 parliament, those of the post-1993 parliament, as well as a ‘citizens’’ draft constitution submitted by ‘Solidarity,’ and approved by over one million citizens. What was common to all the drafts were references to a republican form of government, the democratic character of the ‘State ruled by law,’ the Nation as a sovereign entity, political pluralism, separation and balance of powers, and an extensive catalogue of civil and political rights and freedoms with effective mechanisms for their protection.23

B.  The System of Government The system of government of the Republic of Poland24 is determined by a number of principles entrenched in the 1997 Constitution. Those principles are either specified in its provisions or reconstructed from the reasoning of the judicial branch – with the TK playing a dominant role – and conclusions drawn by the legal scholarship. They transcribe the leading ideas of the constitutional order by entrenching their supremacy within the system of constitutional values, by indicating the mechanism of exercising public authority and the axiology of the system of government. The principles of the system of government therefore constitute the rudiments of state and society. They encapsulate the constitutional identity of the Polish state even though they converge on the catalogue of principles of government characteristic of the contemporary model of constitutional democracy. There is a debate on the exhaustiveness of the catalogue of those principles, but it is sufficient for our purpose to indicate the following list of principles of the system of government: sovereignty of the nation; independence and sovereignty of the state; democratic state ruled by law; political pluralism of the civil society; separation and balancing of powers; social market economy; and the inalienability of human dignity.

1.  The Principle of the Sovereignty of the Nation The 1997 Constitution rejected the, previously binding communist concepts of state power and unity, the supreme role of Parliament as well as acting in the name of the people without direct legal competence. It directly proclaims that the sovereign powers belong to the nation (the Preamble25 in conjunction with Article 4 KRP26).

23 Mirosław Wyrzykowski, ‘Introductory Note to the 1997 Constitution of the Republic of Poland’, (1997) Saint LouisWarsaw Transatlantic Law Journal 3. 24 Małgorzata Masternak-Kubiak and Janusz Trzciński, ‘The System of Government in the Constitution of the Republic of Poland of April 2, 1997 – an Analysis of the Competence of the Sejm’, (1999) (1st special edn) The Sejm Review 39; Jerzy Ciapała, ‘The Position of the Sejm in the Area of Foreign Policy and International Relation’, (1999) (1st special edn) The Sejm Review 83; Paweł Sarnecki, ‘Control Power of the Senate of the Republic of Poland’, (2002) (2nd special edn) The Sejm Review 69; Leszek Garlicki, ‘Control Powers of the Senate of the Republic of Poland Comments to the Art by Prof Paweł Sarnecki’, (2002) (2nd special edn) The Sejm Review 87. 25 The KRP’s Preamble (relevant for interpretation of the constitutional provisions) reads: ‘Having regard for the existence and future of our Homeland … We, the Polish Nation – all citizens of the Republic … recalling the best traditions of the First and the Second Republic … Desiring to guarantee the rights of the citizens for all time, and to ensure diligence and efficiency in the work of public bodies … Hereby establish this Constitution of the Republic of Poland as the basic law for the State.’ 26 1. Supreme power in the Republic of Poland shall be vested in the Nation. 2. The Nation shall exercise such power directly or through their representatives.

Poland  589 The notion of the nation is understood as the community of all citizens of the state.27 The emphasis on ‘citizenship’ indicates that the Constitution provides for the national community to be political rather than ethnic in nature. The nation exercises its sovereign power by means of the instruments of direct and indirect democracy. Direct democracy28 was reduced to a referendum form and public legislative initiative by the further constitutional provisions.29 Only those two forms of exercising power may have legal effect.30 Due to the wording of Article 4 and other provisions on the Sejm and Senate as well as limited constitutional regulation on the institutions of direct democracy, it is said that those forms of democracy are not equal.31 Representative democracy is the primary form of exercising the sovereignty of the nation and bestows the exercise of power on elected representatives. It plays a key role in the constitutional order.32 A referendum is a supplementary form only.33 The 2018–19 constitutional crisis resurrected the old discussion on the nature of the sovereignty of the nation under the Polish constitutional law. The different actors of public life, including the Members of Parliament,34 Sejm Marshal experts35 as well as constitutional law scholars, refereed unexpectedly to the very particular understanding of the sovereignty of the nation in order to justify, legitimise and support the 2015–16 so-called ‘reforms’36 of the TK. According to their view, the sovereign is represented in the Parliament which has the supreme position in the constitutional system. Consequently, they argued that parliament may not only act directly in accordance with the ‘will of the nation’ to achieve political goals, but it may also replace the constitutional court and become a body of ‘last instance’ in all constitutional matters. However, it should be underlined that the argumentation was not supported in the wording of constitutional provisions and was contrary to the Polish constitutional framework for historical 27 ‘In the Preamble to the Constitution, the promulgator unequivocally affirmed that the Polish Nation … is “all citizens of the Republic.” For determination of membership in the Polish Nation, thus conceived, no factors apart from citizenship are therefore relevant – such as nationality, race or creed – and the essence of membership in the Polish Nation thus conceived is a sense of sovereignty manifest in holding Polish citizenship. This understanding of the notion “Polish Nation” under the Constitution would be incomplete. It follows from Art. 5 …, Art. 6 …, and indirectly also from Art. 35 …, that the promulgator treats the Polish nation (lower case) as a community not such much ethnic (in the strict sense of a “community of blood”), which would be moot given the migration and mixing of various nationalities both in the multinational First Republic and Second Republic and as a result of wars and invasions of foreign armies), as cultural’ (KT, 21.09.2015, OTK ZU no 8/A/2015 item 120). 28 See Kazimierz Działocha, in L Garlicki and M Zubik (eds) Konstytucja Rzeczypospolitej Polskiej. Komentarz [Constitution of the Republic of Poland. Commentary] vol 1 (Warszawa, Wolters Kluwer, 2016) 222. 29 There are four kinds of referendum: the ‘nationwide referendum’ – Art 125 KRP; referendum concerning the delegation ‘of certain powers of the competence organs of state authority to an international organization or international institution in relation to certain matters’ – Art 90 KRP; ‘confirmatory referendum’ on amending Chapters I, II or XII of the Constitution – Art 235 KRP; and the local self-government referendum – Art 170 KRP. See also Andrzej Piasecki, ‘Direct Democracy in Poland after 1989’, (2007) (3rd special edn) The Sejm Review 145. 30 Działocha, Konstytucja (n 28) 224. 31 Ibid, 225. 32 Monika Florczak-Wątor, in L Bosek and M Safjan (eds) Konstytucja Rzeczypospolitej Polskiej. Komentarz do art. 1-86 [Constitution of the Republic of Poland. Commentary to Arts 1-86] vol 1 (Warsaw, CH Beck, 2016) 274. 33 Paweł Sarnecki, ‘Central Ideas of the Constitution of the Republic of Poland of 2nd April 1997’, (1997) 17 Droit Polonais Contemporain 113–16; Działocha, Konstytucja (n 28) 225; Florczak-Wątor, Konstytucja (n 32) 276; Michał Pietrzak, ‘Demokracja reprezentacyjna i bezpośrednia w Konstytucji Rrzeczypospolitej Polskiej’ [‘Representative and direct democracy in the KRP’], in Staszewski, Referendum (n 16) 26, 31. 34 See, inter alia, Kornel Morawiecki, Session of the Sejm 25.11.2015, in 2 Sprawozdanie Stenograficzne [Stenographic Report] 78; Marek Ast, Session of the Sejm 02.12.2015, in 3 Sprawozdanie Stenograficzne, 14 f. 35 Arkadiusz Adamczyk et al, Expert Report on Issues Relating to the Constitutional Tribunal, 2016, www. marekkuchcinski.pl/wp-content/uploads/2016/09/EN-Raport-Zespo%C5%82u-Ekspert%C3%B3w-do-sprawproblematyki-Trybuna%C5%82u-Konstytucyjnego-wersja-angielska-1.pdf. 36 For details see Piotr Radziewicz and Piotr Tuleja (eds), Konstytucyjny spór o granice zmian organizacji i zasad działania trybunału konstytucyjnego, czerwiec 2015 – marzec 2016 [Constitutional dispute over the boundaries of changes in the organization and rules of operation of the TK, June 2015-March 2016] (Warszawa, Wolters Kluwer, 2017).

590  Mirosław Wyrzykowski and Michał Ziółkowski reasons. Nor can such a ‘nation’s will argument’ be defended on the ground of constitutional case law. Before the constitutional crisis started in Poland, the TK had no doubt that one of the most important aims of Arts 8 and 10 KRP is to protect against the concentration of powers in one office. Referring to the previous case law, the Tribunal recalled that the Parliament should not have a supreme position in the Polish constitutional system and highlighted that the supremacy of nation principle did not authorise Parliament to assume other constitutional bodies’ competences in the name of the nation.37 Article 4 KRP prohibits, in particular, that the legislative, executive or judicial branch replaces or opposes the Constitution as well as to act in the name of the nation in breach of constitutional law.

2.  The Principle of Independence and Sovereignty of the State History is both the origin and single most important justification for the constitutional provision of the principle of independence of the Polish state as well as the importance ascribed to the notion of independence. After all, the Polish state was subject to foreign occupation from 1795 to 1918 and from 1939 to 1945. Independence is understood as the condition necessary for a state to exist as an entity in relation to other states in both external and internal matters. The state has to exist and exercise functions within specified borders that are not subject to change. State sovereignty, on the other hand, pertains to the possibility of independent and unrestricted decision-making with respect to all matters within the ambit of the powers of state authority. Limitation of sovereignty is itself limited and tightly regulated. First of all, it is only possible to delegate the exercise of state authority based on Article 90 KRP that is dubbed the ‘European Union clause’. Secondly, any such delegation is conditional on being the necessary prerequisite for the benefit of the Republic of Poland. Finally, the delegation of state authority is limited insofar, that it can only take the constitutional form of ‘delegation of certain powers of the competence organs of state authority to an international organization or international institution in relation to certain matters’.

3.  The Principle of Democratic State Ruled by Law As mentioned above, the ‘foundations’ for the constitutional protection of individuals were created by the TK before the Constitution entered into force. The following constitutional principles and requirements had been derived from that general principle before 1997: • • • • •



the general prohibition of retroaction;38 nullum crimen sine lege;39 requirement of intelligibility and clarity of law;40 requirement of sufficient vacatio legis;41 protection of fairly legitimate rights42 and expectations;43

37 TK

22.09.2006, Case U 4/06, OTK ZU 2006, No 8, item 109. 14.05.1995, Case No K 1/94. 26.04.1995, Case No K 11/94. 40 TK 08.03.1995, Case No W 13/94. 41 TK 02.03.1993, Case No K 9/92. 42 TK 04.10.1989, Case No K 3/88. 43 TK 1102.1992, Case No K 14/91. 38 TK 39 TK

Poland  591 • principle of legal certainty;44 • principle of the protection of legitimate interest of individuals in case of intertemporal situations;45 • the prohibition of arbitrary and the concept of discretionary administrative decisions limited by the general constitutional principles;46 • principle of proportionality;47 • right to fair trial.48 That so-called ‘judicial constitution of rights’49 was sustained by the TK in its later case law50 and recognised as the ‘constitutional foundation of the state’.51 The Tribunal underlined its coherence with international human rights standards.52 Having regard to the fact the Constitution had provided a new chapter on human rights and freedoms, the role of the constitutional principle of a democratic state ruled by law was slightly changed. The Tribunal abandoned the activist interpretation of that principle and focused all interpretation efforts on other, more specific and detailed constitutional principles. This did not, however, imply that the catalogue of rights was closed in 199753 and the Rechtsstaat principle lost its meaning and significant role.54 Nowadays the principle of a democratic state ruled by law operates as a constitutional clause that is to be considered a synthesis of the status of the individual and the set of most fundamental directives binding state authorities at state and local level. The clause is best elucidated by indicating its components: the requirement of a legal basis for all acts of public authority; the limitation of the powers of state authorities; the supremacy of the Constitution; separation and balancing of powers; the independence of the judicial branch and the judiciary; the exclusivity of statutes as source of universally binding law; the principle of determinacy of law; the principle of decent legislation covering the principle of protecting the trustfulness of the citizen in respect to the state, protecting the durability of acquired rights, the requirement of due vacatio legis, prohibition of retroaction of law, and the principle of proportionality.

4.  The Principle of Political Pluralism of the Civil Society The idea of civil society was encapsulated in a number of specific constitutional provisions. The Constitution guarantees the right to create political parties and organisations, associations, social movements and public initiatives addressing and voicing needs, concerns and interests of the citizens. The Constitution entrenched pluralism with respect to political, social,

44 TK K 9/92 (n 41). 45 TK 15.07.1996, Case No K 5/96. 46 TK 29.09.1993, Case No K 17/92. 47 TK K 11/94 (n 39). 48 TK 25.01.1995, Case No W 14/94. 49 Elzbieta Morawska, Klauzula demokratycznego państwa prawnego w Konstytucji RP na tle orzecznictwa Trybunału Konstytucyjnego [The clause of the democratic state ruled by law in the KRP in the light of the jurisprudence of the TK] (Toruń, Dom Organizatora TNOiK, 2003) 202, 345. 50 TK 25.11.1997, Case No K 26/97, item 64. 51 TK 21.01.2001, Case No K 24/00, item 51. 52 TK 12.05.2011, Case No P 38/08, item 33. 53 TK K 26/97 (n 50) item 64. 54 TK 13.12.1998, Case No K 5/97, item 3, 21; see also TK 13.04.1999, Case No K 36/98, item 40, 240; and TK 07.10.2008, Case No P 30/07, item 135, 1375.

592  Mirosław Wyrzykowski and Michał Ziółkowski philosophical and religious convictions. Political parties consistent with political pluralism are the form in which Polish citizens can associate as free and equal agents with the aim of influencing the policies defined and implemented by the state by democratic means. Article 13 KRP excludes from this freedom political parties and organisations whose ‘programmes are based upon totalitarian methods and the modes of activity of Nazism, fascism and communism, as well as those whose programmes or activities sanction racial or national hatred, the application of violence for the purpose of obtaining power or to influence the State policy’, as well as such organisations ‘that provide for the secrecy of their own structure or membership’. Finally, the development of civil society is secured by the freedom of establishment and operation of trade unions, organisations of farmers, associations, civic movements as well as other voluntary associations and foundations.

5.  The Principle of Separation and Balancing of Powers The system of government is based on the principle of separation and balancing of powers, with the legislative, the administrative and the judicial branches regulated in detail by the Constitution. Legislative power is bestowed on the Sejm and the Senate. Executive power is vested with the Council of Ministers and the President of the Republic. Judicial power is the exclusive domain of the courts and tribunals. The principle is implemented by specifying the organisation and competences of particular branches of government as well as by providing for mechanisms of checks and balances without separating them in an absolute way. It is important to note that the Constitution provides for the obligation of state authorities to cooperate in as long as such cooperation does not encroach upon the very the essence of a particular branch of government. The judicial branch, however, has a special constitutional status with the Constitution stipulating that courts and tribunals are independent of other branches of government.

6.  The Principle of Social Market Economy The economic system of the Republic of Poland is based on the principle of Social Market Economy, as stipulated in Article 20 KRP (1997). Freedom of economic activity may be limited only by statute and on the condition of important public interest. Statutory limitations need to satisfy the principle of proportionality and cannot encroach upon the very essence of this freedom. Private property is the foundation of societal property relations and is therefore subject to special constitutional protection. It can only be limited by statute for reasons of public interest values covered by Article 31(3) KRP. The conflict of interests between employers and employees is considered to be natural; the Constitution obliges public authorities to provide for negotiating, cooperation and mediating mechanisms.

C.  The Role of the Constitutional Tribunal As mentioned above, the role of the TK was crucial for the progressive constitutional law development before 1997. Since the Constitution entered into force, its role has been analysed from various different perspectives. More sociologically oriented scholars tried to link the role of the TK with the ‘field’ of academic achievements, interests and positions of the Tribunal’s judges, but

Poland  593 the question is still a subject of critical legal studies.55 Not less open is the question of the scope and character of the constitutional ideas and constitutional politics of the TK. More theoretically oriented scholars focused on the question of whether and to what extent the post-1997 Tribunal’s case law was an example of continuity or discontinuity of the old legal regime and its institutions, in particular within private, criminal, labour and insurance law.56 No less important theoretical discussion on the Tribunal’s case law regarded the issues of judicial activism57 and the legitimacy of judicial review in Poland.58 The Constitutional Tribunal was generally seen as tending to adopt an active (or even very active in the human rights field) rather than passive interpretations of constitutional provisions. More doctrinally oriented scholars tried to analyse the TK’s role mainly in the frameworks of incrementalism, living constitutionalism, the public reason concept or the judicial dialogue with the CJEU, the ECtHR or the Polish Supreme Court (hereinafter SCPL). On the one hand, the very important part of the Tribunal’s interpretations was conservative rather than progressive in the field of human rights. In those cases, the Tribunal conducted weak proportionality test, maintained directly or indirectly constitutional limitations of human rights and referred rather to public order and community values rather the protection of individual freedoms and rights. On the other hand, the TK interpretations introduced new elements to the constitutional protection of individuals that became a starting point for legislative reforms or even revolutionised wellestablished statutory mechanisms. Regardless of the different opinions on the TK’s role, three conclusions seem to be uncontroversial. First, the Tribunal’s case law played a pivotal role in the process of European integration. The Tribunal seems to have become a co-creator of Polish constitutional and European ‘imagination’.59 Before the accession of Poland to the EU, the Tribunal reinterpreted constitutional provisions and proclaimed the constitutional principle of the EU law-friendly interpretation.60 Then in the accession judgment, the Tribunal underlined the general convergence61 and consistency of

55 Hanna Dębska and Tomasz Warczok, ‘Sacred Law and Profane Politics. The Symbolic Construction of Constitutional Tribunal’, (2014) 4 Polish Sociological Review 188, passim; Hanna Dębska, Władza, symbol, prawo. Społeczne tworzenie Trybunału Konstytucyjnego [Power, symbol, law. Social creation of the Constitutional Tribunal] (Warszawa, Wydawnictwo Sejmowe, 2015); Hanna Dębska, ‘Legal Doxa as a Form of Neutralization of Values in the Law. The Case of Polish Constitutional Tribunal’, in K Pałecki (ed), Neutralization of Values in Law (Warszawa, Wolters Kluwer, 2013) 303. 56 Aleksandra Kustra, ‘The Polish Constitutional Court and Political “Refolution” after 1989: Between the Continuity and Discontinuity of the Constitutional Narrative’, (2016) 6 Wroclaw Review of Law, Administration & Economics 86. 57 See the contributions to Bogusław Banaszak and Michał Bernadczyk (eds), Aktywizm sędziowski we współczesnym państwie demokratycznym [Judicial activism in a modern democratic state] (Warszawa, Wydawnictwo Sejmowe, 2012) with further references; and Adam Czarnota, ‘Guardians of the Constitution or legislative Bodies?’, (2001) 28 East Central Europe 147. 58 Wojciech Sadurski, Rights Before Courts. A Study of Constitutional Courts in Postcommunist States of Central and Eastern Europe (Dordrecht, Springer, 2005). 59 Martin Loughlin, ‘The Constitutional Imagination’, (2015) 78 The Modern Law Review 1, 13. 60 TK 27.05.2003, Case No K 11/03, item 43. 61 In TK 03.12.2015, Case No K 34/15, item 185, the Tribunal observed that ‘in a majority of European states, the review of the constitutionality of law has become centralised in character. The conduct of such a review has been assigned to one specialised organ of the state the scope of competence of which comprises adjudication on the hierarchical conformity of norms which leads to issuing a final determination that is binding for all … A characteristic of constitutional courts is that the judges of those courts are usually elected by legislative, and in some cases – executive, authorities … Nevertheless, a constitutional court … is always supposed to be an independent organ of the state, and the judges of that court are to be independent as well. These fundamental characteristics of constitutional courts are safeguarded by various solutions specified in a constitution. The procedure for choosing judges … are all to ensure that determinations issued by those judges will arise only from norms, principles and values expressed in a constitution.’

594  Mirosław Wyrzykowski and Michał Ziółkowski constitutional aims of the EU and Poland.62 The Tribunal delimited softly – in comparison to other constitutional courts in the EU – limits for the integration process in European Arrest Warrant case.63 Then, in the Lisbon Treaty judgment, the Tribunal underlined common elements of the Polish constitutional identity and the EU axiology. Once again the Tribunal accepted the multilevel character of European constitutionalism.64 It is true that at the same time, the Tribunal defined itself as a court of the last instance65 also in cases of the constitutionality of the EU law.66 However, the test for applicants who would like to question the EU law before the TK was almost impossible to fulfil. The test was clearly reserved for the ultra vires circumstances only.67 Second, the Tribunal case law became important in providing the effectiveness of ECtHR judgments. Not only did the Tribunal underline the similarities between the scope of the protection of freedom of assemblies68 and right to a fair trial69 under the Constitution and the Convention, but it also adopted the doctrine of dynamic interpretation of certain constitutional right in accordance with the changing ECHR developments.70 It is mainly the Tribunal case law achievement, that parties of civil proceedings in Poland have a right to reopen proceedings after the ECHR judgment.71 Third, the Tribunal has never directly accepted the doctrine of political question or the concept of non-justiciability. The Tribunal ruled in cases concerning all types and generations of constitutional rights (political, social, cultural and economic), cases of high systemic importance (eg voting rights, separation of powers, division of competences) and cases of high importance for moral (eg abortion72), axiological (eg the Martial Law decree73) and political reasons (eg lustration, decommunisation and unconstitutionality of the parliamentary resolutions on special investigations committees74). Instead of directly limiting its jurisdiction the Tribunal rather explored the margin of appreciation doctrine case by case or incidentally, it applied a deference doctrine. That approach is clearly visible in the late case law, when the Tribunal underlined and absolute primacy of the Constitution over the Parliament and other public authorities will. The Tribunal reserved for itself the position of the court of the last instance and guardian of the Constitution in all constitutional law matters. According to the Tribunal, the obligation to observe the Constitution is particularly important with regard to persons in power. This is, inter alia, manifested by an oath of office that must be taken before assuming the office by Sejm Deputies, Senators, the President, members of the Council of Ministers, as well as other officials. What safeguards the principle of the supremacy of the Constitution, and ultimately also the rights and freedoms of the individual, is inter alia the judicial review of the constitutionality of norms, conducted by an independent authority which is separate from the legislature and the executive. Since their origins,

62 TK 11.05.2005, Case No K 18/04, item 49. 63 TK 21.04.2005, Case No P 1/05, item 42. 64 TK 24.11.2010, Case No K 32/09, item 108. 65 Sławomir Dudzik and Nina Półtorak, ‘The Court of the Last Word: Competences of the Polish Constitutional Tribunal in the Review of European Union Law’, (2012) 15 Yearbook of Polish European Studies 225. 66 TK 16.11.2011, Case No SK 45/09, item 97. 67 See Krystyna Kowalik-Bańczyk, ‘Sending Smoke Signals to Luxembourg – the Polish Constitutional Tribunal in a Dialogue with the ECJ’, in M Claes et al (eds), Constitutional Conversations in Europe: Actors, Topics and Procedure (Cambridge, Intersentia, 2012) 267 passim. 68 TK 18.01.2006, Case No K 21/04, item 4. 69 TK 10.07.2000, Case No SK 12/99, item 143; TK 12.12.2005, Case No SK 53/04, item 134; TK 07.12.2010, Case No P 11/09, item 128. 70 TK 19.02.2008, Case No P 49/06, item 5; TK 0711.2004, Case No P 4/04, item 81; TK P 38/08 (n 52) item 33. 71 TK 22.09.2015, Case No SK 21/14, item 122, and cited judgments. 72 TK 28.05.1997, Case No K 26/96, item 19. 73 TK 16.03.2011, Case No K 35/08, item 11. 74 TK 26.11.2008, Case No U 1/08, item 160; TK U 4/06 (n 37).

Poland  595 constitutional courts in European legal culture have been conceived of as ‘safeguards for individuals against the tyranny of a majority’ and guarantors of the precedence of law over power. After the experience of the totalitarian regimes, there is no doubt that even a democratically-elected parliament has no competence to issue determinations that would be contrary to the Constitution, even if they were justified by ‘the good of the Nation’, where the term is understood in an abstract way. Thus, the constitution-maker has delineated substantive and procedural limits for public authorities, within which all their determinations must fall in every case.75

Since the Constitution of 1997 entered into force, the Tribunal has been ‘not only the guarantor of the supremacy of the Constitution, but it also safeguards the tri-division of powers. Any regulations concerning the Tribunal may not lead to a situation where it would lose its capacity to carry out its activity’.76 For the Tribunal, there were no doubts that the aim of Articles 8 and 10 KRP was to protect the Republic from the danger of concentration of powers and competences in one body. In several cases, the Tribunal directly underlined that Parliament should not have a supreme position in the constitutional system77 and underlined the supremacy of the nation principle (Article 4 KRP). However, the role of the TK started to change after the 2015 parliamentary elections, when the political majority in the Parliament adopted a series of Acts to modify competences and structure of the Tribunal. The short period between December 2015 (when Parliament invalidated the election of the Tribunal’s judges and elected its own ‘parallel’ judges) and December 2016 (the end of Andrzej Rzepliński’s term of office as TK President), when the TK had been resisting intensified and unprecedented legislative changes, brought a set of most important decisions and judgments. They summarised the role of the Tribunal in the Polish constitutional system.78 Unfortunately, those cases had not significantly affected the Parliament that adopted the 22 December 2015 amendment to the Constitutional Tribunal Act79 and again changed the procedure of constitutional review. The Tribunal ruled on the unconstitutionality of most of those changes in case No K 47/15. It underlined that the currently binding Constitution does not authorise the legislator to specify the jurisdiction and ­organisational structure of the Constitutional Tribunal since the Constitution itself regulates those issues. Thus, it is constitutionally inadmissible for the legislator to modify the scope of the Tribunal’s competence or to determine its position within the system of the organs of the state. Pursuant to Article 197 of the Constitution, what should be specified by statute is ‘the organization of the Constitutional Tribunal, as well as the mode of proceedings before it’. These are also matters that should be regulated by the constitution-maker in such a way so that it can be ensured that the activity of that public ­institution – as stated in the Preamble to the Constitution – is diligent and efficient … The independence of the Tribunal, which is so strongly emphasised in the present Constitution, requires special protection for two reasons. Firstly, it guarantees that the Tribunal has necessary discretion within the scope of the review of the law-making activity of the Polish Parliament and other organs of public authority as regards the conformity of that activity to the Constitution and other legal acts. Thus, the Tribunal may play the role of the guarantor of the supremacy of the Constitution, which the constitution-maker has

75 TK K 34/15 (n 61) item 186. 76 TK K 34/15 (n 61) item 186. 77 TK U 4/06 (n 37). 78 In K 34/15 the Tribunal pointed out that the new parliament had no power to invalidate the previous Justices election and could not elect Justices for the already occupied sits. In a second case (K 35/15) the Tribunal invalidated legislative provisions referring to the recalling of the TK’s President and the possibility of the re-election of the TK’s President and the conditions of taking an oath of office by a newly elected Justice. The third case (U 8/15) concerned the Tribunal’s competence to assess the parliament’s resolutions on the invalidation of the election of Justices and the appointment of new Justices (TK K 34/15 (n 61); TK 09.12.2015, Case No K 35/15, item 186; TK 07.01.2016, Decision, item 1). 79 DU 2015, item 2217.

596  Mirosław Wyrzykowski and Michał Ziółkowski assigned thereto, and none of the state authorities, including also the legislator, may deprive the Tribunal of that role. Secondly, the said independence guarantees that the Tribunal has the capacity to effectively protect the rights and freedoms of the individual, by adjudicating on the unconstitutionality of provisions that infringe those rights and freedoms. Thus, the Tribunal can fulfil the second role assigned thereto, namely the role of the guarantor of the constitutional rights and freedoms of the individual.80

However, Case No K 47/15 did not stop the Parliament. The new Constitutional Tribunal Law was adopted a few months later and was new only formally.81 It implemented solutions and the legal mechanisms that had been recognised unconstitutional by the Tribunal a few months earlier. Therefore, in the next case, no K 39/16, the Tribunal ruled on unconstitutionality of that formally ‘new’ law because of violations of the principle of separation of powers, the constitutional requirement of co-operation between state authorities and the constitutional guarantee of independence of courts and tribunals. 2015/16 will go down as fundamental in the institutional history of Polish Constitutionalism. On the one hand, the TK answered the most important questions on the basis, scope and function of rule of law, separation of powers as well as judicial review. On the other hand, since the end of December 2016, the legally composed Tribunal had finished its activity.

D.  The 2015–19 Constitutional Crisis: Three Stages of Decomposition of the Judiciary Polish constitutionalism in 2015 faced the most important crisis since the beginning of the democratic transformation in 1989. This crisis raised a number of more detailed questions and constitutional developments explained by the concepts of ‘bypassing the Constitution’82 or ‘democratic backsliding’.83 It was mainly84 focused on the structure and competences of the judiciary power. The ruling party (PiS – Prawo i Sprawiedliwość (Law and Justice)) adopted a number of Acts in order to take control over the judiciary and to change the constitutional order by regular statutes. These phenomena were described as ‘statutory anti-constitutionalism’85 and resulted in the vanishing of the Constitution.86 The process happened in three stages. 80 TK 09.03.2016, Case No K 47/15, item 2. 81 DU 2016, item 1157. It stipulated a) the requirement that a full bench of the Tribunal should adjudicate when three judges of the Tribunal file such a motion within 14 days after receiving certified copies of constitutional complaints, applications, or questions; b) the consideration of a case contingent upon the attendance of the Public Prosecutor-General; c) terms on which judges may raise an objection to a proposed determination with regard to a case considered by a full bench of the Tribunal; d) the obligation of the Tribunal to consider all cases commenced by a constitutional complaint or a question of law within one year, beginning with entry into force of the law. Moreover, it split the Tribunal’s judgments into those to be published in the DU and those not to. The new law also used a deeply stigmatising phrase regarding the Tribunal’s rulings (‘issued in breach of the provisions of the Constitutional Tribunal Act of 25 June 2015’). 82 Mirosław Wyrzykowski, ‘Bypassing the Constitution or Changing the Constitutional Order outside the Constitution’, in A Szmyt and B Banaszak (eds), Transformation of Law Systems in Central, Eastern and Southeastern Europe in 1989–2015, (Gdańsk, Gdansk University Press, 2016) 159. 83 Wojciech Sadurski, How Democracy Dies (in Poland): A Case Study of Anti-Constitutional Populist Backsliding, Sydney Law School Research Paper No 18/01 (2018); Wojciech Sadurski, ‘Polish Constitutional Tribunal Under PiS: From an Activist Court, to a Paralysed Tribunal, to a Governmental Enabler’, (2018) 11 Hague Journal on the Rule Law 63. 84 See also examples of statutory limitation of human rights given in chs 3, 4 and 7 of Wojchiech Sadurski, Poland’s Constitutional Breakdown (Oxford, OUP, 2019). 85 The process of achieving an unconstitutional result by ordinary statutes. Another element is a structural and personal change of the TK, the SCPL and other judiciary bodies like the NCJ. Once this is achieved, the Government can adopt laws that are unconstitutional or act beyond the limits of law, endangering the rule of law and fundamental rights. See Maciej Bernat and Michał Ziółkowski, ‘Statutory Anti-Constitutionalism’, (2019) 28 Washington Law Review 487, 492 f. 86 Lech Garlicki, ‘Die Ausschaltung des Verfassungsgerichtshofes in Polen?’, in A Szmyt and B Banaszak (eds), Transformation of Law Systems in Central, Eastern and Southeastern Europe in 1989–2015 (Gdańsk, Gdansk University

Poland  597 The first stage, focused on the TK, began in December 2015. Since the history of ‘Midnight judges’87 and the decomposition of the Tribunal88 was described in details by scholars,89 we need to underline that as of November 2019, a new President and Vice-President of the Tribunal were unconstitutionally90 elected. For the first time in history, one political party elected 14 of 15 members of the Tribunal. The Tribunal started to act in accordance with the will of the political majority. According to its own statistics, the Tribunal ruled in favour of the Government in almost all cases from 2017–19. The best-known examples are two judgments on the constitutionality of the new National Council of Judiciary, which is the constitutional body crucial for guaranteeing the independence of judiciary and judges appointment. The first judgment was delivered in 2017, the second in 2019.91 In order to declare constitutionality, preferred by the Government, the Tribunal adopted a new – so-called – ‘hostile interpretation’ of the constitutional provisions. It broke with the well-established case law and doctrines. This judgment is controversial for both theoretical and doctrinal reasons. The main argument is that the Tribunal accepted dismissal of all members of the old National Judiciary Council (hereinafter NCJ) before the end of the term, while there was a clear constitutional standard. That standard had prohibited the dismissal of office-holders before the end of their term, except under circumstances explicitly provided for by the law that did not exist. Moreover, the Tribunal ruled on the constitutionality of statutory provisions that excluded the right to question decisions of the NCJ before courts, although according to the Constitution, access to a court of first instance cannot be excluded by any statute. The Tribunal clearly ignored that provision. After 2016, the Tribunal became also a partner in political games played by different actors in Polish public life. After the SCPL preliminary reference to the CJEU, the Minister of Justice asked the TK whether Article 267 TFEU on the Preliminary Ruling Procedure was constitutional. In 2018, the Tribunal was asked to rule on the constitutionality of the elections of all judges of the Tribunal elected before 2015 (read: before PiS came to power). The same year, the Tribunal was also asked to rule on unconstitutionality of the legal grounds of appointment of all judges of the Supreme Court-appointed before 2018. All those cases are formally inadmissible as they obviously lack substantive constitutional grounds. Nevertheless, they are pending. These cases have been used by the PiS officials to undermine the legitimacy of the ‘old’ judges of the SCPL and ‘old’ judges of the Tribunal, as well as those Courts’ former case law. The aberration in the Tribunal’s work is visible also on a procedural level. According to recent research based on the Tribunal files, the President of the Tribunal violated the statute on the

Press, 2016) 63; Anna Chmielarz-Grochal and Jarosław Sułkowski, ‘Appointment of Judges to the Constitutional Tribunal in 2015 as the Trigger Point for a Deep Constitutional Crisis in Poland’, (2018) 2 Przegląd Konstytucyjny, 93; Jerzy Zajadło, ‘Constitution-hostile Interpretation’, (2018) 2 Przegląd Konstytucyjny, 8; Tomasz Koncewicz, ‘The Capture of the Polish Constitutional Tribunal and Beyond: Of Institution(s), Fidelities and the Rule of Law in Flux’, (2018) 43(2) Review of Central and East European Law, 116; Mirosław Wyrzykowski, ‘Antigone in Warsaw, in: Human Rights in Contemporary World’, in M Zubik (ed), Human Rights in Contemporary World. Essays in Honour of Professor Leszek Garlicki (Warszawa, Wydawnictwo Sejmowe, 2017) 372; Ewa Łętowska and Aneta Wiewiórowska-Domagalska, ‘A “Good” Change in the Polish Constitutional Tribunal?’, (2016) 1 Osteuropa Recht, 79. 87 Anna Śledzińska-Simon, ‘Midnight Judges: Poland’s Constitutional Tribunal Caught Between Political Fronts’, Verfassungsblog 2015/11/23. See also ‘An Oxford symposium on the Polish constitutional crisis sparks public debate, 12.05.2017, www.law.ox.ac.uk/news/2017-05-11-oxford-symposium-polish-constitutional-crisis-sparks-public-debate. 88 Sadurski, Polish (n 83). 89 Mirosław Wyrzykowski, ‘The Vanishing Constitution’, (2018) European Yearbook on Human Rights 3. 90 See Michał Ziółkowski, Przywracanie praworządności w TK po kryzysie konstytucyjnym: wybór i powołanie Prezesa TK (analiza sporządzona dla Archiwum Osiatyńskiego) [Restoring the rule of law in the TK after the constitutional crisis: election and appointment of the President of the TK (analysis prepared for the Osiatyński Archives)] (Warszawa, Archiwum Osiatyńskiego, 2019). 91 TK 25.03.2019, Case No K 12/18.

598  Mirosław Wyrzykowski and Michał Ziółkowski Tribunal’s proceedings many times between 2017–19.92 The statute provides that the President should always assign judges to a panel in alphabetical order. Moreover, the President is allowed to change the membership of the panels in limited statutory circumstances only. Unfortunately, the President abused her power to assign and reassign judges to the Tribunal’s panels hearing particular cases. All cases important for the PiS were taken from the old judges and assigned to panels consisting of newly elected ‘judges’, although one of them in November 2019 publicly accused the President of the Tribunal of manipulations and violations of the law. The second stage of the crisis concerned the composition of the courts of general jurisdiction and the NCJ. It started in mid-2017, when Parliament dissolved that body before the end of its term and replaced it by a new council of members elected by the current political majority in the non-transparent proceeding. An NGO asked Parliament for access to information in order to check who supported the candidates to the new NCJ, since there were doubts whether candidates received sufficient support demanded by law. Parliament refused. So the Supreme Administrative Court ordered publication of all files. Parliament questioned its final judgment and asked for an intervention of the Data Protection Officer, dependent on the Government. The Officer started investigations, the files remain unpublished and the new NCJ began its work notwithstanding the doubts regarding its legitimacy and legality. During extraordinary sessions at the end of 2018, following quick and short interviews of candidates for the positions of SCPL Judge, the Council recommended 40 persons for appointment by the President of the Republic. Over the next two months, the President of the Republic appointed 37 candidates as new SCPL judges. In 2019, the Council did not slow down and recommended new candidates who were immediately appointed. At the same time, the new law on disciplinary proceedings for judges enhanced the Minister of Justice’s power to initiate and intervene in disciplinary proceedings.93 Several judges of courts of general jurisdiction, including members of the judges association, were prosecuted for their judgments by the disciplinary officer who was newly appointed by the Minister of Justice. Later, another judge was prosecuted for his preliminary reference to the CJEU. Moreover, this year, after the CJEU judgment from June, the prosecutor officer raised disciplinary charges against judges from Krakow. They tried to execute the CJEU judgment and suspended proceedings in order to wait for the second CJEU judgment. Last but not least, the prosecutor officer suggested in November 2019, that application of the CJEU judgment by judges would automatically activate disciplinary proceedings. Part of the mentioned disciplinary cases are pending, but some judges were convicted (but without punishment being imposed) by the new Disciplinary Chamber of the SCPL. The third stage, concerning the SCPL, began in the middle of 2018 after the Act on the Supreme Court entered into force. The statute lowered the retirement age for SCPL judges from 70 to 65. It was directly applicable to acting judges of the court, without leaving them the right to decide whether or not to exercise the lower retirement age. The new provision imposed on acting judges of the SCPL who were 65 years or older an obligation to obtain the consent of the President of the Republic to exercise their judges’ function. Moreover, the Act created new positions in the SCPL 92 Michał Ziółkowski, ‘Zmiany składu orzekającego Trybunału Konstytucyjnego’ [‘Changes in the adjudication panel of the TK’], in Katarzyna Łakomiec (ed), Funkcjonowanie Trybunału Konstytucyjnego w latach 2014–2017. Raport Zespołu Ekspertów Prawnych Fundacji im. Stefana Batorego [Functioning of the Constitutional Tribunal in 2014–2017. Report of the Team of Legal Experts at the Stefan Batory Foundation] (Warszawa, 2018) 157. 93 ‘In the period between 12 August 2017 and 12 February 2018, over 70 presidents (and 70 vice-presidents) of courts have been dismissed under the six-month transitional regime which gave the Minister of Justice the power to arbitrarily dismiss them without any specific criteria, without justification and without judicial review’ (European Council, Rule of Law in Poland/Art 7(1) TEU Reasoned proposal – European Commission contribution for the hearing of Poland on 26 June 2018, ST 10351 2018 INIT – NOTE, 14).

Poland  599 by adding two new chambers, a Disciplinary Chamber and a Chamber of Extraordinary Control and Public Affairs. The latter aims at controlling elections as well as part of the decisions of other chambers. When the new law on the Supreme Court entered into force, 27 SCPL judges were over 65 years of age, including the First President of the SCPL. The 65+ judges were back to work at the end of 2018 due to the CJEU interim measure, but it had not stopped the ‘court-packing plan’. Facing the above-mentioned acts and violations of constitutional provisions, the judicial branch decided not to stand and watch as the President of the Republic and the new NCJ transformed the personnel and structure of the SCPL and judiciary. The SCPL submitted preliminary references to the Court of Justice regarding the 2018 changes contained in the Supreme Court Act.94 The CJEU was also asked by several courts of general jurisdiction whether the new model of disciplinary proceedings for judges, with a significant role, as provided for the Minister of Justice, is consistent with EU standards. In July 2019 the CJEU ruled in an infringement proceeding that Poland violated the non-discrimination principle.95 In November, the Court answered to the SCPL preliminary references regarding the independence of the newly appointed and added Disciplinary Chamber of the SCPL, as well as the new NCJ.96 The case concerned question whether those two mentioned-bodies satisfy conditions of impartiality and independence.97 In Case C-624/18, the CJEU had three options to settle the case: a) providing a theoretical or constitutional model for judiciary organisation; b) adapting a minimum standard by a direct application of the test of appearance to the Disciplinary Chamber; c) leaving the test of the appearance of independence to the national courts. Option a) would have provoked a flood of preliminary references, since composition and competences of judicial councils vary across Europe and reflect deeply rooted conceptions of the political/judicial power-balance. Moreover, powerful arguments referring to constitutional identity might have been raised. Finally, a judgment would have been addressed to the Polish Parliament and before its implementation, the SCPL would no longer have had competence to finish the case by itself. Paradoxically, had the CJEU applied the test of ‘appearances’ on its own, the impact of the judgment would have been narrow.98 Taking into account the essence of the case and the SCPL preliminary ruling, the CJEU judgment would have been applicable to the Disciplinary Chamber only. It could not have been used to assess the independence of other SCPL judges, newly appointed to others Chambers. Moreover, the hypothetical judgment could not have been used in the assessment of the independence of lower court judges appointed after the non-transparent proceedings before NCJ. Finally, parties of closed cases would have been deprived chances for

94 Stanisław Biernat and Monika Kawczyńska, ‘Why the Polish Supreme Court’s Reference on Judicial Independence to the CJEU is Admissible after all’, Verfassungsblog 2018/8/23. 95 CJEU 11.07.2019 C-619/18 (Commission v Poland) ECLI:EU:C:2019:615. See Maciej Taborowski and Paweł Marcisz, ‘The first judgment of the ECJ regarding a breach of the rule of law in Poland?’, Verfassungsblog 2019/5/29. 96 CJEU C-624/18 (Independence of the Disciplinary Chamber of the SCPL). 97 The first reference was submitted in the case of a Supreme Administrative Court judge reaching the age of 65 before the entry into force of the new law. He declared his will to continue in his position, but the NCJ rejected his application and issued an unfavourable opinion. The judge brought an action before the SCPL, claiming a violation of the nondiscrimination principle provided by EU law. The second and third preliminary references of the SCPL concerned the legal status of two judges of that court. They also reached the age of 65 before the date of the entry into force of the Law on the Supreme Court but refused to ask the President of the Republic for a permission to stay in their position. After the President sent them a private letter informing them on their retirement, they brought actions before the SCPL against the President of the Republic. 98 Michał Krajewski and Michał Ziółkowski, ‘The Power of “Appearances”’, Verfassungsblog, 2019/11/26.

600  Mirosław Wyrzykowski and Michał Ziółkowski reopening proceedings or liability for damages for unlawful actions. The CJEU finally opted for a ‘more complex argumentative and balancing approach’99 largely based on the concept of ‘appearances’ of independence taken from ECtHR case law.100 The CJEU held that the SCPL would have to consider whether the Disciplinary Chamber is independent and impartial in the lights of the test of appearance. At the time of the judgment, the broader context of CJEU case law provided numerous directions for the SCPL regarding the European constitutional standard. It seems that CJEU protected constitutional pluralism (Arts 2 and 3 TEU) postponed the risk of a ‘tyranny of values’101 and safeguarded the future application of the judgment in Poland. It was observed that ‘the ECJ gave to all Polish courts a powerful tool to ensure each citizen’s right to a fair trial before an independent judge, without undermining the systems of judicial appointments in the other Member States.’102 There is no doubt that the CJEU judgment opened a new chapter in the Polish constitutional crisis.

E.  Constitutional Culture and Living Constitutionalism In the Polish constitutional system, the appropriate use of the concept of living constitutionalism should not cause major controversies since the strict originalism, intentionalism and their variations have not achieved a sufficient level of support in the TK case law or constitutional law doctrine.103 There is, however, one clear exception in the Tribunal’s living approach to the constitutional provisions. It is Article 18 KRP, providing that ‘[m]arriage, being a union of a man and a woman, as well as the family, motherhood, and parenthood shall be placed under the protection and care of the Republic of Poland.’ This is one of the few constitutional provisions on human rights that has been ‘unattached’104 or ‘forgotten’105 by the Tribunal. While scholars were calling for a dynamic reading of the constitutional text and equal constitutional guarantee of marriage,106 the TK indirectly accepted the marriage as a union of a man and a woman in the judgment concerning income tax.107 99 Ibid. 100 Michał Krajewski, ‘The AG Opinion in the Celmer Case: Why the Test for the Appearance of Independence is Needed’, Verfassungsblog, 2018/7/05. 101 Armin von Bogdandy, Tyrannei der Werte? Probleme und Wege europäischen Schutzes nationaler Rechtsstaatlichkeit, MPIL Research Paper No. 2019-04. 102 Krajewski/Ziółkowski, ‘Appearances’ (n 98). 103 Since 2015, representatives of public actors and PiS criticised the living constitutionalism concept and accused the TK of judicial activism and the reading of KRP in a way violating Parliament’s right to conduct constitutional policy. 104 The TK adopted a broader concept of ‘family’ in one instance, that included partners living together, but avoided any reference to Art 18 KRP. The Tribunal limited itself and concluded the case on privacy principle only (TK 25.07.2013, Case no P 56/11). 105 Interpretation is unchanged since 2004, when the TK held in the Accession Treaty judgment that ‘marriage, being a union of a man and woman, has acquired a distinct constitutional status … and [a]ny modification of this status would be possible only by the way of an amendment of the Constitution …; in no circumstances would it be possible by way of a ratified international agreement’ (TK K 18/04 (n 62)). 106 See eg Adam Bodnar and Anna Śledzińska-Simon, ‘Between Recognition and Homophobia: Same-Sex Couples in Eastern Europe’, in D Gallo et al (eds), Same-sex Couples before National, Supranational and International Jurisdictions (Heidelberg, Springer; 2014) 211; Ewa Łętowska and Jan Woleński, ‘Instytucjonalizacja związków partnerskich a Konstytucja RP z 1997 r’ [‘Institutionalisation of civil unions and the KRP of 1997’], (2013) 6 Państwo i Prawo 15; Mirosław Wyrzykowski, ‘Publiczne a prywatne w wykładni konstytucyjnej na przykładzie art. 18 Konstytucji RP’ [‘Public versus private in the constitutional interpretation on the example of Art. 18 of the KRP’], in T Giaro (ed), Interes publiczny a interes prywatny w prawie [Public and private interest in law] (Warszawa, SAWPA-WU, 2009) 215; Jakub Pawliczak, Zarejestrowany związek partnerski a małżeństwo [Registered partnership and marriage] (Warszawa, Wolters Kluwer, 2014) 337–67. 107 TK 12.04.2011, Case No SK 62/08. The TK held that ‘the family protection implemented by the public authorities must take into account the vision of family adopted by the Constitution as a lasting union of a man and a woman oriented towards motherhood and responsible parenthood’.

Poland  601

II.  Constitutional Foundations of EU Membership and Closely Related Instruments A.  The So-Called European Union Clause (Article 90 KRP) The years of adopting the new Constitution were paralleled by procedure leading to the accession of the Republic of Poland to the EU, and constitutional discussions addressed this political reality. It was considered essential that the Constitution should provide a legal basis for EU membership. Initially, it was proposed to address it in the first chapter of the Constitution, an idea that, however, could not muster much political support because of the various ideations of sovereignty of the state. It was held improper to ‘pollute’ the Chapter consecrated to the fundamental principles of the system of government with provisions that governed the limitation of the sovereignty that would result from EU-membership. Finally, the ‘European Union Clause’ was placed in Chapter III KRP, and rightly so because this Chapter governs the ‘sources of law’, i.e. regulates the way the legal norms are to be created and applied. Article 90 KRP stipulates that ‘[T]he Republic of Poland may, by virtue of international agreements, delegate to an international organization or international institution the competence of organs of State authority in relation to certain matters’. An international agreement, therefore, is the only legal form capable of producing the effect of delegating the exercise of state powers upon an international organisation or upon the authority of an international organisation in relation to certain matters. According to Article 87, a ratified international agreement is a source of universally binding law in the Republic of Poland, alongside the Constitution, a statute, and a regulation implementing a statute. International agreements are subject to ratification and promulgation and publication in the DU. The Constitution also provides for international agreements to be ratified after Parliament expressed prior consent by the way of a ratification statute, wherever it pertains to (1) peace, alliances, political or military treaties, (2) constitutionally protected freedoms, rights or obligations of citizens, (3) state membership in an international organisation, (4) involves considerable financial responsibilities, or (5) regulates matters for which the Constitution requires the form of a statute. Correspondingly, the Constitution provides for a special procedure of ratification of the international agreement if the agreement delegates the exercise of state powers to an international organisation or its authority. The scope of delegated powers and the procedure of delegating powers are both subject to constitutional regulation. It is essential to note a number of principles – including limits – of delegating the exercise of state powers. The principles are not expressly enumerated but result from the analysis of such matters as the notion of the international agreement transferring state powers, the essence of a delegation of exercise of powers, the typology of state authorities and the notion of ‘certain matters’ in relation to which the delegation of state powers is possible. The specific problem of the international agreement transferring the exercise of state powers to an international organisation is that the transfer of powers can be made by the initial, or primary, agreement (ie the Accession Treaty) as well as in later amendments of this primary agreement. Furthermore, it is possible to transfer powers within a simplified, or fast-track procedure as long as the constitutional requirements governing the ratification of the primary agreement are observed (see below).108



108 TK

K 32/09 (n 64).

602  Mirosław Wyrzykowski and Michał Ziółkowski The notion of transfer of powers entails the limitation of the powers of state authorities so that the powers transferred by a number of countries can be exercised jointly by the international organisation. The notion of ‘organs of state authority’ covers primarily the authorities that the Constitution provides with legislative, executive, judicial and scrutiny powers. In as much a legislature is concerned, only legislative powers are subject to transfer; adjoining powers are outside the scope of the transfer. Thus, Parliament retains the powers of nomination of persons to the government or other offices (eg the Ombudsman, judges of the TK, etc). Also the powers of scrutiny and investigation can still be exercised in the form of parliamentary scrutiny or investigative commissions or through such instruments as the office of the Supreme Chamber of Control (Najwyższa Izba Kontroli). EU policy reflects and extends the original internal policy implemented by the legislative and executive branch. The transfer of the powers of the executive branch, on the other hand, pertains mostly to limiting the attributions of particular Ministers. It stems as much from the material scope of primary EU law as from the evolution of the scope of EU activity. Judicial matters have obviously been included within EU law and EU regulations make inroads into the autonomy of domestic administration of justice with respect to material and procedural law and the regulation of the conflict of laws. While it is difficult to trace the transfer of judicial powers, there is no doubt it includes the transfer of judicial powers of interpretation and the transfer of judicial powers where national courts lose their ‘natural’ powers that are now exercised by the European judiciary. The powers of state authorities can only be transferred in relations to ‘certain matters’. This term has been convincingly specified in the TK case law.109 First of all, it is prohibited to transfer the total attributions of given state authority, to transfer exhaustively all the powers in relation to one certain matter, as well as to transfer the powers in relation to certain matters where those exhaust the function of certain state authority. Secondly, it is constitutionally impossible to ‘hollow out’ a state authority in a way extinguishing its very raison d’être or its functions. Thirdly, the transfer of powers cannot be effectuated by a carte blanche clause. Any specific transfer of powers has to be consistent with the obligation of safeguarding the essence of the constitutional identity of the Polish state. Fourthly, there is no basis for the transfer of the power to transfer powers (‘Kompetenz-Kompetenz’). Any extension of the powers transferred, on the other hand, requires a prior legal basis in the international agreement, made in compliance with the applicable conditions. Finally, the constitutional identity (see below) is co-determined by the limits of legitimate transfer of powers to an international organisation.

B.  The Procedures for the Delegation of Powers The Constitution regulates two procedures in which the President of the Republic can consent to ratify an international agreement transferring certain authority to an international organisation. One procedure requires the consent expressed in the form of a statute, the other requires the consent expressed in the national referendum. The Sejm is the only authority empowered to initiate one or the other procedure. Its decision is made in the form of a resolution voted by an absolute majority in the presence of at least half of the statutory number of Deputies (230). The statute consenting to the ratification of an international agreement can only originate in a bill sponsored by the Government, because only the Government is empowered to make

109 TK

K 18/04 (n 62).

Poland  603 international agreements that are subject to ratification. The bill should be subject to a debate at a special ad hoc committee of the Sejm. It is not permissible that the statute is adopted in the fast-track procedure, because this legislative procedure cannot be applied to statutes governing the organisation or competences of public authorities. The statute to enable the President to ratify an international agreement transferring powers is adopted by the Sejm by a two-thirds majority with at least half of the statutory number of Deputies voting. The same supermajority is required for a constitutional amendment. The qualified-­majority requirement translates into the requirement of building a wide consensus across the political spectrum. In Polish political reality, the supermajority requirement commands cooperation of the parliamentary majority with the opposition.110 Once adopted by the Sejm, the Senate votes the statute again by a two-thirds majority with at least half of the statutory number of Senators voting (50). This supermajority requirement exceeds even the requirements for constitutional amendments, that can be adopted in the Senate by an absolute majority. This supermajority requirement symbolically thus translates concerns for constitutional integrity and constitutional intimacy. The international agreement transferring powers of the state t­ranslates into its profound reorganisation and has huge implications as a symbolic inroad into state sovereignty that apparently results from Poland’s membership in the EU. The second procedure to enable the President of the Republic to ratify an international agreement transferring powers is based on consent expressed by the citizens in a national referendum. This procedure obviously implements the principle of the sovereignty of the Nation and its adjacent mechanism of direct democracy. The decision to initiate this procedure belongs to the Sejm; its decision to hold a national referendum extinguishes the Parliamentary consent-to-ratification procedure. The referendum has a binding effect when at least half of voters cast their votes and the decision is made by a simple majority (Article 125(3) KRP). If the referendum attracts half of the voters and the majority of votes are in favour, the President of the Republic is obligated to ratify the treaty, once the SCPL declares the referendum valid and confirms the vote-count. It is important to note that the Sejm is exclusively competent for initiating one or the other procedure. The subject of both procedures can only be framed in terms of expressing consent. The referendum, in particular, can only produce two votes. If half of the voters participate and a majority vote in favour, the procedure expresses consent. A refusal to give consent, on the other hand, can only result from a referendum where half of the voters take part and a majority of participants vote against ratification of the international agreement. When the voter turnout is below 50 per cent, the Constitution implicitly provides for the possibility, if not an obligation, to proceed with the alternative, parliamentary procedure designed to enable the President to ratify the international agreement by a statute voted by both chambers of the Parliament in accordance with Article 90(2) KRP.111 The Constitutional Tribunal acquiesced in this view concerning the obligation to initiate the parliamentary ratification procedure. The Constitutional Tribunal found the legal basis for such an obligation in the fact that the Constitution commands that the sovereign takes a legally binding decision in one or the other procedures. The refusal to initiate the parliamentary enabling procedure where there is no binding decision thus runs counter to constitutional design. The Polish EU Accession Treaty was signed in 2003 and the Sejm initiated the national referendum ratification procedure. The referendum took place on 7–8 June 2003, asking ‘Do you

110 From 1991–2015 no political party enjoyed an absolute majority in the Sejm. In 2015, PiS was the first party to attain absolute majority and form a government autonomously. 111 TK K 11/03 (n 60).

604  Mirosław Wyrzykowski and Michał Ziółkowski consent for the Republic of Poland to join the European Union?’ With a turnout of 58.85 per cent the referendum was legally binding, 77.45 per cent of the votes being cast in the affirmative. The referendum thus enabled and obliged the President of the Republic to ratify the Treaty that entered into force on 1 May 2004.

C.  The Status of International Law and EU Law in the Internal Legal Order Poland regulates the position of international law according to the monistic philosophy. The Constitution provides for the norms of ratified international agreements to operate in the domestic legal system as universally binding law. This model is comprehensively regulated by the Constitution. The most general principle commands the observance of international law (Article 9). Technically, international law is integrated into the domestic legal order by conferring it the status of a ‘source’ of universally binding law in Poland (Article 87(1)). The Constitution regulates the function and meaning of ratification by compulsory statutory consent for ratification of five categories of international treaties (Article 89(1)). Further, it states in unambiguous terms that upon ratification and publication in the DU, the international agreement is part of the domestic legal order and thus to be applied directly unless the regulation requires statutory transposition. The Constitution also provides for two conflict-of-norms regulations. Article 91(2) provides for the international agreement to prevail over statutes when the statute cannot be applied in compliance with the agreement. Article 91(3), on the other hand, provides for a complementary regime regarding the application of law created by an international organisation. If a ratified international agreement to which Poland is party so provides, the law enacted by an international organisation is to receive direct effect and prevail over statute in cases of conflict. The major legal consequence of the principle of observance of international law as implemented within the 1997 Constitution is that the Constitution itself presumes that the legal system binding on the territory of the Republic of Poland is composed of the norms (legal provisions) created by the national legislator as well as rules (legal provisions) created outside of the system of domestic (Polish) law-giving organs. The constitutional designer of the system of government purposely assumed that the system of law binding on the territory of the Republic should have a composite character. Alongside legal acts made by domestic law-giving organs, acts of international law are also binding and applicable in Poland’.112

The constitutional principle of primacy of international agreements is not absolute, but finds its limits in the Constitution itself. In every situation and without any further conditions, the principle ceases to operate in order to give effect to the primacy of the Constitution. The Constitution is, and remains, the supreme law of the Republic of Poland (Article 8) in respect to all international agreements. The supremacy of the Constitution within the system of sources of law is explicitly regulated in specific and concrete constitutional regulations. Such specific rules are implicit in the justifications and rules to be found in the Preamble, the norm stipulating that the Constitution shall be the supreme law of the Republic of Poland, the principle of sovereignty of the Nation, the specific procedure of constitutional amendment, the Constitution’s specific individuated ‘proper name’ as well as – finally – the Constitution’s



112 TK

K 18/04 (n 62).

Poland  605 contents specifying the system of government of the state, the sovereign and the ways of exercising the sovereign’s power and the fundamental rights, freedoms and obligations of the individual.113

D.  Practice and Doctrinal Debate on these Provisions As stated above, EU law is part of the domestic legal order and is accordingly governed by all constitutional norms, including the principle of supremacy of the Constitution. The TK could not be any clearer in stating that the very idea and model of European law created a new situation where two autonomous legal orders are valid and applicable in the same territory at the same time. Their relationship cannot be described fully with reference to the traditional ideas of monism and dualism and its accessory dichotomy of the domestic and the international legal order. The fact that relatively autonomous legal orders governed by their respective internal principles of hierarchy co-exist does not mean that such legal orders are separate. They do interrelate. The relative autonomy of legal orders also does not foreclose the possibility of collisions between EU law and the Constitution. This situation of conflict can certainly arise in situations where there is an irreconcilable contradiction between the norm of the Constitution and a norm of European Union law and the contradiction is of such a nature as to be not amenable to be resolved by construing domestic law in a way that is respectful of the relative autonomy of domestic law, European law and the dispositions of the Constitution. Such a situation is theoretically possible even though the community of premises and values that underscores both systems of law suggests that it can only materialize in exceptional circumstances. In any event, in the Polish legal system, such a contradiction can never be resolved by resorting to the principle of primacy of European law with respect to the norm of the Constitution. Nor can such a contradiction be resolved by any jurisprudential devices that would remove the validity of a binding constitutional norm by supplanting it by a European norm or restrict the validity of the constitutional norm to the realm outside the scope of European law.114

In such a situation of irreconcilable contradiction between European law and Polish constitutional law, only one solution can be adopted. The TK has upheld the following formulation that was voiced in a judgment of 2005: In such a situation, the Polish legislator is obligated to adopt one of the following solutions. The legislator can amend the constitution so as to remove the irreconcilable contradiction, or is obligated to bring about legislative change in European law and – failing such legislative reform within the EU – take the decision to withdraw membership in the European Union. The decision belongs to the sovereign, i.e. the Polish Nation or a state authority representing the Polish Nation according to the Constitution.115

Such a situation of irreconcilable contradiction has been considered only once by the TK. The Constitution was amended so as to provide a constitutional foundation for the European Arrest Warrant (EAW).116 Article 55 prohibits the extradition of a Polish citizen but now provides for extradition based on the EAW. The TK considered the Code of Criminal Procedure that implemented the EAW’s provisions governing the rendition of persons under the jurisdiction of EU

113 Bogusław Banaszak, Zasada nadrzędności Konstytucji w polskim porządku prawnym [The Principle of Supremacy of the Constitution in the Polish Legal Order] (Wrocław, E-Wydawnictwo, 2015) 47. 114 TK K 18/04 (n 62). 115 Ibid. 116 2002/584/JHA: Council Framework Decision of 13 June 2002 on the European Arrest Warrant and the surrender procedures between Member States, [2002] OJ L 190/1.

606  Mirosław Wyrzykowski and Michał Ziółkowski member states. It stated that the constitutional prohibition of extradition of Polish citizens governed the EAW and would suffer dilution of its protective function unless it was amended to allow the implementation of the EAW.117

III.  Constitutional Limits to EU Integration A.  Constitutional Review of EU Primary Law There is no doubt that constitutional review of primary EU law is not only possible but also only proper. The Constitution explicitly states that there are no exceptions to the power of constitutional review of international agreements and there has never been and there is no scope for debate in this respect. This observation pertains to a priori (preventive) as well as a posteriori (corrective) constitutional review.

B.  Constitutional Review of EU Secondary Law The constitutional status of EU secondary legislation is quite clear. The norms of secondary legislation enjoy primacy in case of conflict with domestic legislation in cases if a contradiction cannot be reconciled by interpreting domestic in way that is sympathetic to European integration. There is no doubt that the TK is competent to review the constitutionality of statutes that implement EU secondary legislation. As has been described, the TK considered the Code of Criminal Procedure that implemented the EAW in 2005. There was much scholarly controversy in European and constitutional law regarding the constitutional review of secondary legislation itself. The matter was apparently also controversial in the reasoning of the TK until the matter was finally settled in the case K 45/09. In the original judgment in case K 18/04, the TK stated that Member States retain the power to evaluate whether or not the law-making authorities of the European Union acted within the powers conferred on them and whether they exercised their powers in accordance with the principle of subsidiarity and proportionality in making a particular legal provision (act). Once this framework has been transgressed, the legal provisions are not subject to the principle of primacy of European law as provided by the Constitution of the Republic of Poland.

This statement has been interpreted as stating that the TK found the legal basis for finding secondary legislation ultra vires. Some years later, however, the Tribunal stated in an obiter dictum that the Constitution does not indicate EU secondary legislation as the object of constitutional review under Article 188(1) and (2) and that the TK thus is not competent to review the constitutionality of secondary legislation.118 There is no clear agreement within the Polish constitutional law community regarding the powers of the TK to undertake constitutional review of secondary EU legislation.119 The main 117 TK P 1/05 (n 63). 118 TK 17.12.2009, Case No U 6/08. 119 For an overview see Monika Krawczyńska, ‘Kontrola zgodności z Konstytucją aktów unijnego prawa pochodnego w postępowaniach przed Trybunałem Konstytucyjnym’ [‘Constitutional Review of Secondary EU Legislation before the TK’], in S Dudzik, N Półtorak (eds), Prawo Unii Europejskiej a prawo konstytucyjne państw członkowskich [The Law of the European Union and the Constitutional Law of Member States] (Warszawa, Wolters Kluwer, 2013) 220.

Poland  607 argument against such powers is based on the observation that the Constitution is exhaustive in regulating the powers of the TK and that the Constitution thus explicitly does not provide for the TK to consider secondary EU legislation. Another argument is made considering the relationship between the provisions of the Constitution governing the general powers of the TK and the provisions governing the constitutional complaint. A constitutional complaint is thus considered an instrument of protection of rights and freedoms of the individual within the limits set by the general powers of the TK. The argument is made that the same is true for proceedings initiated by general and administrative courts within the constitutional preliminary ruling procedure. Many scholars object and support the view that TK has the power to review secondary EU legislation. The argument again draws on the Constitution providing for special regulation of a constitutional complaint. The main argument is formal, observing that the constitutional complaint does not specify the formal catalogue of acts that are subject to review in the constitutional complaint procedure. Accordingly, any legal Act whatsoever can be targeted by a constitutional complaint as long as this Act has been applied by administrative authorities and courts, thereby allegedly abridging constitutional rights or freedoms. It follows that EU secondary legislation can also be subject to constitutional review in the constitutional complaint procedure. Another argument purports the competence of the TK to review the constitutionality of secondary legislation based on the comprehensive nature of constitutional review of norms binding in Poland. The theory proceeds to affirm that the KRP does not tolerate normative acts subject to the Constitution but being immune from being considered by the TK.120 The matter of constitutional review of secondary legislation has been settled in the TK judgment of 6 November 2011 that originated in the constitutional complaint case SK 45/09. The TK the constitutionality of the norm stipulated in Article 41 of Council Regulation No 44/2001.121 Pursuant to this Article, a judgment should be declared enforceable immediately upon the completion of formalities stated in Article 53 and the party against whom enforcement is sought is not – at this stage of the proceedings – entitled to make any submissions on the application. The TK stated that the norm is concordant with the constitutional right to a court (Article 45(1) KRP) and the principle of equality (Article 32). The TK commenced its reasoning, however, by ascertaining that the matter should and can be considered by the TK. After all, the Tribunal was confronted with the constitutional review of a Council Regulation initiated by a constitutional complaint. It was therefore imperative to decide if EU secondary legislation remained within the meaning of a ‘normative Act’ that the Constitution allows reviewing in cases initiated by a constitutional complaint. The TK did not have to engage in soul-searching because a line of case law already considered the notion of a ‘normative Act’ in its material aspect. All it takes for a legal action to fall within the meaning of a ‘normative Act’ is that the Act contains general and abstract legal norms. Any such normative Act falls within TK’s jurisdiction whenever rights and freedoms are concerned and the case is initiated by a constitutional complaint. Considering this interpretation of the term ‘normative Act’ appropriate, the TK stated that EU law does indeed provide for such abstract and general norms.122 In the case considered by the Tribunal, the norm subject to constitutional complaint was created by a Regulation, and Article 288(2) TFEU states that ‘[a] regulation shall have general application.

120 Marcin Wiącek, ‘Glosa do postanowienia TK z dnia 17 grudnia 2009 r. U 6/08’ [‘Commentary on Resolution of the TK of 17 December 2009 U 6/08’], 6 Państwo i Prawo 2010, 123. 121 Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters ([2001] OJ L 012/1). 122 Art 288 para 2–4 TFEU.

608  Mirosław Wyrzykowski and Michał Ziółkowski It shall be binding in its entirety and directly applicable in all Member States’. Regulations are thus legal Acts that have a general character in relation to the facts they regulate and are abstract in relation to the subjects bound by its stipulations. To this extent, therefore, the regulation falls well within the matter covered by the notion of a ‘normative Act’ as an Act containing legal norms of a general and abstract character. The term ‘normative Act’, on the other hand, appears in the Constitution not only in its regulation of the scope of the constitutional complaint (Arts 79 and 188(5)) but also in its regulations of the powers of the TK (Arts 190–91) as well as in relation to the constitutional provisions regulating the essence of the request for a preliminary ruling constitutional review requests directed to the TK by the courts (Article 193). The TK thus had to determine its power to review the constitutionality of such a normative Act in respect to its general powers. The preliminary issue concerned the question, whether the fact that the Constitution does not indicate secondary EU legislation as a category or kind of law subject to constitutional review suggests that the TK does not have the power to review EU secondary legislation in the constitutional complaint procedure. The TK indicated that its powers should be construed under systemic interpretation, taking account of a number of arguments. First, the material scope of normative acts that are subject to constitutional review was autonomously and independently regulated in the dispositions regulating the general powers of the TK on the one hand, and in the separate disposition regulating the constitutional complaint on the other. Secondly, the systematic analysis of the Constitution conclusively suggests that the regulation of the procedure initiated by the constitutional complaint is separate from the regulation of the general powers of the TK. Finally, any restrictive construction of the powers of the TK on general and special regulations would run counter to its position within the system of government that provides that the TK is the guarantor of the Constitution as well as of the constitutional rights and freedoms of the individual. The powers of the TK to review the constitutionality of EU secondary legislation were obviously contingent not only on accepting that the Polish Constitution’s term ‘normative Act’ covers EU regulations. The Constitution does not offer protection of constitutional rights and freedoms by the TK unless the constitutional complaint identifies a normative Act that was the basis of a final decision made by a court or the administration. Such a final decision is to be alleged to abridge a constitutional right or freedom. In other words, the normative Act can only be submitted to constitutional review where it is the basis of a final decision possibly infringing an individual’s constitutional right or freedom. Whereas the term ‘normative Act’ is wide, the additional condition for constitutional review in the constitutional complaint procedure restraints the reach of constitutional review of secondary EU law to acts that are capable of direct effect. Whilst even regulations need to satisfy the criteria for direct effects, in principle EU regulations and decisions can be subject to constitutional review, whereas directives can be subject to constitutional review only to the extent they are clear, precise and unconditional, so as to produce direct effect. The TK could not be any clearer: ‘the constitutional complaint can allege the unconstitutionality of regulations and decisions, and directives in special circumstances’.123 The TK obviously was aware of the trailblazing character of this judgment. It provided for constitutional review of secondary EU legislation in the constitutional complaint procedure. It is hardly surprising, therefore, that the reasons of the decision set out a legal and axiological framework that should govern and restrain constitutional review of secondary EU law. The TK indicated the special position of EU secondary legislation, the special rank accorded to fundamental rights



123 Krawczyńska,

‘Kontrola’ (n 118) 235.

Poland  609 in the EU legal order, the constitutional principle of consideration for European integration (europafreundliche Auslegung) and the principle of the loyalty of Member States towards the EU. Those elements constitute the general normative and axiological framework that governs and limits the operation of constitutional review of secondary EU legislation in the constitutional complaint procedure. This framework translates into specific conditions that the TK adjoins to the procedure of constitutional complaint where it alleges the abridgement of a constitutional right or freedom by the application of secondary EU law. The applicant is required to show not only that the constitutional right or freedom was violated and that the EU secondary law that was at the origin of such a violation is materially noncompliant with the Constitution, but also to demonstrate convincingly that the EU Act of secondary legislation substantially lowers the standard of protection of rights or freedoms in comparison with the standard of protection offered by the KRP. Presenting such evidence and argumentation is part and parcel of the showing of the way in which the constitutional right of freedom was infringed. Furthermore, the requirement to make probable that the level of protection of rights and freedoms has been lowered, in comparison with the level of protection guaranteed by the Constitution, follows from the allocation of the burden of proof in review proceedings commenced by [the] way of constitutional complaint.124

C.  The Effects of the TK Judgment Declaring Secondary EU Law Unconstitutional The TK decision SK 45/09 is a leading case in this area even though the EU regulation was declared consistent with the Constitution. The TK, however, did not shrink from considering the effects of a hypothetical declaration of unconstitutionality in respect of secondary legislation. The Tribunal stated that any such matter should be considered with one basic proposition in the heart, namely that the TK’s powers to review of secondary EU legislation are subsidiary. The TK, in other words, acquiesced in the general setup of the EU division of powers between CJEU and constitutional courts in respect to judicial review of legislation. The CJEU has the exclusive powers of binding interpretation of EU law, the exclusive position to further ‘uniform application and interpretation of European Union law’ and the exclusive power of judicial review of secondary legislation with the Treaties and general principles of EU law. In any case, if a constitutional court has doubts as to the meaning of EU law that cannot be reasonably overcome by interpretation, the constitutional court should make a preliminary reference to the CJEU. If the CJEU determines that the meaning of a particular Act is such that there is no longer incompatibility with the Polish Constitution or the if it declares the secondary law purported to violate the Polish Constitution is non-consistent in respect to primary EU law and thus void, the case should be discontinued by the TK.125 In principle, the declaration of unconstitutionality of a normative Act entails the abrogation of the norm deemed unconstitutional. This is not possible with respect to normative acts that belong to the EU legal order: the CJEU is the only authority empowered to produce such an effect. It would thus seem that a declaration of unconstitutionality by the TK should produce the effect of barring the application of secondary EU legislation deemed unconstitutional by the domestic authorities. Far from voiding the norm, the effect would limit the territorial applicability of this



124 TK

SK 45/09 (n 66) para 8.5. para 2.6.

125 Ibid,

610  Mirosław Wyrzykowski and Michał Ziółkowski unconstitutional norm provided by EU secondary legislation. The TK envisaged such an effect and considered it not only unfortunate and but also unacceptable. The TK thus excluded such an interpretation of the Constitution so as to give effect to the obligations arising under the Treaties under Arts 258–60 TFEU and under the principle of loyalty. The TK considered that the effects of its judgment should operate neither on validity nor on the applicability of the norm of secondary EU legislation. Technically, therefore, the TK considered that the Constitution requires it to defer the effects of its judgment to allow space for the political process to restore constitutionality by legislative or constitutional amendment. After declaring the non-conformity of particular norms of EU secondary legislation to the Constitution, measures should be undertaken forthwith in order to eliminate the conflict. The constitutional principle of the favourable predisposition of the Republic of Poland towards the process of European integration and the Treaty’s principle of the loyalty of the Member States towards the Union requires that the effects of the Tribunal’s ruling be deferred in time, pursuant to Article 190(3) of the Constitution.126

This analysis of the leading case in which the TK declared itself competent to review the constitutionality of secondary EU legislation in constitutional complaints would be incomplete unless complemented by the following observations that are essential for understanding the mechanism of constitutional review as well as the effects of the judgment for the applicant.127 First it is important to emphasise that the constitutional complaint is the only procedure in which the TK is competent to undertake a constitutional review of secondary EU legislation. The Tribunal expressly stated that the Constitution does not provide for constitutional review in the constitutional preliminary reference procedure or in the abstract constitutional review procedure. This is important despite the possibility that legislation might be alleged to violate constitutional rights or freedoms but will not be subject to constitutional review unless initiated by an individual in the constitutional complaint procedure. Secondly, deferring the effect of the declaration of unconstitutionality bars the possibility of reopening proceedings in the case. This is the practical side of things because this deferment is provided by the Constitution to allow the legislator to amend unconstitutional legislation and give effect to the judgment of the TK. The legislative amendment, however, substitutes the effect of the declaration of unconstitutionality and formally the unconstitutional norm disappears independently of the TK judgment. In this situation, the constitutionality of legislation results from a legislative amendment rather than invalidation by the TK. The proceedings, therefore, cannot be reopened and the decision violating constitutional rights cannot be quashed as provided by Article 190(4) KRP, because there is no formal invalidation of the norm by the TK. In the scenario of a constitutional amendment made before the judgment takes effect, this amendment would entail an implicit validation of the hitherto unconstitutional norm and an implicit invalidation of the invalidation. Thirdly, the right to reopen proceedings or quash a decision is largely theoretical in cases where the TK judgment takes the form of a judgment on conforming interpretation. This situation concerns two techniques of adjudication called interpretative judgment and partial

126 Ibid, para 2.7. 127 Agnieszka Kastelik-Smaza, ‘Skutki stwierdzenia przez Trybunał Konstytucyjny niezgodności z Konstytucją unijnego prawa pochodnego z punktu widzenia skarżącego’ [‘The Effects of TK’s Decision on Unconstitutionality of Secondary EU Law in the Eyes of the Applicant’], in S Dudzik, N Półtorak (eds), Prawo Unii Europejskiej a prawo konstytucyjne państw członkowskich [The Law of the European Union and the Constitutional Law of Member States] (Warszawa, Wolters Kluwer, 2013) 244.

Poland  611 judgment. The former case relates to judgments where the TK declares a specified interpretation of the legal disposition unconstitutional (ie the TK specifies that the specific understanding of the wording of a normative Act is unconstitutional), the latter relates to judgments in which the Tribunal declares unconstitutional only specified words or wording (ie the TK specifies the unconstitutional phrase or words). This unfortunate situation results directly from a line of SCPL cases expressing the theory that the SCPL has the exclusive power of interpretation and that the TK no longer has the powers of interpretative judgments under KRP of 1997. Finally, it is practically impossible to find a legal basis to claim compensation for the so-called ‘statutory unlawfulness’ of Polish or EU authorities, since ‘the unconstitutionality of the law of the European Union in a particular Member State does not fall within the ambit of European Union liability’.128 These observations seem to corroborate the statement that ‘a constitutional complaint filed against EU secondary legislation may in certain cases contribute to the review of the conformity of the Constitution to the EU law rather than the another way round, and result in adjusting the constitutional standard to the EU standard’.129

IV.  Constitutional Rules and/or Practice on Implementing EU Law A.  Conflicts of Authority Poland’s EU membership transforms many aspects of the domestic system of government. One such aspect concerns the understanding of the meaning of the dual executive established by the Constitution. The 1997 constitutional design includes the distribution of executive power between the President of the Republic and the Council of Ministers (Prime Minister). In 2007, there was a conflict of powers between those two authorities regarding the power to represent Poland at the European Council. This dispute arose against a peculiar background: In 2006–07, the Government was backed by a coalition parliamentary majority uniting the parties of PiS, Samoobrona (Self-Defence) and Liga Polskich Rodzin (League of Polish Families). PiS was the dominant party and nominated the Prime Minister, Kazimierz Marcinkiewicz, who was soon replaced by Jarosław Kaczyński, the president of the dominant party. Furthermore, just before the parliamentary elections, his twin brother Lech Kaczyński was elected President of the Republic. He was also politically affiliated to the PiS party. The situation thus saw both constitutional authorities forming the dualist executive representing one political party. The look-alikes in both offices of the dual executive were quick to short-circuit any conflict of powers. It was established, for example, that the President should represent Poland at the European Council and the Prime Minister should represent the country at other fora, such as NATO. This is an example of implementing the principle of cooperation of the public authorities, although facilitated by the trust emanating from family bonds. In autumn 2007, however, parliamentary elections changed the political situation. The Government was formed by the dominant Platforma Obywatelska (Civic Platform) party and the marginal Polskie Stronnictwo Ludowe (Polish Folk Party). Donald Tusk, President of Platforma Obywatelska, was at the helm of the government. Shortly, the new Government found itself in conflict with the



128 Ibid,

257.

129 Dudzik/Półtorak,

‘Court of Last Word’ (n 65) 256.

612  Mirosław Wyrzykowski and Michał Ziółkowski President of the Republic concerning the authority to represent Poland at the European Council. Both authorities claimed the exclusive power to represent the country at the European Council. The Prime Minister petitioned the TK to adjudicate on the distribution of powers between the two offices of the executive. The TK asserted its authority over the question of authority because it was a material constitutional problem regardless of its significant political and emotional import. The essence of the constitutional question concerned two constitutional matters: First the ‘determination’ of the central constitutional authority of the Republic, empowered to represent the state at the European Council. The second concerned ‘presenting’ the stance of the Republic of Poland at the European Council. The problem was related to the situation that the President intended to take part in the European Council against the will of the Council of Ministers. The Constitution does not expressly regulate the matter. The TK was thus confronted with the task of a systemic interpretation of the Constitution. The analysis of the Constitution led to the following conclusions. The Council of Ministers conducts internal and foreign policy according to Article 146(1) KRP, whereas EU affairs are linked to internal affairs in the traditional understanding of the term. Depending on the subject matter discussed at the European Council, therefore, the powers of the President of the Republic suffer limitation, whereas the Council of Ministers can conduct internal policy and be politically accountable. The Prime Minister is responsible to implement the policies of the Council of Ministers and make all decisions as to its ways and means. Under Article 148 KRP, s/he has the power to ‘represent the Council of Ministers’ (item 1) and to ‘ensure the implementation of the policies adopted by the Council of Ministers and specify the manner of their implementation’ (item 4). The powers of the Prime Minister are, therefore, implicit in the authority’s constitutional function, also if the Council of Ministers decides its representation at the European Council. The Council of Ministers exercises general control in the field of external relations (Article 146(4) item 9). Further, there is a presumption of non-specified powers in favour of the Council of Ministers (‘all the affairs of State not reserved to other State organs or local government’, Article 146(2)) that supplements the general power of conducting the internal affairs and foreign policy (Article 146(1)). It is, therefore, the exclusive power of the Council of Ministers to determine the position of the Republic of Poland to be adopted at the European Council. The participation of the Prime Minister at the European Council, on the other hand, flows directly and naturally from the Prime Minister’s functions in the Government as established by the Constitution. The TK stated, moreover, that the President of the Republic has no autonomous constitutional powers regarding foreign policy and relations of the state with international organisations. This statement was important because it contradicted the view expressed in the memorandum presented by the then President of the Republic. According to the Constitution, the President is the ‘the supreme representative of the Republic of Poland and the guarantor of the continuity of State authority’ (Article 126(1)). This status is accompanied by the duty to ‘ensure observance of the Constitution, safeguard the sovereignty and security of the State as well as the inviolability and integrity of its territory’ (Article 116(2)). Those duties, however, do not imply any prerogatives or powers. To the contrary, the President of the Republic is obliged by the Constitution to cooperate and communicate with other constitutional authorities. The President’s conduct, furthermore, has to be based on and constrained to the norms that regulate the powers and duties of the President of the Republic. The TK indicated that there is very little scope for EU membership to trigger presidential powers related to dangers to ‘sovereignty and security of the State’ or the ‘inviolability and integrity of its territory’. Unless such circumstances arise, there is no competence of the President of the Republic to attend meetings of the European Council.

Poland  613 The function of the ‘guarantor of the constitution’, on the other hand, cannot justify the President’s intrusion into the domain of the Council of Ministers. This function is specified by powers relating to internal affairs, such as the power to initiate constitutional review of statutes consenting to the ratification of EU Treaties. The only instance where the ‘guarantor of the constitution’ function of the President of the Republic implies the power to attend meetings of the European Council is when the Council’s agenda includes amending the EU Treaties, but only if those amendments encroach upon the sovereignty of the Republic. It is important to note that the exceptional circumstances empowering the President to attend the meeting of the European Council does not give rise to any prerogative of the President. To the contrary, the decision that the President of the Republic should attend the meeting has to be made jointly by the President and the Prime Minister. The President, on the other hand, is obliged to present the position of Poland determined by the government. It is natural that this position will be subject to prior understanding reached in accordance with the principle of cooperation of constitutional authorities.130 The resolution of the TK settled this question of authority that characterised the situation of cohabitation, when the dualist executive was in radical discord as to the conduct of European affairs. No such conflict has arisen ever since.

B.  Attempts at Constitutional Reform As mentioned above, the European Union clause in the Constitution of 1997 reflected a political compromise and also the contemporary understanding of the EU’s character. It was understood, however, that the clause is way too modest a constitutional basis for consequences and processes brought about by the progress of European integration or for the consequences related to EU powers and procedures within the Polish system of government. In 2009–10, attempts were made to draft a special ‘European Union Chapter’ that eventually could become part of the Constitution. The initiative came from parliamentary parties131 and the President of the Republic.132 A special Constitutional Committee was established as a forum for hammering out a modicum of political compromise capable to command the constitutional majority in the Parliament. The Committee presented a report and a draft of the constitutional amendment.133 The draft constitutional amendment presented by the Committee was a comprehensive regulation to govern Polish EU membership. It should be inserted into the Constitution as a new Chapter Xa, titled ‘Membership of the Republic of Poland in the European Union’. The draft itemised specific regulations proposing to establish a clear-cut constitutional basis for EU membership, which it called the ‘limitation clause’.134 This ‘limitation clause’ would stipulate the supremacy of the KRP. The supremacy of the Constitution is already stipulated as a principle

130 CT, decision of 20 May 2009, Case No Kpt 2/08, paras 4.3, 5.5, 5.7–5.8 and 6.1. 131 ‘Bill Amending the Constitution of the Republic of Poland’ (Druk sejmowy [Sejm Paper, DS] 2989), submitted by the Civic Platform party and three bills submitted by the PiS (‘Bill amending the Act on the Constitutional Tribunal and the Act on Publishing Normative Acts and other Acts’ [DS 3395]; ‘Bill Amending the Constitution of the Republic of Poland’ [Druk sejmowy 3399]; ‘Bill Amending the Constitution of the Republic of Poland’ [DS 3687]). 132 ‘Draft Bill Amending the Constitution of the Republic of Poland’ (DS 3598). 133 DS 4450 and DS 4450-A. 134 The analysis follows Zmiany Konstytucji Rzeczypospolitej Polskiej w latach 1997–2011 w świetle projektów ustaw oraz uchwalonych nowelizacji [Draft Amendments to the KRP in the years 1997–2011 in the Light of Statutes and Amendments Passed], Biuro Analiz i Dokumentacji OT – 605, Kancelaria Senatu (2011) 9.

614  Mirosław Wyrzykowski and Michał Ziółkowski in Article 8 KRP; the new clause proposed to specify that supremacy covers domestic and external legal acts. The drafters also sought to emphasise the principle that Polish membership in the EU serves the purposes of the Constitution and violates none of its fundamental principles. The proposed clause also mentioned the principle according to which the EU respects the sovereignty and national identities of Member States. The draft proposed to provide the constitutional basis for the delegation of competences in certain matters and provided for amending Article 90 KRP, that was to be no part of Chapter Xa, but would need to be coordinated with the suggested regulations. The proposal also sought to establish new procedures to express consent to the ratification of international treaties transferring the exercise of state authority in certain matters onto the EU. Chapter Xa was to establish principles of decision-making regarding the position of Poland in in the simplified Treaty change procedure under Article 48(6) TEU. Finally, the draft set out the constitutional bases necessary for the Sejm and the Senate to exercise the powers conferred on Parliaments by the Lisbon Treaty. The Committee also proposed to determine the pertinent competences of the Council of Ministers, the President and other public authorities. Interestingly, the Committee also sought to regulate the competences to decide on a possible withdrawal from the EU, to entrench the right of Polish citizens to vote and stand as candidates in European elections and the right of foreign EU citizens residing in Poland to vote and stand as candidates in the European Parliament and local self-government elections. The report and draft marked the demise of the initiative of the constitutional amendment. The Parliament elected in 2011 did not take any interest in the matter.

C.  The Principle of Interpretation Favourable to European Integration The TK has never departed from its original case law that stated that the judicial branch is obliged to develop jurisprudential mechanisms integrating both the basic principles of EU law and the Polish constitutional design. The Tribunal, on the other hand, always accepted the principles of primacy of EU law, of uniform and effective application of EU law, and of loyal cooperation. It also emphasised the exclusive powers of the CJEU to interpret and review EU law. The domestic constitutional setup, on the other hand, was considered limited to the principles of supremacy of the Constitution and of safeguarding Poland’s constitutional identity. To reconcile these principles, the TK relies on the concept of interpretation favourable to European integration. EU-sympathetic interpretation originated in the very first EU decisions of the TK, has not suffered any evolution in terms of constitutional law and is rooted in several fundamental elements of the constitutional system. The KPR’s Preamble emphasises the ‘need for cooperation with all countries for the good of the Human Family’. The basic principles of the Republic entrenched in Chapter I command observance of international law binding the Republic (Article 9) and respect for the Constitution of the Republic as the supreme and directly applicable law in the Republic (Article 8).135 The constitutional basis of the duty of EU law conforming interpretation (sometimes termed ‘EU-sympathetic’, ‘European-law-friendly’ ‘pro-European’ interpretation) grants the TK considerable discretion in establishing its meaning and scope. In particular, the meaning of ‘constitutional identity’ limiting the transfer of powers to the EU allows for considerable leeway. Limits of EU-conforming interpretation of domestic law were first drawn in the EAW case, when the TK stated that

135 TK

K 11/03 (n 60).

Poland  615 the obligation of EU-conforming interpretation of domestic law has limits. It might be noted that those limits have already been indicted by the European Court of Justice. When EU-conforming interpretation could lower the status of the individual, and in particular when the matter concerns matters of criminalization or matters of criminal responsibility, EU-law conforming interpretation expires. There is no doubt that rendition of a person under the European Arrest Warrant procedure to a foreign state in relation to the person’s action that is not criminal in Poland lowers the legal protection of the person thus prosecuted.136

The TK subscribed to the same analysis when declaring the constitutionality of the Polish Accession Treaty. The Tribunal stated that the principle of interpreting domestic law in a manner ‘sympathetic to European law’ as formulated within the Constitutional Tribunal’s jurisprudence, has its limits. In no event may it lead to results contradicting the explicit wording of constitutional norms or being irreconcilable with the minimum guarantee functions realised by the Constitution. In particular, the norm of the Constitution within the field of individual rights and freedoms indicate a minimum and unsurpassable threshold which may never be lowered or questioned as a result of the introduction of Community provisions.137

The reasoning formulates a principle that henceforth governed conflict-of-norm situations: Such a collision would occur in the event that an irreconcilable inconsistency appeared between a constitutional norm and a Community norm of such a nature that the contradiction could not be eliminated by means of applying an interpretation which respects the mutual autonomy of European law and national law. Such a collision may in no event be resolved by assuming the supremacy of a Community norm over a constitutional norm. Furthermore, it may not lead to the situation whereby a constitutional norm loses its binding force and is substituted by a Community norm or whereby the constitutional norm’s scope of application is restricted to areas beyond the scope of Community law. In such an event, the Nation as the sovereign, or a State authority organ authorized by the Constitution to represent the Nation, would need to decide on amending the Constitution, bringing modifications in Community provision or, ultimately, on Poland’s withdrawal from the European Union.138

D.  Constitutional Identity139 Constitutional identity is not a concept directly provided for in the KRP. The notion was not encountered in prior constitutional provisions or treated extensively in the constitutional literature before 1997. When ruling on the constitutionality of the Treaty of Lisbon140 and secondary EU law,141 and assessing the constitutionality of not notifying national regulations to EU bodies,142 the TK has employed the expression ‘constitutional identity’ (tożsamość konstytucyjna), while at the same time attempting to outline a new framework for dialogue with the CJEU.143

136 TK P 1/05 (n 63), para 3.4. 137 TK K 18/04 (n 62). 138 Ibid. 139 This section is based on Michał Ziółkowski’s contribution to the never-published draft-paper (The Politics of Constitutional Identity. Between Constitutional Essentials and Unconstitutional Capture, Polish experience A.D. 2017), presented on the ICON Society annual conference in Copenhagen in 2017. 140 TK K 32/09 (n 64). 141 TK SK 45/09 (n 66). 142 TK 11.03.2015, Case No P 4/14. 143 Krzysztof Wójtowicz, Constitutional courts and European Union law, (Wrocław, E-Wydawnictwo, 2014) 155.

616  Mirosław Wyrzykowski and Michał Ziółkowski In TK case law, ‘identity’ appears most often in three meanings: a) as a synonym for equivalence;144 b) as the equivalent of the boundaries for exercise of competencies;145 c) as a synonym for the essence of a constitutional right of an individual.146 These notions of ‘identity’ correspond to the various meanings that have been ascribed to ‘constitutional identity’ in the constitutional literature, equating it with the appropriate method of argumentation of constitutional courts, or regarding it as a set of fundamental values or principles of a legal system, to the identification of ‘constitutional identity’ as a norm or set of norms comparatively constant in a legal system.147 These meanings correspond to the contexts in which the constitutional courts cite ‘constitutional identity’ when ruling on the constitutionality of actions by the constitutional legislature, conducting comparative analyses and adaptation of foreign solutions within their own jurisdiction, and setting the bounds for delegation of competencies to supranational and international organisations.148 In the EU context, the term ‘constitutional identity’ appeared for the first time in Case K 32/09 – the Treaty of Lisbon case – in a reference to the state as a subject of international law under Article 90(1) KRP. Citing the views in the legal literature, the Tribunal recognised that ‘constitutional identity’ is a concept laying down the scope of ‘exclusion from the competence of delegating matters belonging … to the “hard core”, cardinal for the foundations of the system of the given state’, the delegation of which would not be possible on the basis of Article 90 KRP. Notwithstanding the difficulties inherent in establishing a detailed catalogue of non-delegable competencies, the provisions defining the paramount principles of the Constitution and the provisions concerning the rights of the individual defining the identity of the state should be counted among the matters subject to a complete prohibition of delegation, including more particularly the requirement of ensuring protection of human dignity and constitutional rights, the principle of sovereignty, the principle of democracy, the principle of the rule of law, the principle of social justice, the principle of subsidiarity, as well as the requirement to ensure better realization of constitutional values and a prohibition on delegating legislative authority and competencies to create competencies.149

Citing Article 4(2) TEU and the phrase ‘national identity’ used there, the TK stated: ‘[t]he counterpart of the concept of constitutional identity in primary European law is the concept of national identity’.150 In SK 45/09 the Tribunal stated that constitutional identity is an ‘essential component of the Treaty concept referred to in Article 4(2) TEU,151 although the Tribunal’s statements do not provide sufficient grounds to equate the constitutional term ‘constitutional identity’ with the Treaty notion of ‘national identity’ or the EU concept of ‘constitutional identity’ derived from it. Apart from the autonomy of constitutional concepts and the separateness of the area of constitutional law, the scope of the EU requirement to respect the national identity and the scope of the constitutional requirement to respect constitutional identity are not the same. Article 4(2) TEU

144 TK 27.03.2007, Case No SK 3/05, item 32; TK SK 62/08 (n 107), item 22. 145 TK 26.11.2001, Case No K 2/00, item 254. 146 TK 22.10.2001, Case No SK 16/01, item 214; TK 15.05.2006, Case No P 32/05, item 56; Anna Śledzińska-Simon and Michał Ziółkowski, ‘Constitutional Identity in Poland: Is the Emperor Putting On the Old Clothes of Sovereignty?’, in C Calliess and G van der Schyff (eds), Constitutional Identity in a Europe of Multilevel Constitutionalism (Cambridge, Cambridge University Press, 2019) 248. 147 Anna Śledzińska-Simon, ‘Constitutional identity in 3D: a model of individual, relational and collective self and its application in Poland’, (2015) 13 International Journal of Constitutional Law 124. 148 Ibid, 128. 149 TK K 32/09 (n 64), point III.2.1 of the justification. 150 Ibid. 151 TK SK 45/09 (n 66).

Poland  617 recognises a national identity which is ‘inherent’ only in the ‘fundamental structures, political and constitutional’ of the member states.152 For example, in Michaniki, constitutional provisions on the prohibition of concentrations of media enterprises were not found by the Court of Justice to qualify as an element of the constitutional identity of Greece, despite arguments presented by the Greek Government.153 Nor is every instance will constitutional provisions on innate and inalienable human dignity be included in the constitutional identity of a member state as is the case with the German Constitution.154 The normative scope of constitutional identity in the understanding adopted by the TK is delineated from provisions of the Constitution, but the Tribunal itself does not limit it to fundamental norms or institutions. Second, the addressees of the command to respect the constitutional identity and the Treaty command to respect national identity are different. The TK indicates that constitutional identity, on the one hand, establishes boundaries of competence for the Polish legislature, and on the other hand empowers the Tribunal to evaluate whether these bounds were overstepped. In turn, the Treaty notion of national identity, on one hand, contributes to drawing boundaries of competence for the EU legislature, and on the other hand, addresses the CJEU as the organ standing guard over the realisation of the principle of the primacy of EU law.155 Third, from the perspective of TK case law, protection of constitutional identity takes precedence over a ratified international agreement. But the view of the Tribunal that due to the wording of Article 8(2) KRP, the Constitution ‘enjoys priority in force and application in the territory of the Republic of Poland’156 remains valid. While the conception of constitutional identity adopted by the Tribunal assumes that the legal system in force in Poland is made up of many elements, including constitutionally subordinated subsystems of law originating from various legislative sources,157 the notion of national identity, including constitutional identity, adopted by the CJEU fits within the concept of constitutional pluralism. These notions differ importantly from one another, as, among other things, the former assumes the hierarchical supremacy of force and application of the Constitution, while the latter employs, inter alia, the conception of the primacy of application of Treaty law and the equal weight of the constitutional law of the Member States. Fourth, the command flowing from the TEU and the requirement to protect constitutional identity flowing from the KRP are applied differently in the case law. Under the current case law of the CJEU, the Treaty requirement of respect for national identity, including constitutional identity, does not in itself justify refusal by organs of a Member State to apply EU law which in the view of those organs may infringe such identity.158 Consideration of protection of national identity is sometimes treated as a normatively significant argument in the proportionality test when the CJEU assesses limitations introduced by the legislation of the member states.159 Under the TK case law, the requirement to protect constitutional identity is not considered in the context of the proportionality test as it constitutes an impassable boundary for delegation of competencies. These arguments support a distinction in the analysis in the CJEU and TK case law of concepts of constitutional identity and the Treaty concept of national identity. The TK created

152 See i.a. Elke Cloots, National Identity in EU Law (Oxford, OUP, 2015) 121 f. 153 CJEU 16.12.2008 C-213/07 (Michaniki) ECLI:EU:C:2008:731. 154 Cloots, National Identity (n 152) 169. 155 See Opinion of AG Poiares Maduro, 20.09.2005, CJEU C-53/04 (Marrosu and Sardino) ECLI:EU:C:2005:569 para 40. 156 TK K 32/09 (n 64), point III.2.5 of the justification. 157 TK K 18/04 (n 62). 158 CJEU 22.12.2010 C-208/09 (Sayn-Wittgenstein) ECLI:EU:C:2010:806 para 84; CJEU 12.05.2011 C-391/09 (RunevičVardyn and Wardyn) ECLI:EU:C:2011:291 paras 86 f. 159 Opinion of AG Poiares Maduro, 08.10.2009, CJEU C-213/07 (Michaniki) ECLI:EU:C:2008:544 para 33.

618  Mirosław Wyrzykowski and Michał Ziółkowski a freestanding construction of Polish constitutional law, without expressly determining how its content is related to the Treaty notion of national identity.160 Nevertheless, analysing references to ‘constitutional identity’ it might be argued that the TK’s concept of constitutional identity shall generally be seen as Euro-friendly and open for future interpretation. Firstly, as an element of a substantive norm addressed to the legislature, the obligation to protect constitutional identity is strictly referred to the integration process. According to the Tribunal, the substance of constitutional identity should be deduced from the competencies subject to the prohibition of delegation.161 The legal basis is thus Article 90 KRP, as the substance of the constitutional norm empowering delegation of competencies is a limitation on such delegation by a command to maintain constitutional identity. In the Tribunal’s view, ‘Article 90 of the Constitution and the bounds on delegation of competencies specified therein are the guarantees of maintaining the constitutional identity of the Republic.’162 In analysing the decisions of constitutional courts from other member states, the Tribunal, stated, ‘A common characteristic of these rulings is … to stress the importance of constitutional identity and systemic identity … respect for which excludes the possibility of any implied change in the national Constitution, specifically with respect to the rules for delegation of competencies.’163 Secondly, mentioning constitutional identity protection, the Tribunal underlined an openness of the Polish constitutional order for the integration process164 as well as a common axiological foundation for EU law and the KRP. The Tribunal rejected the traditional sovereignty concept and emphasised that there shall be no contradiction between the principle of sovereignty and the principle of conferral competence165 as long as Poland is recognised as a state under international law, with its own constitution and authorities with the minimum scope of competence.166 Thirdly, acting in accordance with the obligation to protect constitutional identity the TK is also bound by the other constitutional norms, in particular with the constitutional duty to interpret national, including constitutional law in accordance with the EU law. Therefore there is a narrow scope for a collision that could not be ruled in a judiciary dialogue. The fourth reason is that one of the rationes of the identity concept was to protect the national constitutional order against indirect constitutional changes. Having regard to the fact that the Tribunal concept was strictly connected to the principle of conferral of competences, and taking into account the lack of the eternity clause in the Constitution (Article 235), it may be argued that constitutional identity in a normative sense shall not be seen as an objection for the national parliament acting changing the Constitution. Therefore, in the light of the Lisbon Treaty judgment analysed in connection with EAW judgment, changing the Constitution in accordance with its Chapter XII might be one of the possible ways to remove the constitutional identity collision with the EU law. In other words, the obligation to protect constitutional identity is relative in time due to the current wording of the constitutional provisions. The fifth reason is that the procedure of controlling normative acts in conformity with the constitutional identity requirement may be initiated in the selected (narrow) context only. Generally, there are no legal grounds for controlling the EU secondary law with the constitutional 160 Śledzińska-Simon/ Ziółkowski, ‘Constitutional Identity’ (n 146) 254 f. 161 ‘Competencies subject to the prohibition on delegation determine constitutional identity, and thus reflect the values the Constitution is founded on’ (KT K 32/09 (n 64), point III.2.1 of the justification). 162 Ibid, point III.2.1 of the justification. 163 Ibid, point III.3.1 of the justification. 164 Ibid, point III.2.2 of the justification. 165 Ibid, point III.2.1 of the justification. 166 Ibid.

Poland  619 identity requirement. According to the Tribunal interpretation, the EU directives or regulations may be controlled in a constitutional complaint proceedings only.167 However, it does not imply the admissibility of rising similar allegations under the constitutional complaint proceedings. According to Article 79 KRP, individuals right to appeal to the TK depends on human rights violation. Since a complainant must specify the provisions from Chapters I or II KRP, the complaint referring to the constitutional identity protection shall be also dismissed. Therefore the Tribunal may truly hear the constitutional identity violation case in two situations only: applications against the EU treaties or national statutes mentioned in Article 90 KRP.

V.  The Resulting Relationship between EU Law and National Law After a long judicial dialogue between the TK and the CJEU as well as the harmonious development of the Polish constitutional law and European constitutionalism, the relationship between EU law and national law has been radically changing. From 2016–19, EU institutions formally accused Poland of a serious breach of the rule of law principle. Following the widely discussed168 European Commission’s Communication from 2014,169 which aimed at a ‘pre-Article 7 procedure’170 and at filling the gap between the infringement procedure applied in cases of breach of EU law and the political procedure under Article 7 TEU,171 on 13 January 2016, for the first time ever, the Commission announced that it would carry out a preliminary assessment of the situation of the TK under the Rule of Law Framework.172 The Polish Government referred to the Council Legal Service’s arguments173 and claimed that time that the Commission had acted ultra vires and unlawfully. Those arguments had been strongly criticised earlier by legal scholars.174 167 TK SK 45/09 (n 66). 168 Eg Christophe Hillion, ‘Overseeing the Rule of Law in the EU Legal Mandate and Means’, in C Closa and D Kochenov (eds), Reinforcing rule of law oversight in the European Union (Cambridge, Cambridge University Press, 2016) 59, 77; Dimitry Kochenov and Laurent Pech, ‘Monitoring and Enforcement of the Rule of Law in the EU: Rhetoric and Reality’, (2015) 11 European Constitutional Law Review 512; Armin von Bogdandy, Carlino Antpöhler and Michael Ioannidis, ‘Protecting EU values Reverse Solange and the Rule of Law Framework’, in A Jakab and D Kochenov (eds), The Enforcement of EU Law and Values Ensuring Member States’ Compliance (Oxford, OUP, 2017) 211, 228; Dimitry Kochenov and Laurent Pech, ‘From bad to worse? On the Commission and the Council’s rule of law initiatives’ Verfassungsblog, 2015/1/20; Maciej Taborowski and Paweł Marcisz, ‘Nowe ramy Unii Europejskiej na rzecz umocnienia praworządności. Krytyczna analiza analizy krytycznej (artykuł polemiczny)’ [‘A new framework for the European Union to strengthen the rule of law. Critical analysis of critical analysis (polemical article)’], (2017) 12 Państwo i Prawo 100; Jan‐Werner Müller, ‘Should the EU Protect Democracy and the Rule of Law inside Member States?’, (2015) 21 European Law Journal 141, 146. 169 Communication of the Commission to the European Parliament and the Council. A New EU Framework To Strengthen The Rule Of Law, Brussels, 11.3.2014, COM(2014) 158 final. 170 Vivian Reding, A new Rule of Law Initiative, Speech 14/202, Strasbourg, 11 March 2014. 171 According to the Commission’s opinion, ‘[t]he framework seeks to resolve future threats to the rule of law in Member States before the conditions for activating the mechanisms foreseen in Article 7 TEU would be met. It is therefore meant to fill a gap. It is not an alternative to but rather precedes and complements Article 7 TEU mechanisms. It is also without prejudice to the Commission’s powers to address specific situations falling within the scope of EU law by means of infringement procedures …’ (n 169, para 1). 172 Read out by First Vice-President Timmermans of the College Meeting of 13 January 2016, Brussels, 13 January 2016 (http://europa.eu/rapid/press-release_SPEECH-16-71_en.htm?locale=en%3E). 173 The Council Legal Service arguments were later used by the Commission’s opponents – see eg Thomas Würtenberger and Jan W Tkaczyński, ‘Nowe ramy Unii Europejskiej na rzecz umocnienia praworządności (analiza krytyczna)’ [‘A new framework for the European Union to strengthen the rule of law (critical analysis)’] (2017) 9 Państwo i Prawo 16. 174 Armin von Bogdandy, Carlino Antpöhler and Michael Ioannidis, ‘A New Page in Protecting European Constitutional Values: How to best use the new EU Rule of Law Framework vis-a-vis Poland’, Verassungsblog, 2016/1/24; Leonard Besselink, ‘The Bite, the Bark and the Howl: Article 7 TEU and the Rule of Law Initiatives’ in A Jakab and D Kochenov (eds), The Enforcement of EU Law and Values Ensuring Member States’ Compliance (Oxford, OUP, 2017) 128;

620  Mirosław Wyrzykowski and Michał Ziółkowski In its first recommendation of 27.07.2016,175 the Commission explained the circumstances of the Law Framework in which it adopted an Opinion concerning the rule of law in Poland. The Commission found that there was a systemic threat to the rule of law in Poland and recommended that the Polish authorities take appropriate action to address this threat as a matter of urgency. In particular, the Commission recommended that the Polish authorities: • implement fully the judgments of the TK of 3 and 9 December 2015, requiring that the three judges that were lawfully nominated in October 2015 can take up their judicial functions in the TK, and that the three judges nominated afterwards and occupying posts without a valid legal basis do not take up their judicial functions; • publish and implement fully the judgments of the TK of 9 March 2016 and ensure that publication of future judgments is automatic and does not depend on any decision of the executive or legislative powers; • ensure that any reform of the law on the TK respects its judgments, including those of 3 and 9 December 2015 and of 9 March 2016, and takes the opinion of the Venice Commission fully into account; • ensure that the effectiveness of the TK as a guarantor of the Constitution is not undermined by requirements inconsistent with the rule of law; • ensure that the TK can review the compatibility of the new law adopted on 22 July 2016 on the TK before its entry into force and publish and implement fully the judgment of the Tribunal in that respect; • refrain from actions and public statements which could undermine the legitimacy and efficiency of the Constitutional Tribunal. In its reply of 27 October 2016, the Polish Government176 failed to acknowledge the majority of concerns expressed by the Commission and by the Venice Commission.177 The reply mainly contested that the TK was prevented from exercising an effective review by referring to the fact that the Tribunal had been able to issue rulings during the so-called constitutional crisis.178 In the statement published on the website of the Polish Ministry of Foreign Affairs on 27 October 2016, the Ministry denounced the ‘interferences into Poland’s internal affairs’ as violations of the principles of ‘objectivism, or respect for sovereignty, subsidiarity, and national identity.’179 Kochenov/Pech, ‘Monitoring’ (n 168); Roberto Baratta, ‘Rule of Law “Dialogues” Within the EU: A Legal Assessment’, (2016) 8 Hague Journal on the Rule of Law 357. 175 Recommendation regarding the rule of law in Poland No 2016/1374, [2016] OJ L 217/53. 176 See Poland – Position regarding draft Opinion of the Venice Commission on the Act on the Constitutional Tribunal of 22 July 2016, CDL-REF(2016)060-e. According to the Polish Government ‘the draft Opinion of the Venice Commission … repeats the errors of the Opinion of 11 March 2016, reinforces them and, in so doing, tries to exert pressure on the Government of the Republic of Poland so as to reaffirm the past and currently planned illegal actions taken by the Constitutional Tribunal. The Opinion … reveals a clear political commitment of the experts who drafted it on the side of the opposition’. 177 From 2016–17 the Venice Commission adopted four opinions regarded the current rule of law crisis in Poland. The first referred to the TK statutory changes as well as practices of Polish constitutional authorities, including the rejection of promulgation of judgments. The second opinion was focused on statutory changes in the Police and limitations of right to privacy protection. The third opinion’s subject was narrowed to the statute on the Public Prosecutor’s Office. The fourth opinion regarded the rule of law assessment of the statutes on judiciary branch (the SCPL, NCJ and ordinary courts). See Venice Commission, Opinion no 833/2015, cdl-ad(2016)001; Opinion no 839/ 2016, cdl-ad(2016)012; Opinion 892 / 2017, cdl-ad(2017)028; Opinion 892 / 2017, cdl-ad(2017)028. 178 Statement on the Polish government’s response to Commission Recommendation of 27.07.2016, www.msz.gov. pl/en/p/msz_en/news/mfa_statement_on_the_polish_government_s_response_to_commission_recommendation_ of_27_07_2016. 179 Ibid.

Poland  621 In its second recommendation of 21 December 2016,180 the Commission mainly focused on new Polish constitutional review legislation as well as the President’s competence to appoint a judge acting as a President of the Constitutional Court.181 In an official letter of 20 February 2017, the Polish Government replied to this, disagreeing with the assessments and refusing to address the concerns identified by the Commission. The reply emphasised that the appointment of the new President of the TK as well as entry into force of the three new laws governing its functioning created the conditions for the functioning of the Tribunal after a period of paralysis caused by politicians of the opposition in which the former President of the Tribunal was also engaged. In a third recommendation of 26 July 2017,182 the Commission recalled the previous recommendations but mainly focused on new Polish legislation on NCJ, SCPL and ordinary courts. Taking into account the legal and political situation of the TK as well as the response of the Polish Government to the previous recommendations, the Commission pointed out that the independence and legitimacy of the Constitutional Tribunal are seriously undermined and, consequently, the constitutionality of Polish laws can no longer be effectively guaranteed. This situation is particularly worrying for the respect of the rule of law …. Moreover, the adverse impact on the rule of law of the lack of an independent and legitimate constitutional review in Poland is now seriously aggravated by the fact that the constitutionality of the new laws relating to the Polish judicial system.183

In his reply of 26 July 2017, the Polish Minister of Foreign Affairs noted ‘with concern that the European Commission once again talks about a dialogue, but instead, it uses a language of ultimatums, arbitrarily setting the agenda of talks.’184 Officially the Polish Government communicated its response to the European Commission on 29 August 2017.185 The fourth Commission recommendation of 20 December 2017 was widely based on the Venice Commission findings.186 At the same time, the Commission adopted a reasoned proposal in accordance with Article 7(1) TEU (regarding the rule of law in Poland no 2017/0360).187 In its reply of 20 December 2017, the Polish Government questioned again the legitimacy of the Commission’s acts, underlined that recommendations were based on misunderstandings and alleged that Commission’s proceedings were entered into with a lack of goodwill. Therefore, the Government decided to defend its claims at the CJEU.188 In 2018 the European Commission brought an infringement case against Poland to the Court of Justice of the EU, accusing Poland of violation of the equal treatment of judges of ordinary

180 Recommendation regarding the rule of law in Poland No 2017/146 complementary to Commission Recommendation (EU) 2016/1374, [2016] OJ L 22/65. 181 The Commission was strongly criticised for giving more time to the Polish authorities to implement both the ‘old set’ and the ‘new set’ of rule of law recommendations. (Laurent Pech and Kim Lane Scheppele, ‘Poland and the European Commission, Part I: A Dialogue of the Deaf?’, Verfassungsblog 2017/1/03). 182 Recommendation regarding the rule of law in Poland No 2017/1520 complementary to Commission Recommendations (EU) 2016/1374 and (EU) 2017/146, OJ L 228/19. 183 Ibid, paras 10–11. 184 Statement following the EU Comission’s Recommendation of 26 July 2017, www.msz.gov.pl/en/news/mfa_statement_ following_the_european_commission_s_recommendation_of_26_july_2017_regarding_the_rule_of_law_in_poland_. 185 MFA statement regarding Polish response to European Commission Recommendation of 26 July 2017, https:// brusselseu.mfa.gov.pl/en/news/mfa_statement_regarding_polish_response_to_european_commission_recommendation_ of_26_july_2017?printMode=true. 186 Recommendation regarding the rule of law in Poland 2018/103 complementary to Commission Recommendations (EU) 2016/1374, (EU) 2017/146 and (EU) 2017/1520, OJ L 17/50, para 47. 187 Proposal for a Council Decision on the determination of a clear risk of a serious breach by the Republic of Poland of the rule of law, COM/2017/0835 final – 2017/0360 (NLE). 188 MFA statement on the EU Commission’s decision to launch the process against Poland laid out in Art 7 TEU, www. msz.gov.pl/en/p/msz_en/news/mfa_statement_on_the_european_commission_s_decision_to_launch_the_disciplinary_ process_against_poland_laid_out_in_article_7_of_the_teu.

622  Mirosław Wyrzykowski and Michał Ziółkowski courts,189 as well as violation of the independence of the SCPL.190 In 2019 the Court of Justice ruled twice in June and in November that Poland had violated the EU law. Moreover, the Court of Justice replied for the first preliminary questions of the SCPL on 19 November and opened a new chapter in the Polish rule of law saga.191 The three other preliminary references of the SCPL are still waiting for the Court of Justice response. Moreover, the preliminary references of courts of general jurisdictions are also pending before the Court of Justice.192

References B Ackerman, The Rise of World Constitutionalism (Yale Occasional Papers No 4, 1996). B Ackerman, Revolutionary Constitutions Charismatic Leadership and the Rule of Law (Cambridge MA, Harvard University Press, 2019). A Adamczyk et al, Expert Report on Issues Relating to the Constitutional Tribunal, 2016 (www. marekkuchcinski.pl/wp-content/uploads/2016/09/EN-Raport-Zespo%C5%82u-Ekspert%C3%B3w-dospraw-problematyki-Trybuna%C5%82u-Konstytucyjnego-wersja-angielska-1.pdf). B Banaszak, Zasada nadrzędności Konstytucji w polskim porządku prawnym [The Principle of Supremacy of the Constitution in the Polish Legal Order] (Wrocław, E-Wydawnictwo, 2015). B Banaszak and M Bernadczyk (eds), Aktywizm sędziowski we współczesnym państwie demokratycznym [Judicial activism in a modern democratic state] (Warszawa, Wydawnictwo Sejmowe, 2012). R Baratta, ‘Rule of Law “Dialogues” Within the EU: A Legal Assessment’, (2016) 8 Hague Journal on the Rule of Law 357. M Bernat and M Ziółkowski, ‘Statutory Anti-Constitutionalism’, (2019) 28 Washington Law Review 487. L Besselink, ‘The Bite, the Bark and the Howl: Article 7 TEU and the Rule of Law Initiatives’ in A Jakab and D Kochenov (eds), The Enforcement of EU Law and Values Ensuring Member States’ Compliance (Oxford, OUP, 2017) 128. S Biernat and M Kawczyńska, ‘Why the Polish Supreme Court’s Reference on Judicial Independence to the CJEU is Admissible after all’, Verfassungsblog 2018/8/23, https://verfassungsblog.de/why-the-polishsupreme-courts-reference-on-judicial-independence-to-the-cjeu-is-admissible-after-all/. A Bodnar and A Śledzińska-Simon, ‘Between Recognition and Homophobia: Same-Sex Couples in Eastern Europe’, in D Gallo et al (eds), Same-sex Couples before National, Supranational and International Jurisdictions (Heidelberg, Springer; 2014) 211. M Brzezinski, ‘The Emergence of Judicial Review in Eastern Europe: The Case of Poland’, (1993) 41(2) American Journal of Comparative Law, 173. M Brzezinski and L Garlicki, ‘Judicial Review in Post-Communist Poland: The Emergence of a Rechtsstaat?’, (1995) 31 Stanford Journal of International Law 13. A Chmielarz-Grochal and Jarosław Sułkowski, ‘Appointment of Judges to the Constitutional Tribunal in 2015 as the Trigger Point for a Deep Constitutional Crisis in Poland’, (2018) 2 Przegląd Konstytucyjny, 93. R Chruściak, ‘The Constitution of the Republic of Poland of 2 April 1997 – The Course of Parliamentary Work from 1995–1997’, (1997) 113–16 Droit Polonais Contemporain 163. J Ciapała, ‘The Position of the Sejm in the Area of Foreign Policy and International Relation’, (1999) (1st special edn) The Sejm Review 83.

189 CJEU 05.11.2019 C-192/18 (Commission v Poland [Independence of ordinary courts]) ECLI:EU:C:2019:924. 190 CJEU 24.06.2019 C-619/18 (Commission v Poland (Independence of the Supreme Court)) ECLI:EU:C:2019:531. 191 CJEU C-624/18 (n 96). 192 04.09.2018 by the Regional Court in Warsaw; 31.08.2018 by the Regional Court in Łódź and a similar submission at the end of September 2018 by the Regional Court in Gorzów Wielkopolski.

Poland  623 E Cloots, National Identity in EU Law (Oxford, OUP, 2015). A Czarnota, ‘Guardians of the Constitution or legislative Bodies?’, (2001) 28 East Central Europe 147. Z Czeszejko-Sochacki, ‘The Origins of Constitutional Review in Poland’, (1996) Saint Louis-Warsaw Transatlantic Law Journal 24. H Dębska, ‘Legal Doxa as a Form of Neutralization of Values in the Law. The Case of Polish Constitutional Tribunal’, in K Pałecki (ed), Neutralization of Values in Law (Warsaw, Wolters Kluwer, 2013) 303. H Dębska, Władza, symbol, prawo. Społeczne tworzenie Trybunału Konstytucyjnego [Power, symbol, law. Social creation of the Constitutional Tribunal] (Warszawa, Wydawnictwo Sejmowe, 2015). H Dębska and T Warczok, ‘Sacred Law and Profane Politics. The Symbolic Construction of Constitutional Tribunal’, (2014) 4 Polish Sociological Review 188. S Dudzik and N Półtorak, ‘The Court of the Last Word: Competences of the Polish Constitutional Tribunal in the Review of European Union Law’, (2012) 15 Yearbook of Polish European Studies 225. K Działocha, in L Garlicki and M Zubik (eds) Konstytucja Rzeczypospolitej Polskiej. Komentarz [Constitution of the Republic of Poland. Commentary] vol 1, ed (Warszawa, Wolters Kluwer, 2016). M Florczak-Wątor, in L Bosek and M Safjan (eds) Konstytucja Rzeczypospolitej Polskiej. Komentarz do art. 1-86 [Constitution of the Republic of Poland. Commentary to Arts 1-86] vol 1, (Warsaw, CH Beck, 2016). K Gebert, ‘Poland since 1989: Muddling through, Wall to Wall’, in S Ramet and C Hassenstab (eds), Central and Southeast European Politics since 1989 (Cambridge, CUP, 2019) 137. S Gebethner, ‘Political Reform in the Process of Round Table Negotiations’, in G Sanford (ed), Democratization in Poland. 1988–90 (London, Palgrave Macmillan, 1992) 50. S Gebethner, ‘The 1997 Referendum on the Constitution in Poland. The Controversies and the Compromise’, (1997) 113–16 Droit Polonais Contemporain 135. C Hillion, ‘Overseeing the Rule of Law in the EU Legal Mandate and Means’, in C Closa and D Kochenov (eds), Reinforcing rule of law oversight in the European Union (Cambridge, CUP, 2016) 59. K Jasiewicz and T Żukowski, ‘The Elections of 1984–89 as a Factor in the Transformation of the Social Order in Poland’, in G Sanford (ed), Democratization in Poland. 1988–90 (London, Palgrave Macmillan, 1992) 98. A Kastelik-Smaza, ‘Skutki stwierdzenia przez Trybunał Konstytucyjny niezgodności z Konstytucją unijnego prawa pochodnego z punktu widzenia skarżącego’ [‘The Effects of TK’s Decision on Unconstitutionality of Secondary EU Law in the Eyes of the Applicant’], in S Dudzik and N Półtorak (eds), Prawo Unii Europejskiej a prawo konstytucyjne państw członkowskich [The Law of the European Union and the Constitutional Law of Member States] (Warszawa, Wolters Kluwer, 2013) 244. D Kochenov and L Pech, ‘Monitoring and Enforcement of the Rule of Law in the EU: Rhetoric and Reality’, (2015) 11 European Constitutional Law Review 512. D Kochenov and L Pech, ‘From bad to worse? On the Commission and the Council’s rule of law initiatives’ Verfassungsblog, 2015/1/20, https://verfassungsblog.de/bad-worse-commission-councils-rulelaw-initiatives/. T Koncewicz, ‘The Capture of the Polish Constitutional Tribunal and Beyond: Of Institution(s), Fidelities and the Rule of Law in Flux’, (2018) 43 (2) Review of Central and East European Law, 116. K Kowalik-Bańczyk, ‘Sending Smoke Signals to Luxembourg – the Polish Constitutional Tribunal in a Dialogue with the ECJ’, in M Claes et al (eds), Constitutional Conversations in Europe: Actors, Topics and Procedure (Cambridge, Intersentia, 2012) 267. M Krajewski, ‘The AG Opinion in the Celmer Case: Why the Test for the Appearance of Independence is Needed’, Verfassungsblog, 2018/7/05, https://verfassungsblog.de/the-ag-opinion-in-the-celmer-casewhy-the-test-for-the-appearance-of-independence-is-needed/. M Krajewski and M Ziółkowski, ‘The Power of “Appearances”, Verfassungsblog, 2019/11/26, https:// verfassungsblog.de/the-power-of-appearances/. M Krawczyńska, ‘Kontrola zgodności z Konstytucją aktów unijnego prawa pochodnego w postępowaniach przed Trybunałem Konstytucyjnym’ [‘Constitutional Review of Secondary EU Legislation before the TK’], in S Dudzik, N Półtorak (eds), Prawo Unii Europejskiej a prawo konstytucyjne państw członkowskich [The Law of the European Union and the Constitutional Law of Member States] (Warszawa, Wolters Kluwer, 2013) 220.

624  Mirosław Wyrzykowski and Michał Ziółkowski L Garlicki, ‘Das Polnische Gesetz über den Verfassungsgerichtshof vom 29. April 1985’, (1987) 26 Der Staat 279. L Garlicki, in Norske videnskaps-akademi (ed), The Role of the Constitution in a Changing Society. Joint Polish-Norwegian Conference. Oslo, 14–16 May 1991 (Norwegian Academy of Science and Letters, Oslo, 1991) 18. L Garlicki, ‘Control Powers of the Senate of the Republic of Poland Comments to the Article by Prof. Paweł Sarnecki’, (2002) (2nd special edn) The Sejm Review 87. L Garlicki, ‘Constitutional Court of Poland: 1986–2009’, in P Pasquino/F Billi (eds), The Political Origins of Constitutional Courts (Rome, Fondazione Adriano Olivetti, 2009) 13. L Garlicki, ‘Die Ausschaltung des Verfassungsgerichtshofes in Polen?’, in A Szmyt and B Banaszak (eds), Transformation of Law Systems in Central, Eastern and Southeastern Europe in 1989–2015 (Gdańsk, Gdansk University Press, 2016) 63. L Garlicki, ‘Constitutional Court and Politics. The Polish Crisis’, in C Landfried (ed), Judicial Power How Constitutional Courts Affect Political Transformations (Cambridge, CUP, 2019) 144. L Garlicki and Z Garlicka, ‘Constitution Making, Peace Building, and National Reconciliation. The Experience of Poland’, in L Miller (ed), Framing the State in Times of Transition: Case Studies in Constitution Making (Washington DC, US Institute of Peace Press, 2000) 394. A Kustra, ‘The Polish Constitutional Court and Political “Refolution” after 1989: Between the Continuity and Discontinuity of the Constitutional Narrative’, (2016) 6 Wroclaw Review of Law, Administration & Economics 86. E Łętowska and A Wiewiórowska-Domagalska, ‘A “Good” Change in the Polish Constitutional Tribunal?’, (2016) 1 Osteuropa Recht, 79. E Łętowska and J Woleński, ‘Instytucjonalizacja związków partnerskich a Konstytucja RP z 1997 r’ [‘Institutionalization of civil unions and the KRP of 1997’], (2013) 6 Państwo i Prawo 15. M Loughlin, ‘The Constitutional Imagination’, (2015) 78 The Modern Law Review 1. M Masternak-Kubiak and J Trzciński, ‘The System of Government in the Constitution of the Republic of Poland of April 2, 1997 – an Analysis of the Competence of the Sejm’, (1999) (1st special edn) The Sejm Review 39. M Mazurkiewicz, ‘The Draft of the Constitution of the Republic of Poland (1996) – Its Political Conditions and Expectations’, in M Wyrzykowski (ed), Constitution-making Process (Warsaw, Instytut Spraw Publicznych, 1998) 115. E Morawska, Klauzula demokratycznego państwa prawnego w Konstytucji RP na tle orzecznictwa Trybunału Konstytucyjnego [The clause of the democratic state ruled by law in the KRP in the light of the jurisprudence of the TK] (Toruń, Dom Organizatora TNOiK, 2003). JW Müller, ‘Should the EU Protect Democracy and the Rule of Law inside Member States?’, (2015) 21 European Law Journal 141. W Osiatyński, ‘The Constitution-Making Process in Poland’, (1991) 13 (2) Law & Policy, 129. W Osiatyński, ‘Rights in New Constitutions of East Central Europe’, (1994) 26 Columbia Human Rights Law Review 119. W Osiatyński, ‘Poland’s Constitutional Ordeal’, (1994) 3 (2) East European Constitutional Review, 29. W Osiatyński, ‘A Brief History of the Constitution’, (1997) 6 (2/3) East European Constitutional Review, 67. J Pawliczak, Zarejestrowany związek partnerski a małżeństwo [Registered partnership and marriage] (Warszawa, Wolters Kluwer, 2014). L Pech and K Scheppele, ‘Poland and the European Commission, Part I: A Dialogue of the Deaf?’, Verfassungsblog 2017/1/03, https://verfassungsblog.de/poland-and-the-european-commission-part-i-adialogue-of-the-deaf/. A Piasecki, ‘Direct Democracy in Poland after 1989’, (2007) (3rd special edn) The Sejm Review 145. M Pietrzak, ‘Demokracja reprezentacyjna i bezpośrednia w Konstytucji Rrzeczypospolitej Polskiej’ [‘Representative and direct democracy in the KRP’], in M Staszewski (ed), Referendum Konstytucyjne w Polsce [The Constitutional Referndum in Poland] (Warszawa, Instytut Studiów Politycznych PAN, 1997) 26.

Poland  625 P Radziewicz and P Tuleja (eds), Konstytucyjny spór o granice zmian organizacji i zasad działania trybunału konstytucyjnego, czerwiec 2015 – marzec 2016 [Constitutional dispute over the boundaries of changes in the organization and rules of operation of the TK, June 2015–March 2016] (Warszawa, Wolters Kluwer, 2017). W Sadurski, Rights Before Courts. A Study of Constitutional Courts in Postcommunist States of Central and Eastern Europe (Dordrecht, Springer, 2005). W Sadurski, How Democracy Dies (in Poland): A Case Study of Anti-Constitutional Populist Backsliding, Sydney Law School Research Paper No 18/01 (2018). W Sadurski, ‘Polish Constitutional Tribunal Under PiS: From an Activist Court, to a Paralysed Tribunal, to a Governmental Enabler’, (2018) 11 Hague Journal on the Rule Law 63. W Sadurski, Poland’s Constitutional Breakdown (Oxford, OUP, 2019). A Sajó, ‘Constitution without the constitutional moment: a view from the new member states’, (2005) 3 International Journal of Constitutional Law 243. P Sarnecki, ‘Central Ideas of the Constitution of the Republic of Poland of 2nd April 1997’, (1997) 113–16 Droit Polonais Contemporain 17. P Sarnecki, ‘Control Power of the Senate of the Republic of Poland’, (2002) (2nd special edn) The Sejm Review 69. A Śledzińska-Simon, ‘Constitutional identity in 3D: a model of individual, relational and collective self and its application in Poland’, (2015) 13 International Journal of Constitutional Law 124. A Śledzińska-Simon, ‘Midnight Judges: Poland’s Constitutional Tribunal Caught Between Political Fronts’, Verfassungsblog 2015/11/23, https://verfassungsblog.de/?p=11945. A Śledzińska-Simon and M Ziółkowski, ‘Constitutional Identity in Poland: Is the Emperor Putting On the Old Clothes of Sovereignty?’, in C Calliess and G van der Schyff (eds), Constitutional Identity in a Europe of Multilevel Constitutionalism (Cambridge, CUP, 2019) 248. W Sokolewicz, ‘The Legal-Constitutional Bases of Democratisation in Poland: Systemic and Constitutional Change’, in G Sanford (ed), Democratization in Poland. 1988–90 (London, Palgrave Macmillan, 1992) 97. M Staszewski (ed), Referendum Konstytucyjne w Polsce [The Constitutional Referndum in Poland] (Warszawa, Instytut Studiów Politycznych PAN, 1997). M Taborowski and P Marcisz, ‘Nowe ramy Unii Europejskiej na rzecz umocnienia praworządności. Krytyczna analiza analizy krytycznej (artykuł polemiczny)’ [‘A new framework for the European Union to strengthen the rule of law. Critical analysis of critical analysis (polemical article)’], (2017) 12 Państwo i Prawo 100. M Taborowski and P Marcisz, ‘The first judgment of the ECJ regarding a breach of the rule of law in Poland?’, Verfassungsblog 2019/5/29, https://verfassungsblog.de/the-first-judgment-of-the-ecj-regardinga-breach-of-the-rule-of-law-in-poland/. A von Bogdandy, Tyrannei der Werte? Probleme und Wege europäischen Schutzes nationaler Rechtsstaatlichkeit, MPIL Research Paper No 2019-04. A von Bogdandy, C Antpöhler and M Ioannidis, ‘A New Page in Protecting European Constitutional Values: How to best use the new EU Rule of Law Framework vis-a-vis Poland’, Verassungsblog, 2016/1/24, https:// verfassungsblog.de/a-new-page-in-protecting-european-constitutional-values-how-to-best-use-thenew-eu-rule-of-law-framework-vis-a-vis-poland/. A von Bogdandy, C Antpöhler and M Ioannidis, ‘Protecting EU values Reverse Solange and the Rule of Law Framework’, in A Jakab and D Kochenov (eds), The Enforcement of EU Law and Values Ensuring Member States’ Compliance (Oxford, OUP, 2017) 211. M Wiącek, ‘Glosa do postanowienia TK z dnia 17 grudnia 2009 r. U 6/08’ [‘Commentary on Resolution of the TK of 17 December 2009 U 6/08’], (2010) 6 Państwo i Prawo, 123. K Wójtowicz, Constitutional courts and European Union law, (Wrocław, E-Wydawnictwo, 2014). T Würtenberger and J W. Tkaczyński, ‘Nowe ramy Unii Europejskiej na rzecz umocnienia praworządności (analiza krytyczna)’ [‘A new framework for the European Union to strengthen the rule of law (critical analysis)’] (2017) 9 Państwo i Prawo 16. M Wyrzykowski, ‘Constitutional Changes in Poland 1989–1991’, (1992) 17 Bulletin of the Australian Society of Legal Philosophy 25.

626  Mirosław Wyrzykowski and Michał Ziółkowski M Wyrzykowski, ‘Reflections on Some Recent Constitutional Developments in Eastern Europe’, (1992) 2 Tilburg Foreign Law Review, 164. M Wyrzykowski, ‘Introductory Note to the 1997 Constitution of the Republic of Poland’, (1997) Saint LouisWarsaw Transatlantic Law Journal 3. M Wyrzykowski, ‘Publiczne a prywatne w wykładni konstytucyjnej na przykładzie art. 18 Konstytucji RP’ [‘Public versus private in the constitutional interpretation on the example of Art. 18 of the KRP’], in T. Giaro (ed), Interes publiczny a interes prywatny w prawie [Public and private interest in law] (Warszawa, SAWPA-WU, 2009) 215. M Wyrzykowski, ‘Bypassing the Constitution or Changing the Constitutional Order outside the Constitution’, in A Szmyt and B Banaszak (eds), Transformation of Law Systems in Central, Eastern and Southeastern Europe in 1989–2015, (Gdańsk, Gdansk University Press, 2016) 159. M Wyrzykowski, ‘Antigone in Warsaw’, in M Zubik (ed), Human Rights in Contemporary World. Essays in Honour of Professor Leszek Garlicki (Warszawa, Wydawnictwo Sejmowe, 2017) 372. M Wyrzykowski, ‘The Vanishing Constitution’, (2018) European Yearbook on Human Rights 3. J Zajadło, ‘Constitution-hostile Interpretation’, (2018) 2 Przegląd Konstytucyjny 8. M Ziółkowski, ‘Constitutional Moment and the Polish Constitutional Crisis 2015–2018 (a few Critical Remarks)’, (2018) 4 Przegląd Konstytucyjny 76. M Ziółkowski, ‘Zmiany składu orzekającego Trybunału Konstytucyjnego’ [‘Changes in the adjudication panel of the TK’], in Katarzyna Łakomiec (ed), Funkcjonowanie Trybunału Konstytucyjnego w latach 2014–2017. Raport Zespołu Ekspertów Prawnych Fundacji im. Stefana Batorego [Functioning of the Constitutional Tribunal in 2014–2017. Report of the Team of Legal Experts at the Stefan Batory Foundation] (Warszawa, 2018) 157. M Ziółkowski, Przywracanie praworządności w TK po kryzysie konstytucyjnym: wybór i powołanie Prezesa TK (analiza sporządzona dla Archiwum Osiatyńskiego) [Restoring the rule of law in the TK after the constitutional crisis: election and appointment of the President of the TK (analysis prepared for the Osiatyński Archives)] (Warszawa, Archiwum Osiatyńskiego, 2019).

22 Portugal ANA MARIA GUERRA MARTINS

I.  Main Characteristics of the National Constitutional System A.  The Constitution 1.  The Revolution and the Genesis of the Constitution of 1976 The current Constitution of the Portuguese Republic (CPR) of 19761 was mainly conceived in the first two years after the Revolution (25 April 1974). Its elaboration was largely influenced by the Movimento das Forças Armadas (Armed Forces Movement) that had freed Portugal from the dictatorship, and by the power constellation of the political parties seated in the Constituent Assembly (Assembleia Constituinte), which had meanwhile emerged. According to the Movimento das Forças Armadas’ programme, a Constituent Assembly should be elected within a year after the Revolution by direct, secret and universal suffrage with the aim of drafting a Constitution. Without clearly imposing any particular solution, the programme foresaw, however, the existence of a directly elected President beside Parliament. Elected on 25 April 1975, the Constituent Assembly was exposed to huge political pressure.2 This came especially from the Communist Party, which tried to attain in the streets the power it had failed to win in the ballots. In spite of the tense political climate, the Assembly was able to draw up a Constitution and to approve it on 2 April 1976.3

2.  Seven Constitutional Revisions Once the excitement of the Revolution was over, it was necessary to rethink the constitutional system. As a consequence, the CPR has been amended seven times.4

1 An English version of the Constitution can be consulted at www.parlamento.pt. 2 See, above all Jorge Miranda, Da Revolução à Constituição (Cascais, Principia, 2015) 223 ff. 3 On the drafting of the Constitution see Jorge Miranda, Manual de Direito Constitucional, vol I.2, 10th edn (Coimbra, Coimbra Editora, 2014) 101 ff; Jorge Miranda, ‘Decisões políticas: aprovação, abstenção e rejeição no momento constituinte de 1976’, (2000) 46 Studia Juridica. Universidade de Coimbra. Boletim da Faculdade de Direito 177; José Joaquim Gomes Canotilho and Vital Moreira, Fundamentos da Constituição (Coimbra, Coimbra Editora, 1991) 9 ff; Francisco Lucas Pires, Teoria da Constituição de 1976. A transição dualista (Coimbra, Edition of the author, 1988) 125 ff; José Joaquim Gomes Canotilho, Direito Constitucional e Teoria da Constituição, 7th edn (Coimbra, Almedina, 2003) 200–06. 4 See Constitutional Laws No 1/1982, No 1/1989, No 1/1992, No 1/1997, No 1/2001, No 1/2004 and No 1/2005.

628  Ana Maria Guerra Martins The first constitutional revision occurred in 1982. It not only sought to reduce the ideological content of the CPR, allowing for greater flexibility in the economic system, but also redefined the structures for the exercise of political power.5 In contrast to the first revision, the second one (1989) was centred on economic issues, allowing for greater openness in the economic system and rejecting the principle of irreversibility of nationalisations.6 The subsequent revision in 1992 adapted the CPR to the principles of the Treaty of European Union (Maastricht Treaty), introducing, amongst other things, developments in the field of fundamental rights and strengthening the exclusive legislative powers of the Assembly of the Republic (AR).7 The constitutional revision of 1997 was characterised by the development of fundamental rights matters, increasing the powers of the AR, the reinforcement of Constitutional Court competences, the development of the powers of the autonomous regions, and strengthening the private initiative in the economy.8 In 2001 the CPR was once again amended. This was done in order to allow Portugal to ratify the Convention creating the International Criminal Court, by altering the rules on extradition.9 The sixth revision, approved in 2004, extended the political and administrative powers of the autonomous regions and amended the rules on international relations and the applicability of EU law within Portugal’s domestic legal system.10 Finally, the seventh revision was passed in 2005. By adding a new article to the CPR, it allowed the holding of referenda on the approval of treaties aimed at the construction and deepening of the European Union (EU).11 Since then, the CPR has not been amended, which can be considered a factor of its maturity.12

B.  Political System The influence of the post-revolutionary period on the 1976 Constitution is in many aspects remarkable. In its original form, it reflected the political and ideological options of the revolutionary period, enshrining a transition to socialism based on nationalisation of the principal means of production. Perhaps even more remarkable was that the Movimento das Forças Armadas maintained involvement in the exercise of political power through the Conselho da Revolução (Revolutionary Council). This Council was created during the Revolution when it represented, on the one hand, the Movimento das Forças Armadas and, on the other hand, the armed forces in general. It became a constitutional organ with extensive powers.

5 See Miranda, Manual I.2 (n 3) 216–24; Gomes Canotilho, Direito Constitucional (n 3) 208–09. 6 See Miranda, Manual I.2 (n 3) 224–32; Gomes Canotilho, Direito Constitucional (n 3) 209–13. 7 See Miranda, Manual I.2 (n 3) 232 ff; Gomes Canotilho, Direito Constitucional (n 3) 214. 8 See Miranda, Manual I.2 (n 3) 234–45; Gomes Canotilho, Direito Constitucional (n 3) 211–14; Alexandre Sousa Pinheiro and Mário João de Brito Fernandes, Comentário à IV Revisão Constitucional (Lisboa, AAFDL, 1999) 11–31. 9 See Miranda, Manual I.2 (n 3) 245 f. 10 For more details see ibid, 246–50. 11 Compare ibid 250 f. 12 One has to mention that some scholars have criticised the frequency of constitutional amendments in Portugal. See especially Jorge Miranda, ‘Acabar com o frenesim constitucional’, in AAFDL (ed), Nos 25 anos da Constituição da República Portuguesa de 1976 (Lisbon, AAFDL, 2001) 5; Ana M Guerra Martins, ‘La Constitution portugaise: Une Constitution inflationniste’, in G Amato et al (eds), The Constitutional Revision in Today’s Europe (London, Esperia Publications, 2002) 129 ff.

Portugal  629 Inspiration for the final choice of political system came also from two former Portuguese constitutional experiences and from foreign constitutions,13 such as Germany’s Weimar Constitution of 1919 and the French Constitution of 1958, both sharing political power between three organs – Parliament, Government and President – as occurred in the Portuguese Constitution.14 Notwithstanding these influences, as is stressed by Jorge Miranda, the political system in the 1976 Constitution was the outcome of the power constellation in Parliament and the political environment.15 The political system created by the 1976 Constitution is based on a separation of effective – positive and negative – powers between President, Parliament and Government, each one possessing identical democratic legitimacy. That system does not match well with any classical system – parliamentary or presidential. As a result, the majority of the Portuguese doctrine classifies it as a semi-presidential system.16 Throughout the application of the 1976 Constitution, the roles played by those three main actors had changed considerably. The powers of the President were reduced while the parliamentary dimension of the system was reinforced.17 However, in the last two years, due to the personal characteristics of the current President of the Republic – Marcelo Rebelo de Sousa – the office of President has been playing a more interventionist role.

C.  Type of State According to Article 1 CPR, ‘Portugal is a sovereign Republic, based on the dignity of the human person and the will of the people and committed to building a free, just and solidary society’. That means that the human being is the first priority of the state, ranking above economic and political organisation of the state.18 In accordance with Article 2 CPR, the Portuguese Republic is a democratic state based on the rule of law, the sovereignty of the people, plural democratic expression and political organisation, respect for and the guarantee of the effective implementation of the fundamental rights and

13 See Gomes Canotilho, Direito Constitucional (n 3) 597–98. 14 See especially Vitalino Canas, ‘Sistema semi-presidencial’, in Dicionário Jurídico da Administração Pública – 1º Suplemento (Coimbra, Almedina, 1998) 472; Jorge Reis Novais, Semipresidencialismo. Teoria do sistema de governos semipresidencial, vol I (Coimbra, Almedina, 2007) 232 ff; Ana M Guerra Martins, ‘The Portuguese Semi-Presidential System – About Law In the Books and Law In Action’, (2006) 2 European Constitutional Law Review, 81, 96 f. 15 Compare Miranda, Manual I.2 (n 3) 126. 16 See, eg Miranda, Manual I.2 (n 3) 194–99, 251–62; Carlos Blanco de Morais, ‘As metamorfoses do semipresidencialismo português’, in (1997) 22 Revista Jurídica 141; Canas, ‘Sistema’ (n 14) 490 ff; Marcelo Rebelo de Sousa, O Sistema de Governo Português, 4th edn (Lisbon, AAFDL, 1992) 14, 71; Pires, Teoria (n 3) 226 ff; Guerra Martins, ‘Semi-Presidential System’ (n 14) 96 ff. Gomes Canotilho argues against this classification, qualifying the Portuguese system as a parliamentarypresidential mix (Gomes Canotilho, Direito Constitucional (n 3) 598–605). Paulo Otero, O poder de substituição em Direito Administrativo, vol II (Lisboa, Lex, 1995) 792 and Direito Constitucional Português. Organização do Poder Político, vol II (Coimbra, Almedina, 2010) 486 ff qualifies the Portuguese political system as a parliamentary rationalised system. Adriano Moreira, ‘O regime: presidencialismo do Primeiro-Ministro’, in M Baptista Coelho (ed), Portugal – O Sistema Político e Constitucional 1974/87 (Lisbon, ICSUL, 1989) 31 supports the idea of a prime-ministerial presidential system. Cristina Queiroz, ‘A classificação das formas de governo. Em particular, o sistema de governo “semi-presidencial”’, in J Miranda (ed), Estudos em homenagem ao Prof. Doutor Armando Marques Guedes (Lisbon, Almedina, 2004) 321, argues that the main feature of the semi-presidential system is the domination of the parliamentary majority by the President. She concludes that the sole country in Europe that has such a system is France. 17 See especially Guerra Martins, ‘Semi-Presidential System’ (n 14). 18 See especially Paulo Otero, Direito Constitucional Português. Identidade Constitucional, vol I (Coimbra, Almedina, 2010) 34 ff.

630  Ana Maria Guerra Martins freedoms, and the separation and interdependence of powers, with a view to achieving economic, social and cultural democracy and deepening participatory democracy. The structure of the state is unitary with autonomous regions (Article 6(2) CPR) respecting the autonomous island system of self-government, and the principles of subsidiarity, the autonomy of local authorities and the democratic decentralisation of public administration (Article 6(1) CPR).19

D.  Separation and Interdependence of Powers As mentioned above, the political system is founded on separation of effective powers – both positive and negative – between President, Parliament and Government, which possess identical democratic legitimacy. The President of the Republic is elected by direct, secret and universal suffrage under a majority two-round electoral system, independently of political parties (Article 121 CPR). He or she exercises effective positive and negative political powers, such as appointing the Prime Minister, having the right to veto legislation, dismiss the Government, and dissolve the AR.20 The AR (Parliament) is elected by direct, secret and universal suffrage, under a proportional representation system (Article 149 CPR) and exercises legislative and political control powers (Article 149(1) CPR). The AR has the power to: • amend the Constitution (Article 161(a) CPR); • enact legislation on any subject other than those in the exclusive mandate of the government under the Constitution (Article 161(c) CPR); • delegate to the government the power to legislate (Article 161(d) CPR); • approve international conventions (Article 161(i) CPR); and • propose to the President of the Republic that referenda be held in cases where they are warranted by national interest (Article 161(j) CPR). The AR also has exclusive legislative powers in some matters, and partially exclusive legislative powers in others. For instance, the Assembly has exclusive legislative powers with regard to rights, freedoms, and guarantees, except where legislative power is delegated to the government.21 The government conducts the country’s general policy and is the supreme authority in public administration (Article 182 CPR). It also enjoys legislative powers in matters not within the exclusive mandate of the AR, and in matters that are within the exclusive power of the AR but which have been delegated to it.22

E.  Judicial System The courts are the entities that exercise sovereignty with the competence to administer justice in the name of the people (Article 202 CPR). Constitutionality and legality control is multi-shaped 19 On the unitary structure of the state see Jorge Miranda, Manual de Direito Constitucional, vol III, 6th edn (Coimbra, Coimbra Editora, 2010) 301–29; Gomes Canotilho, Direito Constitucional (n 3) 359–62. 20 See Otero, Organização do Poder Político (n 16) 209–80; Gomes Canotilho, Direito Constitucional (n 3) 619 ff; José Joaquim Gomes Canotilho and Vital Moreira, Os poderes do Presidente da República (Coimbra, Coimbra Editora, 1991) 30. 21 See Otero, Organização do Poder Político (n 16) 281–328; Gomes Canotilho, Direito Constitucional (n 3) 627 ff. 22 See Otero, Organização do Poder Político (n 16) 329–407; Gomes Canotilho, Direito Constitucional (n 3) 640 ff.

Portugal  631 and any court (including the Constitutional Court) shall not apply rules violating the CPR or the principles enshrined therein (Article 204 CPR).23 The Constitutional Court administers justice in matters of legal and constitutional nature (Article 221 CPR). It exercises abstract (Article 281 CPR) and concrete (Article 280 CPR) constitutionality control. The former can be ex ante (Article 278 CPR) or ex post (Article 281 CPR).24 The Constitutional Court has an active and influential role made even more evident during the Sovereign Debt Crisis.

II.  Constitutional Culture A.  The Constitution as a Result of Political Commitments The Portuguese Constitution is the outcome of certain political commitments25 between the Movimento das Forças Armadas and the political parties, on the one hand, and between the different political parties represented at the Constituent Assembly, on the other to achieve a common constitutional text.26 These political commitments have normative expression in the CPR through certain rules and principles, which are, however, hardly reconcilable and even contradictory. The drafters of the CPR left open the possibility for future political commitments to be made between the sovereign organs and within member of collegial organs. They did so by allowing for constitutional rules and principles to be further implemented through statutes.27

B.  The Constitution as a Living Document Although the Constitution is a rather rigid instrument, due to the existence of temporal,28 formal,29 material30 and circumstantial31 limits32 to constitutional amendments, it is open to 23 On the Portuguese judicial control of constitutionality see, among others Otero, Organização do Poder Político (n 16) 409 ff; Gomes Canotilho, Direito Constitucional (n 3) 657 ff; Jorge Miranda, Manual de Direito Constitucional, vol VI, 4th edn (Coimbra, Coimbra Editora, 2013) 189 ff. 24 On the Constitutional Court see among others Otero, Organização do Poder Político (n 16) 421–23; Gomes Canotilho, Direito Constitucional (n 3) 677 ff; Miranda, Direito Constitucional VI (n 23) 181 ff. 25 See Otero, Identidade Constitucional (n 18) 155–71. 26 See, above all Miranda, Revolução (n 2) 47–62, 96–106, 190–222, 241–85, 292–326, 337 ff. 27 See Otero, Identidade Constitucional (n 18) 163–71. 28 Art 284(1) CPR states ‘the Assembly of the Republic may revise the Constitution five years after the date of publication of the last ordinary revision law’ and para 2 of the same provision continues by stating that ‘however, by a majority of at least four-fifths of all the Members in full exercise of their office, the Assembly of the Republic may take extraordinary revision powers at any time’. 29 Art 285 CPR stipulates that ‘the competence to initiate revisions pertains to Members of the Assembly of the Republic and once a draft revision of the Constitution has been submitted, any others have to be submitted within a time limit of thirty days’; Art 286 CPR regulates that ‘amendments to the Constitution require passage by a majority of two-thirds of the Members of the Assembly of the Republic in full exercise of their office’ and that ‘amendments that are passed shall be combined in a single revision law’. According to para 3 ‘the President of the Republic may not refuse to enact the revision law’. 30 Article 288 CPR imposes that the Constitutional revision laws respect: a) national independence and the unity of the state; b) the republican form of government; c) the separation between church and state; d) citizens’ rights, freedoms and guarantees; e) the rights of workers, workers’ committees and trade unions; f) the coexistence of the public, private and cooperative and social sectors of ownership of the means of production;

632  Ana Maria Guerra Martins other solutions and legal orders. As such, the Constitution is not a definitive, closed, or static document, but a living instrument, open to new and better responses to solve the problems coming to light. These responses are reached through constitutional revision and/or reception of international or EU law. The living character of the CPR stems also from the interpretation of constitutional rules and principles by the judiciary and the legislature, the latter often enjoying large margin of discretion.33 Accordingly, the Portuguese Constitution, as many other constitutions, experienced an evolution leading to many divergences between the wording and the practice. To put it in other words, ‘law in books’ does not always coincide with ‘law in action’.34

C.  The Active and Very Influential Role of the Constitutional Court Undoubtedly, the Constitutional Court has been one of the most interventionist constitutional actors in Portugal. It has had an active and influential role ever since its creation.35 The Constitutional Court contributed crucially to establishing constitutional norms and principles by defining fundamental rights that are expressly mentioned in the CPR, ‘creating’ implied fundamental rights, explaining the human dignity concept, clarifying some aspects of the political system, and interpreting the rules and principles on the repartition of powers between the Republic and the autonomous regions. Giving an example that relates to fundamental rights, the Constitutional Court, beyond the ‘positive dimension’ of social rights (which includes a facere duty to the state), also acknowledged a ‘negative dimension’ (which includes a non facere duty to the state). In some cases the CPR establishes the concrete mechanisms for the realisation of social rights, leaving little discretion to the legislator. In these cases the Court decided that once a certain constitutional social right has been established by the legislator, it cannot subsequently be eliminated. For instance, according to Article 64(2)(a) CPR, the right to health care shall be fulfilled by a national health service. In April 1984, the Court accepted that the legislator was relatively free

g) the existence of economic plans, within the framework of a mixed economy; h) the appointment of the elected officeholders of the entities that exercise sovereignty, of the organs of the autonomous regions and of local government organs by universal, direct, secret and periodic suffrage; and the proportional representation system; i) plural expression and political organisation, including political parties, and the right of democratic opposition; j) the separation and interdependence of the entities that exercise sovereignty; l) the subjection of legal norms to review of their positive constitutionality and of their unconstitutionality by omission; m) the independence of the courts; n) the autonomy of local authorities; o) the political and administrative autonomy of the Azores and Madeira archipelagos. 31 Art 289 CPR reads ‘no act involving the revision of the Constitution may be undertaken during a state of siege or a state of emergency’. 32 On the Portuguese constitutional revision procedure and its limits see Jorge Miranda, Manual de Direito Constitucional, vol II, 7th edn (Coimbra, Coimbra Editora, 2013) 206 ff; Jorge Miranda and Rui Medeiros, Constituição Portuguesa Anotada, vol III (Coimbra, Coimbra Editora, 2007) 890–956; José Joaquim Gomes Canotilho and Vital Moreira, Constituição da República Portuguesa Anotada, vol II, 4th edn (Coimbra, Wolters Kuwer/Coimbra Editora, 2006) 995–1020; Gomes Canotilho, Direito Constitucional (n 3) 1059–68, Carlos Blanco e Morais, Curso de Direito Constitucional. Teoria da Constituição em Tempo de Crise do Estado Social, vol II.2 (Coimbra, Coimbra Editora, 2014) 268 ff. 33 See Otero, Identidade Constitucional (n 18) 173 ff. 34 Ibid, 207 ff. 35 Miranda, Manual I.2 (n 3) 271–73.

Portugal  633 to change the way the National Health Service was organised. However, it also underlined that such change cannot lead to the disappearance of the service, as that would be unconstitutional. Another example relates to Article 63(1) CPR. According to it, ‘everyone has the right to social security’ and para 3 adds ‘the social security system shall protect citizens … in any other situation that entails a lack of or reduction in means of substance’.36 In December 2002, the Court ruled that the legislator could not deprive people aged 18 to 25 from the right to the minimum guaranteed income.37 This change was derogated from a previously established regime, which was based, first, on the principle of respect of human dignity (Articles 1 and 2 CPR) and, secondly, on the right to social security and solidarity mentioned above. This influential and active role of the Constitutional Court increased further, during the public debt crisis, due to the so-called ‘case law of the crisis’, which will be briefly studied below.38

III.  Constitutional Foundations of EU Integration A. Accession After the entry into force of the CPR, and to some extent the stabilisation of the political situation, in March 1977, the Portuguese Government under Prime Minister Mário Soares applied for full membership to the European Communities (EC). The accession negotiations lasted almost 10 years due, inter alia, to economic disparities between Portugal and the EC Member States as well as the relatively recent democratisation of the state. Spain’s membership application of June 1977 also complicated the Portuguese negotiations. This was because the EC decided that the two states should negotiate together, leading to the problem of both states cross-influencing the negotiations. The accession treaty between Portugal and Spain and the EC was signed in Lisbon and Madrid on 13 June 1985. Subsequently, on 1 January 1986, Portugal joined the EC. The accession treaty followed the constitutional proceedings applicable to any other international treaty, which were similar to the ones applicable today. That is, the treaty was negotiated by the Government (former Article 200(1)(b), current Article 197(1)(b) CPR), approved by the AR (former Article 164(i), current Article 161(i) CPR) by a simple majority, and ratified by the President of the Republic (former Article 138(b), current Article 135(b)CPR).

B.  Amendments to the Constitution Due to European Integration Portugal’s EC accession necessitated changes in the Portuguese Constitution,39 because the ­original version made reference only to international law. The constitutional revision of 1982

36 Tribunal Constitucional Portugal (hereinafter TCPT) 11.04.1984, Acórdão 39/1984. Portuguese versions of all judgments can be found at www.tribunalconstitucional.pt/tc/acordaos/ under ‘Pesquisa’. English summaries of selected judgments are provided at www.tribunalconstitucional.pt/tc/en/acordaos/. 37 TCPT 19.12.2002, Acórdão 509/2002 (English summary available). 38 Section V.C. 39 On the constitutional consequences of Portugal’s accession to the EC see especially António Vitorino, ‘A adesão de Portugal às Comunidades Europeias’, (1984) 3 Estudos de Direito Público 9; Maria Isabel Jalles, Implications juridicoconstitutionnelles de l’adhésion aux Communautés européennes – Le cas du Portugal (Brussels, Bruylant, 1981) 165 ff; Maria Isabel Jalles, ‘Primado do Direito Comunitário sobre o Direito nacional dos Estados membros’, Documentação e Direito Comparado Separata No 4, Lisbon 1980.

634  Ana Maria Guerra Martins amended the constitutional provision on reception of international law (Article 8) by aiming to automatically internalise and integrate secondary EU law into the national legal order. Paragraph 3 of that provision reads: ‘[R]ules issued by the competent bodies of international organisations to which Portugal belongs shall come directly into force in Portuguese internal law, on condition that this is laid down in the respective constituent treaties.’ Although this provision was introduced because of Portugal’s accession to the EC, its wording is broad and unrestricted to the EC and the EU. It could also be applied to any other international organisation that would share direct applicability with EC law, which at that time was not very common, but nowadays is more frequent. The constitutional revision of 1989 amended the constitutional provision dealing with international relations (Article 7 CPR) by providing for Portugal’s participation in EU integration. Paragraph 5 of that Article states: ‘Portugal is committed to reinforcing the European identity and to strengthening the European states’ actions in favour of democracy, peace, economic progress and justice in the relations between peoples.’ The evolution and especially the deepening of EU integration in the early 1990s led to constitutional changes in Portugal and other Member States. In particular, the Portuguese Constitution had to be adapted to the Treaty of Maastricht.40 The constitutional revision of 1992 introduced such modifications. First, an ‘EU clause’ was added to Article 7 CPR stating that ‘Portugal may, subject to reciprocity conditions and to respect for the principle of subsidiarity, and with a view to the achievement of the economic and social cohesion, enter into agreements for the exercise jointly, of the powers needed to construct the European Union’. Second, a paragraph on foreigners and stateless persons was added to Article 15 CPR asserting that ‘under reciprocal terms, the law may also accord citizens of European Union Member States who reside in Portugal the eligibility to vote for and stand for election as Members of the European Parliament’. Furthermore, paragraph (f) was included in Article 166 CPR (today Article 163), stating that ‘[a]s laid down by law, the Assembly of the Republic has the competences to monitor and consider Portugal’s participation in the process of constructing the European Union’. Finally, paragraph (i) was added to in Article 200 CPR (today Article 197), so reading that the Government ‘for the purpose of Article 166 (f) and in good time, [has] to submit information concerning the process of constructing the European Union to the Assembly of the Republic’. The constitutional revision of 1997 did not aim at adapting the CPR to EU integration. However, two of the changes in that revision were directly connected to that matter. First, powers were conferred to the autonomous regions ‘to participate, when matters that concern their specific interest are at stake, in the process of constructing the European Union by means of their representation in the respective regional institutions and in the delegations involved in European Union decision-making processes’.41 Second, competence was conferred to the AR ‘to pronounce, as laid down by law, on matters awaiting decision by European Union organs that concern the sphere of its exclusive legislative competence’.42 Furthermore, the constitutional revision of 1997 introduced amendments allowing referenda on treaties that concern Portugal’s participation in international organisations.43

40 On the influence of the Treaty of Maastricht in the Portuguese Constitution see Jorge Miranda, ‘O Tratado de Maastricht e a Constituição Portuguesa’, in P de Pitta e Cunha (coord), A União Europeia na encruzilhada (Coimbra, Almedina, 1996) 45. 41 Art 227(1)(x) CPR. 42 Art 161(n) CPR. 43 Art 115 CPR.

Portugal  635 The constitutional revision of 2001 dealt mainly with adapting the CPR to the Rome Statute. In particular, a paragraph was added to Article 7 CPR, stating that With a view to achieving an international justice that promotes respect for the rights of the human person and of peoples, and subject to the provisions governing complementarity and the other terms laid down in the Rome Statute, Portugal may accept the jurisdiction of the International Criminal Court.

Article 7, however, was also amended to take account of the Area of Freedom, Security and Justice, which was introduced by the Treaty of Amsterdam. This amendment read as follows: Subject to reciprocity and to respect for the fundamental principles of a democratic state based on the rule of law and for the principle of subsidiarity, and with a view to the achievement of the economic and social cohesion and of an area of freedom, security and justice, Portugal may enter into agreements for the exercise jointly or in cooperation, of the powers needed to construct and deepen the European Union.

The constitutional revision of 2004 amended Article 8 CPR by adding that The provisions of the treaties that govern the European Union and the rules issued by its institutions in the exercise of their respective responsibilities shall apply in Portuguese internal law in accordance with Union law and with respect for the fundamental principles of a democratic state based on the rule of law.

This amendment implicitly consecrated the principle of primacy of EU law over domestic law, which was included – as is well-known – in Article I-6 of the (failed) Treaty establishing a Constitution for Europe (TECE). Although the TECE failed due to the negative outcomes of the referenda in France and the Netherlands, this constitutional amendment remained in force and now applies to the Treaty of Lisbon. That means ‘the Treaties and the law adopted by the Union on the basis of the Treaties have primacy over the law of Member States, under the conditions well settled by the case law of the Court of Justice of the European Union’, as Declaration 17 concerning primacy recalls. That amendment of Article 8 was strongly criticised by the most traditional part of the doctrine,44 due to the loss of sovereignty involved. Furthermore, the increased importance of the EU common foreign security and defence policy prompted an amendment of Article 7(6) CPR, where it is now mentioned. The last constitutional revision happened in 2005 with the exclusive aim of allowing referenda on European treaties. Taking into consideration that the TECE was signed in December 2004 and had introduced deep and wide changes to the European treaties, a strong political consensus emerged in Portugal for a referendum on accepting the treaty. However, a constitutional basis for such referendum was lacking. Consequently, the constitutional revision of 2005 added Article 295 CPR, stating that ‘The provisions of Article 115(3) do not prejudice the possibility of calling and holding referenda on the approval of treaties concerning the construction and deepening of the European Union.’

C.  Current Provisions Specifically on EU-Membership The current EU Membership CPR provisions are Article 7(5) and (6), containing the so-called ‘EU clause’, and Article 8(2), (3) and (4) relating to the reception of primary and secondary EU

44 See Jorge Miranda, ‘Artigo 8’, in J Miranda and R Medeiros, Constituição Portuguesa Anotada, vol I, 2nd edn (Coimbra, Coimbra Editora, 2010) 172.

636  Ana Maria Guerra Martins law in the Portuguese legal order, and the relationship between EU law and Portuguese domestic law, including the CPR. Article 7(5) and (6) CPR states 5. 6.

Portugal shall make every effort to reinforce the European identity and to strengthen the European states’ actions in favour of democracy, peace, economic progress and justice in the relations between peoples. Subject to reciprocity and to respect for the fundamental principles of a democratic state based on the rule of law and for the principle of subsidiarity, and with a view to the achievement of the economic, social and territorial cohesion of an area of freedom, security and justice and the definition and implementation of a common external, security and defence policy, Portugal may enter into agreements for the exercise jointly, in cooperation or by the Union’s institutions, of the powers needed to construct and deepen the European Union.

The current wording of Article 8(2)-(4) CPR is the following: 2.

The norms contained in duly ratified or approved international conventions come into force in Portuguese internal law once they have been officially published, and remain so for as long as they are internationally binding on the Portuguese state. 3. The norms issued by the competent organs of international organisations to which Portugal belongs come directly into force in Portuguese internal law, on condition that this is laid down in the respective constituent treaties. 4. The provisions of the treaties that govern the European Union and the norms issued by its institutions in the exercise of their respective competences are applicable in Portuguese internal law in accordance with Union law and with respect for the fundamental principles of a democratic state based on the rule of law.

The CPR also refers to EU integration in other provisions, such as Article 15(5) regarding EU citizens; Articles 161(n) and 163(f) on the pronouncement of the AR on matters awaiting decision by EU organs that concern the sphere of its exclusive legislative competence and on the monitoring and consideration of Portugal’s participation in the process of constructing the EU, respectively; Article 197(i) relating to the duty of the government to submit, in good time, information concerning the process of constructing the EU to the AR; Article 227(1)(x) on the participation of autonomous regions in certain circumstances in the process of constructing the EU; and Article 295 on the EU treaties referenda. We will return to some of these provisions below.

D.  Judicial Control As already underlined,45 since 1997 the CPR contains a provision on referenda concerning the approval of treaties relating to Portugal’s participation in international organisations (except peace treaties and treaties rectifying borders). Such treaties have to be approved by the AR. The treaty in question should already have been signed by the Government but not yet formally approved by the AR and ratified by the President. However, no referendum on the ‘European question’ is allowed to question Portugal’s past participation or future membership in the EU. According to current Article 115(8) CPR, ‘the President of the Republic shall submit all draft referenda submitted to him by the Assembly of the Republic or the government, to compulsory prior review of



45 Section

III.B.

Portugal  637 their constitutionality and legality’. In recent Portuguese constitutional history, the President of the Republic has used this prerogative twice in cases related to European treaties. First, a proposal for a referendum on the Amsterdam Treaty submitted by the AR was referred by the President to the Constitutional Court. The question proposed for referendum was: ‘Do you agree that Portugal should continue to participate in the process of construction of the EU through the Treaty of Amsterdam?’ The proposal also stipulated that citizens registered as voters within the national territory and Portuguese citizens resident in the Member States of the EU would be entitled to take part in the referendum. In Ruling No 531/98 of 27 July 1998,46 the Constitutional Court found that the question did not meet the requirements of clarity and exactitude because it was open to more than one interpretation and the wording made it impossible to determine the significance of the reference to the Treaty of Amsterdam. The question also did not fulfil the objectivity requirement, because it was worded in such a way as to force voters who were in favour of Portugal’s continued participation in the process of construction of the EU to vote yes, and the key issue of the referendum, namely the changes brought about by the adoption of the Treaty of Amsterdam, was played down. The Court therefore declared the proposed referendum unconstitutional and unlawful and as a result the President rejected the proposal.47 Secondly, the question of the preliminary review of the constitutionality and legality of an EU Treaty came up again with the TECE. Taking into account that, in accordance with Article 115(6) CPR and Article 7(2) of the Organic Law on Referenda, the question forming the subject of the referendum must be formulated in an objective, clear, and precise way and must permit a ‘yes’ or ‘no’ answer, without suggesting, directly or indirectly, the significance of the answer, in December 2004 the Constitutional Court checked whether these conditions were met.48 In the case in point, the Constitutional Court decided that the question ‘Do you agree with the Charter of Fundamental Rights, the qualified majority voting rule and the new institutional framework of the European Union under the Constitution for Europe?’ combines three questions within one, for which a single answer is required. The question was problematic with respect to the constitutional and legal provisions requiring a clear, explicit, and unambiguous question. The lack of clarity stemmed mainly from the fact that there were three questions in one. Furthermore, the fact of assigning more than one meaning to the question established its ambiguity and lack of clarity. Neither was the question as a whole worded in a clear, explicit, and unambiguous manner, nor were the individual questions. Combining the three questions into one infringed the constitutional and legal requirements of allowing for a ‘yes’ or ‘no’-type answer. In conclusion, the Constitutional Court ruled that the proposal for a referendum on a Constitution for Europe was neither constitutional nor legal.

E.  Relevance of (General) Provisions on International Law As regards international law, the CPR includes a fundamental principle of ‘openness’ to international law.49 According to Article 8(1) CPR, ‘the rules and principles of general or common

46 TCPT 27.07.1998, Acordão 531/98 (English summary available). 47 The decision of the Constitutional Court was taken by a majority. Six judges delivered dissenting opinions. 48 TCPT 17.12.2004, Acordão 704/04 (English summary available). 49 Compare Ana Maria Guerra Martins and Miguel Prata Roque, ‘Chapter 18 – Universality and Binding Effect of Human Rights from a Portuguese Perspective’, in R Arnold (ed), The Universalism of Human Rights (Dordrecht, Springer, 2013) 297, 307; Ana Maria Guerra Martins, ‘A Portuguese Perspective of the Accession on the European Union to the European Convention of Human Rights’, in I Iliopoulos-Strangas, V Pereira da Silva and M Potacs (eds), The Accession

638  Ana Maria Guerra Martins international law shall form an integral part of Portuguese law’. Furthermore, Article 16(1) CPR states that ‘the fundamental rights enshrined in this Constitution shall not exclude such other rights as may be laid down by law and in the applicable rules of international law’ and Article 16(2) CPR confers a supra-constitutional value to the Universal Declaration of Human Rights (UDHR), given that ‘the provisions of this Constitution and of laws concerning fundamental rights shall be interpreted and construed in accordance with the Universal Declaration of Human Rights’. In other words, Article 16(1) CPR foresees an ‘opening clause’ in matters of fundamental rights. This means that the CPR receives all human rights contained in other sources of law (including in international law), as if they were part of the CPR. Also, Article 16(2) CPR expressly assures that all constitutional principles and norms shall be interpreted and applied in the light of the UDHR. In a 2010 ruling on same-sex marriage law, the Constitutional Court clarified that Article 16(2) CPR could only be used with the aim of increasing the protection of human rights, and not to diminish it.50 According to the Court, any restriction of a human right should be directly and expressly foreseen by the CPR and not by means of an indirect interpretation of an international text, even when it is binding. Hence, the Court stated that the UDHR could not be used as a tool to restrain the full range of the constitutional protection to the right to marry.51 As regards international covenants, Article 8(2) CPR states that ‘the rules set out in duly ratified or passed international agreements shall come into force in Portuguese internal law once they have been officially published, and shall remain so for as long as they are internationally binding on the Portuguese State’. That is to say that all international conventions are binding on the Portuguese Republic, when duly ratified52 or signed53 by the Head of State, on the condition that they have also entered into force at the international level and for as long as they remain so. The ratification or the signature of the Head of State depends on them being considered ‘treaties’ or ‘agreements’.54 Regarding Article 8(3) CPR, ‘rules issued by the competent bodies of international organizations to which Portugal belongs shall come directly into force in Portuguese internal law, on condition that this is laid down in the respective constituent treaties’. As mentioned above, this provision was introduced by the constitutional revision of 1982, against the background the Portuguese accession to the EC. It aimed to automatically incorporate secondary EU law and integrate it into the national legal order without any procedure of reception. However, taking the wording into consideration, the provision can also be applied outside EU law.

of the European Union to the ECHR (Baden-Baden, Nomos, 2013) 201; Rui Medeiros, A Constituição Portuguesa num Contexto Global (Lisbon, UCP, 2015) 291 ff. 50 TCPT 08.04.2010 Acordão 121/2010 (English summary available). 51 For a different view, Decision 121/2010 even quotes a judgment of the South African Constitutional Court (01.12.2005, Minister of Home Affairs and Another v Fourie and Another (CCT 60/04) [2005] ZACC 19), whereby it considered that the wording of Art 16(1) UDHR should be interpreted as only allowing an opposite-sex marriage (although it also stated that this conception was cultural and historically dated). 52 See Art 135(b) CPR, stating ‘In international relations the President of the Republic has the competences: … b) Once they have been duly approved, to ratify international treaties’. 53 See Art 134(b) CPR, reading ‘In the practice of personal acts the President of the Republic has the competences: […] to sign both resolutions of the Assembly of the Republic that approve international agreements and the rest of the Government’s decrees’. 54 Under Art161(i) CPR, the AR has the competence to approve treaties, particularly those that concern Portugal’s participation in international organisations and the ones concerning friendship, peace, defence, the rectification of borders or military affairs, as well as international agreements that address matters in which the AR has exclusive competence (see Arts 164 and 165 CPR) or which the Government deems fit to submit to the AR for consideration.

Portugal  639

F.  Practice and Doctrinal Debate on General International Law The position of rules and principles of general international law within the hierarchy of the Portuguese sources of law is governed by Article 8(1) CPR. It states that they are considered ‘an integral part of Portuguese law’, without responding to the question whether they prevail or not over national law, including the CPR. There is broad consensus that peremptory norms of international law (jus cogens)55 prevail over national law, including constitutional law, and that rules and principles of general international law only prevail over national legislative acts and they must respect the CPR.56 Regarding international covenants, Article 8(2) CPR does not give any clear solution to the problem of possible conflicts between them and constitutional law. In order to prevent these conflicts, Article 278(1) CPR57 foresees an abstract preventive control (ex-ante review) of constitutionality. However, the fact is that the ex-ante constitutionality review does not completely prevent conflicts. They can emerge and the constitutional provisions do not provide a clear solution. As a consequence, Portuguese scholars and judicial authorities have been expressing different points of view. The major stream supports the view that international sources of law prevail over ordinary law but they do not prevail over the CPR,58 based on the following arguments: • Article 204 CPR concerning compliance with the CPR reads: ‘In matters that are brought to trial, the courts shall not apply rules that contravene the provisions of this Constitution or the principles enshrined therein’. As the provision does not distinguish between internal sources of law and international ones, the scholars consider that the courts shall not apply any legal source, including an international one that contravenes the CPR; • Article 277 CPR allows the Constitutional Court to control the compatibility of any rule – including international ones – with the principles and rules included in the CPR; • Article 278 CPR permits an abstract preventive control of constitutionality of rules of the international conventions. These conclusions do not apply to the relationship between primary EU law and domestic law, because, as mentioned above,59 the CPR contains a specific rule on this issue, studied deeper below.60 To sum up, for the major stream of scholarship, international rules, with the only exception of those containing jus cogens, do not prevail over constitutional law.

In accordance with Art 197(1)(b)–(c) CPR, the Government has the competence to negotiate and finalise international conventions and to approve international agreements whose approval is not within the competences of, or which have not been submitted to, the AR. 55 According to Arts 53 and 64 VCLT, some provisions of the universal international human rights law are qualified as peremptory norms of international law (jus cogens), eg the prohibition of torture or genocide and the right to selfdetermination. They prevail over every single rule and they do not allow their removal through any kind of corrective interpretation. 56 Jorge Miranda, Curso de Direito Internacional Público, 4th edn (Cascais, Principia, 2009) 116, 153; José Joaquim Gomes Canotilho and Vital Moreira, ‘Commentary to Article 8’, in J J Gomes Canotilho and V Moreira (eds), CRP – Constituição da República Portuguesa – anotada, vol 1 (Coimbra, Coimbra Editora, 2014) 258 f. 57 ‘The President of the Republic may ask the Constitutional Court to conduct a prior review of the constitutionality of any rule laid down by an international treaty that is submitted to him for ratification, … or by any international agreement, the decree passing which is sent to him for signature’. 58 Miranda, Curso (n 56) 155 f; Gomes Canotilho/Moreira, ‘Article 8’ (n 56) 260; Eduardo Correia Baptista, Direito Internacional Público, vol I, Conceito e fontes (Lisbon, Lex, 1998) 438–442. 59 Sections III.B. and III.C. 60 Section V.A.

640  Ana Maria Guerra Martins

IV.  Constitutional Limits to EU Integration A.  Limits to the Further Transfer of Powers to the EU Through Treaty Amendments The constitutional limits to further transfer of powers to the EU through Treaty amendments derive from Article 7(6) CPR – the ‘EU clause’. Additionally, one can mention Article 288 CPR – the ‘material limits of constitutional revision clause’ – as it can also be considered relevant in this context.61 According to Article 7(6) CPR, the first limit to the further transfer of powers to EU integration is the principle of reciprocity, which is founded on the principle of equality between states mentioned in Article 7(1) CPR. That is to say the CPR solely accepts the transfer of powers from Portugal to the EU under equal circumstances with other Member States. To put it in other words, all Member States have to transfer similar powers to the EU. However, this principle should not be interpreted too rigidly. This is necessary to allow the opting out of some Member States in relation to certain policies, as agreed in past Treaty revisions. In these cases, reciprocity should be assessed in general terms and not in an individualised manner, checking and comparing every single transfer of power. Similar powers may not exactly mean the same. The second limit included in Article 7(6) CPR concerns the fundamental principles of a democratic state based on the rule of law,62 which, according to Article 2 CPR, defines the model of the Portuguese political power. It is founded on four assumptions: • the sovereignty of the people;63 • the plural democratic expression and political organisation; • the respect for and the guarantee of the effective implementation of the fundamental rights and freedoms; and • the separation and interdependence of powers.64 Moreover, in accordance with Article 2 CPR, the democratic state based on the rule of law aims at the achievement of an economic, social, and cultural democracy and the deepening of participatory democracy. Consequently, the further transfer to the EU must respect the democratic principle,65 the pluralist principle,66 the fundamental rights and freedoms, and principle of separations and interdependence of powers, as well as the guiding principles of a democratic state based on the rule of law, which, for the CPR, are the welfare state principle67 and the principle of citizens’ participation.68 61 See Carlos Blanco de Morais, ‘A sindicabilidade do direito da Unão Europeia pelo Tribunal Constitucional Português’, in J Miranda (ed), Estudos em Homenagem ao Prof. Doutor Sérvulo Correia vol I (Coimbra, Coimbra Editora, 2011) 221, 250. 62 Otero, Identidade Constitucional (n 18) 51–55; Miranda, Manual III (n 19) 67 ff. 63 Miranda, Manual III (n 19) 759–767; Gomes Canotilho, Direito Constitucional (n 3) 292. 64 Otero, Organização do Poder Político (n 16) 11–16; Alexandre Sousa Pinheiro and Pedro Lomba, ‘Commentary to Article 111’, in P Otero (ed), Comentário à Constituição Portuguesa (Coimbra, Almedina, 2008) 55–68. 65 Otero, Identidade Constitucional (n 18) 119–37; Pedro Lomba, ‘Commentary to Article 108’, in P Otero (ed), Comentário à Constituição Portuguesa (Coimbra, Almedina, 2008) 11–24; Gomes Canotilho, Direito Constitucional (n 3) 287–91; Gomes Canotilho/Moreira, Fundamentos (n 3) 76–85. 66 Otero, Identidade Constitucional (n 18) 56–75. 67 Ibid, 100–06. 68 Alexandre Sousa Pinheiro, ‘Commentary to Article 109’, Pedro Lomba, ‘Commentary to Article 108’, in P Otero (ed), Comentário à Constituição Portuguesa (Coimbra, Almedina, 2008) 25–35; Gomes Canotilho, Direito Constitucional (n 3) 301.

Portugal  641 The third limit to further transfer of powers to the EU in Article 7(6) CPR is the principle of subsidiarity, which is mentioned in Article 6(1) CPR, on the characterisation of the Portuguese state as a unitary state. Under this principle, Portugal should not accept the transfer of powers to the EU if the exercise of the proposed powers could be better achieved at the national level. Furthermore, according to the CPR, further transfer of powers to the EU should be teleologically oriented – it should respect certain objectives. Thus, future Portuguese agreements to jointly exercise the powers needed to construct and deepen the EU must respect the aim of the achievement of the economic, social, and territorial cohesion of an area of freedom, security and justice, and the definition and implementation of a common foreign, security and defence policy. Accordingly, Article 7(6) CPR does not give a blank cheque to the political power to further transfer powers to the EU. Instead, several constitutional limits must be respected. So far, no serious conflicts regarding the observance of the ‘EU clause’ have emerged. This has been due to the existence of wide consensus among politicians and the civil society that Portugal should participate in all EU integration developments. However, recently, an initial lack of consensus between the two major political parties represented in the AR occurred concerning the participation of Portugal in the permanent structured cooperation (PESCO). The Government and the Socialist Party’s Members of the Parliament supported the participation of Portugal, whereas the Members of Parliament belonging to the Social Democratic Party recommended non-participation. When the Council Decision (CFSP) 2017/2315 on PESCO was adopted, this divergence was overcome and according to its Article 2, Portugal is one of the participating Member States.69 Apart from Article 7 CPR, the ‘material limits of constitutional revision clause’ – Article 288 CPR – also contain constitutional limits to further EU integration. Article 288 CPR includes a significant number of areas that, in principle, cannot be modified even by a constitutional amendment. This leads to the question whether it is possible to transfer powers under those areas to the EU, without convening a new pouvoir constituant, to adopt another constitution. In particular, given that the CPR formally has higher standing than treaties and imposes limits on the amendment of certain matters, how could these limits be overcome by the transfer of powers in such fields to the EU? At first glance that would be impossible unless another Constitution is enacted, following, for instance, a constitutional rupture, such as a revolution. Nevertheless, this result is not that obvious. First, the real meaning of Article 288 CPR must be understood in its historical context. After almost 50 years of a dictatorial government, the members of the Constituent Assembly intended to prevent at any cost a regression of democracy. Therefore, many limits had been included, which nowadays do not make sense. As a consequence, the current meaning of the clause is still disputed among Portuguese scholars.70 For some, the clause itself may be amended. This can happen by deleting a certain material limit and amending a substantive provision in the same or future constitutional amendment procedure. This is the thesis of the double constitutional revision.71 Other scholars maintain that the material limits may be interpreted in a rather restrictive way in order to allow extensive changes of the CPR.72 Very few scholars support a rigid interpretation of this clause. Unfortunately, this matter cannot be discussed in further detail here.

69 Council Decision (CFSP) 2017/2315 of 11 December 2017 establishing permanent structured cooperation (PESCO) and determining the list of participating Member States, [2017] OJ L 331/57. 70 On the terms of this debate see Miranda, Manual II (n 32) 221 ff; Miranda/Medeiros, Constituição Anotada vol III (n 32) 927–51; Gomes Canotilho/Moreira, Constituição Anotada vol II (n 32) 1010–19; Gomes Canotilho, Direito Constitucional (n 3) 1064–68; Blanco e Morais, Curso (n 32) 275 ff. 71 Miranda, Manual II (n 32) 259 ff; 72 See Medeiros, Constituição Portuguesa (n 49) 207 ff.

642  Ana Maria Guerra Martins Secondly, Article 288 CPR was not conceived as a clause limiting the external power of the Portuguese state. Instead, it was mean to limit the internal power of constitutional revision. Therefore, it is disputable whether it can be invoked with the aim to prevent further transfer of powers to the EU. Notwithstanding, in our opinion, future steps towards further EU integration must, at least, respect the existence and the main features of the Portuguese state structure. As a consequence, the national independence and the unity of the Portuguese state (Article 288(a) CPR), the republican form of government (Article 288(b) CPR), the autonomy of local authorities (Article 288(n) CPR), and the political and administrative autonomy of the Azores and Madeira archipelagos (Article 288(o) CPR) should be respected not because they belong to the ‘material limits of constitutional revision clause’ but because they constitute the constitutional identity of Portugal. To give an example, the CPR could not accommodate changes such as Portugal accepting to become part of a Federal State, losing its independence, or modifying its republican form of government and turning into a monarchy. A new constitution would be needed instead. Similar reasoning applies to material limits relating to the respect for the fundamental principles of a democratic state based on the rule of law, because they constitute the core of the CPR and the Portuguese constitutional identity. This core and identity cover the separation between church and state (Article 288(c) CPR); citizens’ rights, freedoms, and guarantees (Article 288(d) CPR); the rights of workers, workers’ committees, and trade unions (Article 288(e) CPR); plural expression and political organisation, including political parties, and the right of democratic opposition (Article 288(i) CPR); the separation and interdependence of the entities that exercise sovereignty (Article 288(j) CPR); and the independence of the courts (Article 288(m) CPR). To conclude, unless the Portuguese people decide otherwise by creating a new Constitution, the current constitutional identity of the Portuguese State, which is contained in Article 7(6) CPR and Article 288(a) to (e), (i), and (j) to (o) CPR must be respected.

B.  Scrutiny of Secondary Legislation, Especially Ultra Vires Doctrine Although neither case law nor Portuguese scholarship have so far deeply discussed the ultra vires doctrine, in our view, it also applies in Portugal, taking into consideration Article 8(4) CPR. This provision states that ‘the rules issued by its institutions in the exercise of their respective responsibilities shall apply in Portuguese internal law’. That means that the CPR only assures the application of secondary EU law rules over national law provided that those rules were adopted by the EU institutions in the exercise of their respective competences. If EU institutions exceed their powers, the CPR does not guarantee the application of secondary EU law in Portugal. However, the CPR does not (and cannot) define the competence limits of EU institutions. This is because the competences of EU institutions are established in the EU treaties and not in the Member States’ Constitutions. When these competences are not respected, there is, first, a breach of those treaties and not a breach of the CPR. Looking at it differently, in principle, when EU institutions adopt an ultra vires act, the powers they are exercising still belong to the Member States. This is because, according to Article 4(1) TEU, the ‘competences not conferred upon the Union in the Treaties remain with the Member States’. That means that ultra vires acts interfere with the competences that the Member States did not transfer to the EU. Taking this into account, EU ultra vires acts constitute not only a problem of EU law, but also of Member States’ constitutional law. Therefore, a conflict can arise and its solution is far from simple. Even identifying the final arbiter to solve this conflict is rather complex.

Portugal  643 As already pointed out,73 in the Portuguese constitutional system the guardians of Portuguese sovereignty are either the Constitutional Court or the national courts altogether. They have this role because they may not apply rules and principles violating the CPR. Consequently, from a strictly constitutional point of view, an ultra vires act is an act that concerns a matter that was not transferred to the EU. That is to say, provided it contradicts the CPR, it may be judged or declared unconstitutional. Nevertheless, from an EU perspective, an ultra vires act is a question internal to Article 19(1) TEU, if a conflict arises it shall be resolved by the CJEU, which is the ultimate guardian of EU law. The CJEU has recently re-confirmed the exclusivity of its jurisdiction in EU law matters.74 These two apparently contradictory perspectives must be reconciled. In federal states, this reconciliation presupposes a judicial hierarchy between the national courts and the federal courts, the latter having the last word. However, such hierarchy does not exist within the EU. Therefore, it is necessary to find a solution elsewhere. According to Article 8(4) CPR, national courts, whatever they are, may not apply domestic law, including constitutional law, which contravenes EU law, as long as ‘the provisions of the treaties that govern the European Union and the norms issued by its institutions … are applicable in Portuguese internal law in accordance with Union law’. In other words, the CPR incorporates the principle of supremacy/primacy75 of law over national law.76 Any national court suspecting an ultra vires act and considering that a decision on this question is necessary for it to deliver a judgment, may make a preliminary reference to the CJEU, under Article 267 TFEU (including the Constitutional Court).77 Thus, the national court is not allowed to declare an ultra vires act either invalid or unconstitutional. If such question is raised in a case pending before a court or tribunal of a Member State against whose decisions there is no judicial remedy under national law, the reference to the CJEU is obligatory. Even where the CJEU delivers its preliminary ruling, it is not guaranteed that the ultra vires problem is solved. However, given that such a question has never arisen in Portugal, it is difficult to elaborate more on this issue.

C.  Limits to European Integration Outside the EU Legal Order Besides integration within the EU legal order, the Member States have always been cooperating in fields connected to EU law. This had, for instance, originally been the case with the Schengen agreements, before their inclusion in EU law. However, in recent years such cooperation also concerns core subjects of EU integration, such as the EMU. Therefore, it is necessary to assess whether there are limits to EU integration outside the EU legal order. Above all, it needs to be highlighted that the CPR contains some rules on the financial and fiscal system (Articles 101 to 107 CPR). However, none of them refers either to the Stability and Growth Pact (SGP), to the Treaty on the European Stability Mechanism (TESM), or to the Treaty on the Stability, Coordination and Governance in the EMU (TSCG).

73 Section I.E. 74 CJEU 18.12.2014 Avis 2/13 (Accession of the Union to the ECHR) ECLI:EU:C:2014:2454. 75 As we will see in section V.A., in Portugal there is no discussion on the difference between supremacy and primacy as occurs in other Member States. 76 For further developments on the supremacy/primacy of EU law over Portuguese law, see Ana Maria Guerra Martins, Manual de Direito da União Europeia, 2nd edn (Lisbon, Almedina, 2017) 517 ff. 77 The TCPT has made a first preliminary reference to the CJEU on 9 December 2020 (Acórdão 711/20), as we will see in section VI.B.

644  Ana Maria Guerra Martins Due to the budgetary content of these instruments, especially the latter one, some scholars supported the idea that the CPR should be amended.78 This was corroborated by the XIX Constitutional Government, which argued that the ‘golden rule’ should be included in the CPR. However, it was impossible at the time to reach an agreement with the opposition on this matter. As a consequence, the TESM and the TSCG were approved by two resolutions79 of the AR, on 13 April 2012, almost without discussion.80 The reason for the lack of discussion was that the same day involved a debate on municipalities, which drew much more attention. Actually, the political debate on the TESM was almost non-existent and the discussion on the TSCG was dominated by two main topics – convening a referendum, which was supported by the Communist Party,81 the Left Block82 and the Green Party,83 and a recommendation, supported by the Socialist Party, to the Government to propose to the other signatory states, and within the framework of the EU, the adoption of measures or the negotiation of an additional protocol with the aim to promoting economic growth and employment.84 All of them were rejected by the majority that supported the Government. None of these instruments were challenged before the Constitutional Court. Eventually, a reference to the ESM and the TSCG, was inserted into the new Budget Framework Law (Law No 151/2015 of 11 September 201585). This law has an enhanced value, in a double sense. On the one hand, under the terms of Article 106(1) CPR, the Budget Law shall be in accordance with the applicable framework law, and, on the other hand, Article 4 of Law No151/2015 expressly asserts that this law ‘prevails over all rules establishing particular budgetary regimes not compatible with’ (Article 4). The new Portuguese Budget Framework Law86 is much more inspired by the treaties approved by Member States outside the EU legal order than by the EU legal order. The CPR thus plays an insignificant role in that regard.

V.  Constitutional Rules and Practice on Implementing EU Law A.  Participation of the Assembly of the Republic in Secondary Law-making As already mentioned,87 under the terms of Articles 161(n), 163(f), 164(p) and 197(1)(i) CPR, the AR has competences with regard to monitoring, assessment and pronouncement on Portugal’s participation in the construction of the EU. These provisions were implemented by Law No 43/2006 of 25 August 2006, as amended by Law No 21/2012 of 17 May 2012,88 and more recently 78 Vital Moreira, ‘A CRP e a União Europeia’, in Estudos em homenagem ao Conselheiro Presidente Rui Moura Ramos, vol I (Coimbra, Almedina, 2016) 869, 898. 79 www.parlamento.pt/ActividadeParlamentar/Paginas/DetalheIniciativa.aspx?BID=36902. 80 www.parlamento.pt/ActividadeParlamentar/Paginas/DetalheIniciativa.aspx?BID=36902. 81 Draft Resolution No. 279/XII, 1ª, of 5/4/2012, www.parlamento.pt/ActividadeParlamentar/Paginas/DetalheIniciativa. aspx?BID=36902. 82 Draft Resolution No. 268/XII, 1.ª, of 28/3/2012, www.parlamento.pt/ActividadeParlamentar/Paginas/ DetalheIniciativa.aspx?BID=36902. 83 Draft Resolution No.281/XII, 1.ª, of 5/4/2012, www.parlamento.pt/ActividadeParlamentar/Paginas/DetalheIniciativa. aspx?BID=36902. 84 Draft Resolution No. 283/XII, 1.ª of 5/4/2012, www.parlamento.pt/ActividadeParlamentar/Paginas/DetalheIniciativa. aspx?BID=36902. 85 Diário da República, 1st série, No 178 of 11 September 2015. 86 Compare www.en.parlamento.pt/Legislation/Law151_2015.pdf. 87 See section III.C. 88 Available at https://dre.pt/application/file/551921.

Portugal  645 by Law No 18/2018 of 2 May 2018.89 This law also provides for a regular process of consultation with the Government (Article 1(2)), which assumes particular relevance when comes to PESCO (see Article 4(1)(j)–(k) introduced by Law No 18/2018). The AR shall also pronounce itself on other initiatives of EU institutions (Article 1-A), specifically on conformity of EU legislative proposals with the principle of subsidiarity (Article 3) and, in general terms, on EU policy documents and proposals for action (Articles 4 and 7), and sets out the parliamentary assessment process. When matters falling within the AR’s exclusive legislative competence are pending before EU bodies, it shall pronounce itself thereon in accordance with the procedure foreseen in Article 2 of the law. In that case, the Government must inform the AR and ask for an opinion, and shall in a timely manner send an information note containing a summary of the draft act, an analysis of its implications and, if already defined, the position the Government wishes to adopt (Article 2(2) of the law). This process involves the whole Parliament, from the committees with responsibility for the matter in question to the European Affairs Committee (EAC), and the Plenary itself. The AR, via the EAC, which is the parliamentary committee with the competence to monitor and assess European affairs in overall terms (Article 6(1) of the law), receives the initiatives from the various European institutions, and refers them, depending on their subject matter, to the various parliamentary committees, which may decide to draft a Report. At the same time, if the subject matter in question falls within the competence of the Legislative Assemblies of the Autonomous Regions, they should be consulted (Article 3(4) of the law). If the draft acts fall within the sphere of the AR’s exclusive legislative competence, the Government shall request an Opinion (Article 2(2)). When the standing parliamentary committee approves a Report, it is sent to the EAC (Article 2(3)). The EAC then prepares a final Written Opinion, in which it pronounces on the compliance of a particular initiative with the principle of subsidiarity and reflects any concerns about the substance of the European proposal expressed by the committee with responsibility for the matter in question (Article 6(3)). This Written Opinion is considered and voted upon in an EAC meeting. Finally, the Written Opinion, approved by the EAC, together with the Report of the competent committee, is sent by the President of the AR to the European institutions and the Government (Article 7(7)). In the event that the proposal under consideration is deemed to breach the principle of subsidiarity or has politically important implications for Portugal, the EAC may draw up a draft resolution for submission to the Plenary for approval (Article 3(3)).The process described above must be completed within eight weeks, as foreseen by the Treaty of Lisbon’s Protocol on Subsidiarity. The scrutiny of the other European initiatives follows mutatis mutandis the same procedure. The EAC has also adopted a specific procedure, the so-called ‘enhanced scrutiny’, for what it regards as priority issues. Since the entry into force of the law in 2006, the AR has successfully implemented a methodology for the systematic monitoring of European initiatives. This methodology resulted in the scrutiny of a significant number of European initiatives. It has also generated important internal momentum involving the EAC and the standing parliamentary committees. The result of this work is reflected in the progressive increase in the number of parliamentary opinions sent to European institutions. The result is reflected also in the fact that the AR has gained a reputation as the most active national Parliament in terms of participation in the European legislative

89 Available

at https://dre.pt/application/conteudo/115203957.

646  Ana Maria Guerra Martins process, having sent the most Written Opinions to the European Commission among all national parliaments, between 2008 and 2011.90

B.  Techniques of Implementing Secondary EU Legislation One has to distinguish between legislative and administrative implementation. Starting with the legislative one, according to Article 112(8) CPR, the transposition of EU legal Acts, especially of directives, into the internal legal order shall take the form of a law, a decree-law, or a regional legislative decree, depending on the content of the Act. First, one has to decide whether the subject matter of the EU legal act covers regional issues. If it does, one has to look at the repartition of powers between the Republic and the autonomous regions, in order to decide which one is competent to implement the Act. Provided it is the latter, the regional legislative decree will be the appropriate form to implement the EU Act. Should the matter be implemented by the Republic, it is necessary to clarify whether it is an exclusive competence of the AR or not. If it is, the implementation Act would be a law. Should the content of the EU Act come under concurrent competence of the AR and the Government, the implementation may take the form of either a law or a decree-law. Considering that in Portugal the Government is the main legislator, the majority of the EU Acts, especially directives, are implemented by decree-laws. Finally, the implementation of EU legal Acts may also need administrative execution. This execution may depend either on administrative sources or administrative practices. In this case, administrative regulations may be the most adequate means to implement EU Acts.

C.  Enforcement Through the Courts Portuguese courts, like any EU Member State courts, should interpret and apply EU law in their decisions, since they are the common judges of EU law.91 In some occasions they have to directly apply EU law and in others they apply the domestic law that transposes EU legislation. There are also cases where the courts use EU law as a rhetoric ‘reinforcement tool’ or as an ‘interpretative tool’ in order to strengthen their final decisions by showing that they have conducted extensive research on the matter before them. The latter position is very common within the Portuguese Constitutional Court, which does not directly apply a provision to a concrete case, but controls the constitutionality of norms. Admittedly, at the very beginning of Portugal’s EC membership, the courts had resisted EC law mainly due to the lack of knowledge. In fact, they decided cases without taking into consideration EEC law. Nowadays the influence of EU law on national law is such that it has become impossible to ignore it. Therefore, the judges overcame the initial resistance and they now usually take EU law into account. It must be underlined that the Charter on Fundamental Rights of the European Union (CFR) has been quoted by the Constitutional Court even before it had entered into force.92 The situation is similar with CJEU’s case law. The interpretation and validity of EU norms as decided by the CJEU is not challenged. Portuguese Courts accept that the CJEU’s jurisprudence on the interpretation and validity of EU law is mandatory.93 The highest courts (Supreme Court 90 For more information see www.en.parlamento.pt/EuropeanAffairs/EuropInitiativesScrutiny.html. 91 For a comprehensive study see Francisco Pereira Coutinho, Os tribunais nacionais na ordem jurídica da União Europeia – o caso português (Coimbra, Coimbra Editora, 2013) 331 ff. 92 See, among other decisions, TCPT 03.03.2001, Acórdão 90/2001; and TCPT 12.12.2003, Acórdão 614/2003. 93 For further developments see Pereira Coutinho, Os tribunais nacionais (n 91) 399 ff.

Portugal  647 of Justice and Supreme Administrative Court) and the Constitutional Court – even when they do not interpret or apply EU law – usually refer to it in order to reinforce the argumentation and motivation of their decisions. The Constitutional Court in particular increased considerably its EU law references in general and to the CFR in particular in the last decade. This increase has been due to many factors, such as the enlargement of the material scope of EU law, the increased awareness of the judges relating to EU law and, last but not least, the composition of the Court, which is much more ‘EU law-friendly’ than in the past.94 However, the Constitutional Court has never considered EU law an autonomous constitutional parameter.

VI.  Relationship between EU Law and National Law A.  Supremacy/Primacy of EU Law Over Domestic Law, Including Constitutional Law First of all, it must be recalled that – in contrast to other Member States – neither case law nor doctrine have so far explored the potential difference between supremacy and primacy of EU law over national law in Portugal. Until the constitutional amendment of 2004, the CPR did not provide any criteria to solve potential conflicts between domestic and primary or secondary EU law. However, some scholars accepted the primacy of the EU law even over the CPR.95 In a strict constitutional perspective, it was rather difficult to accept the primacy over the CPR. It could only be inferred from the treaties and from the jurisprudence of the CJEU. After the constitutional revision of 2004, Article 8(4) CPR recognises that the solution of conflicts between national and primary and secondary EU law adopted by EU institutions within their competences is a matter of EU law. Therefore, EU law prevails over every single internal rule, including constitutional ones,96 unless, according to Article 8(4) CPR, it violates the fundamental principles of a democratic state based on the rule of law. From a Portuguese perspective this rationale does not only result from EU law, but also from a constituent decision indirectly taken by Portuguese citizens, through their representatives. Nevertheless, some scholars97 still sustain that the CPR prevails over the EU law. In our opinion,98 the CPR does not leave any margin for doubt. EU law prevails over the domestic legal order, including the CPR, except if it violates the fundamental principles of a democratic state based on the rule of law. Taking into account that these principles also bind the EU through Article 2 TEU, the adoption of an EU measure violating these limits will be extremely rare, as the Constitutional Court has recently accepted.99 However, even for such rare cases, one has to identify the competent jurisdiction to deal with this issue. Under Portuguese law, the Constitutional Court and the national courts in general are the guardian(s) of the principles of a democratic state based on the rule of law (Article 204 CPR). Under EU law, however, the CJEU is the guardian of EU law. Taking into consideration the 94 For more details see Ana Maria Guerra Martins and Miguel Prata Roque, ‘Judicial Dialogue in a Multilevel Constitutional Network – the Role of the Portuguese Constitutional Court’, in M Andenas and D Fairgrieve (eds), Courts and Comparative Law (Oxford, OUP, 2015) 300. 95 See André Gonçalves Pereira and Fausto de Quadros, Manual de Direito Internacional Público, 3rd edn (Coimbra, Almedina, 1993) 124 ff. 96 Otero, Identidade Constitucional (n 18) 133; Gomes Canotilho/Moreira, ‘Article 8’ (n 56) 265. 97 Blanco de Morais, ‘Sindicabilidade’ (n 61) 242 ff. 98 Guerra Martins, Manual (n 76) 511 ff; Guerra Martins, ‘Portuguese Perspective’ (n 49); Guerra Martins/Roque, ‘Chapter 18’ (n 49) 310 ff. 99 TCPT 15.07.2020, Acórdão 422/20.

648  Ana Maria Guerra Martins fact that there is no judicial hierarchy between the national courts and the CJEU, but a system of cooperation, the principle of supremacy of EU law over domestic law shall be invoked. As a consequence, EU law shall not be declared unconstitutional and national courts shall not apply domestic law that contravenes EU law. The Portuguese Constitutional Court, in a rather long and well-reasoned ruling, decided that, under the terms of Article 8(4) of the Constitution, it can only assess and refuse to apply a EU rule, if the rule is incompatible with a fundamental principle of a democratic state based on the rule of law, which, within the scope of the EU law – including, therefore, the jurisprudence of the CJEU – does not enjoy a status materially equivalent to that recognised in the Constitution. On the contrary, whenever the assessment of a EU norm is at stake in the light of a (fundamental) principle of the democratic rule of law that, within the European Union law, does enjoy a status materially equivalent to that recognised in the Portuguese Constitution, functionally guaranteed by the CJEU, the Constitutional Court refrains from assessing the compatibility of that rule with the Constitution.100 In our view, the case law of the Portuguese Constitutional Court is in line with the case law of other European Constitutional Courts, first and foremost, the German and the Italian Constitutional Courts, which are referred to in the ruling. Summing up, in the current stage of evolution of the relationship between EU law and domestic law, the solution for such a controversial question can solely be found in the cooperation between the CJEU and domestic courts. Thus, in case of doubt on the interpretation or validity of EU law, the latter (including the Constitutional Court) may or shall request a preliminary ruling from the CJEU, under the terms of Article 267 TFEU. However, the supremacy of EU law over domestic law does not lead to the invalidity of the national rule, but only to its inapplicability.101 Actually, no mechanism exists which would automatically repeal conflicting national law.

B.  Judicial Dialogue between the Portuguese Constitutional Court and the CJEU The absence of judicial hierarchy between the CJEU and national courts requires the development of judicial dialogue between them – with special emphasis to the dialogue between the CJEU and Constitutional Courts – which may be exercised by several means. The most important one is the preliminary ruling procedure. This dialogue is also relevant in Portugal.102 The Portuguese Constitutional Court has very recently referred its first preliminary question to the CJEU.103 However, it has ever been aware of the need to conduct a judicial dialogue with the CJEU. First, the Constitutional Court has acknowledged its competence to make a preliminary reference to the CJEU, since Ruling No 163/1990 of 23 May 1990104 and reaffirmed this position, more 100 Ibid. 101 Otero, Identidade Constitucional (n 18) 133 f; Gomes Canotilho/Moreira, ‘Article 8’ (n 56) 266. 102 For further developments see Guerra Martins/Prata Roque, ‘Judicial Dialogue’ (n 96) 314 ff; Catarina Sarmento e Castro and Filipa Vicente Silva, ‘The Portuguese Constitutional Court’, in VfGH (ed), XVIe Congrès de la Conférence des Cours Constitutionnelles Européennes, Vol II: La coopération entre les Cours constitutionnelles en Europe: situation actuelle et perspectives (Vienna, Verlag Österreich, 2014) 763. 103 TCPT 09.12.2020, Acórdão 711/20. By contrast, the Portuguese Supreme Court of Justice and the Portuguese Administrative Supreme Court have already referred to the CJEU. According to some studies the Administrative Supreme Court is the highest Portuguese court using the preliminary reference procedure more frequently. In this direction see Pereira Coutinho, Os tribunais nacionais (n 91) 369 ff. 104 TCPT 23.05.1990, Acórdão 163/1990 (English summary available).

Portugal  649 recently, among others, in Ruling No 391/2012 of 9 August 2012.105 Secondly, the Court has discussed the possibility to request a preliminary ruling from the CJEU many times. However, it has always concluded that it was not needed mainly based on three reasons: • the lack of an EU law question of interpretation;106 • no need for clarifying the EU law question for solving the constitutionality/illegality problem;107 and • a decision of the CJEU on the preliminary question is not necessary to enable the Constitutional Court to give its judgment.108 More recently, in its Ruling No 141/2015 of 3 March 2015,109 the Court discussed whether, in order to scrutinise the constitutionality of a domestic rule that required one year of legal residence in Portugal for EU citizens as a prerequisite to benefit from the social insertion income (a social benefit), it was necessary to request a preliminary ruling from the CJEU, since the right to move and reside freely within the Portuguese territory of the EU citizens and their families could be jeopardised. The Court considered that, in light of Directive 2004/38/EC and the case law of the CJEU, especially the judgment in Dano,110 there was no doubt that EU law tolerates the national provision and that, consequently, there was no need to request the preliminary ruling. Another way of conducting judicial dialogue is the strong commitment of the Constitutional Court to openness vis-à-vis binding EU law. As already mentioned, the Constitutional Court uses EU law, particularly the CFR as a ‘reinforcement tool’ or as an ‘interpretative tool’. Actually, the Constitutional Court’s decisions frequently quote CJEU case law.111

C.  Case Law of the Portuguese Constitutional Court During the Sovereign Debt Crisis The impact of EU law and international law on the jurisprudence of the Constitutional Court is particularly visible in the so-called ‘case law of the crisis’. The financial crisis in general and the sovereign debt crisis in particular (2010–14) led the Portuguese legislature to adopt a package of austerity measures. Some of them were challenged before the Constitutional Court, which had to deal, on the one hand, with the EU and international commitments of the Portuguese state and, on the other hand, with the highly protective character of the CPR concerning fundamental rights, especially social rights. Throughout this case law, the Constitutional Court accepted that the lawmaker has a wide margin of appreciation to reverse the level of protection of social rights, including the right to a fair wage, the right to a pension, and the right to a social benefit, which constitute the hard core of social rights. The Court also admitted that the control of the public debt strongly lies in the public interest, and consequently the respective EU and international commitments must be respected.

105 TCPT 09.08.2012, Acórdão 391/2012. 106 TCPT 163/1990 (n 104); TCPT 17.12.1999, Acórdão 658/1999; TCPT 21.12.2004, Acórdão 717/2004; TCPT 08.03.2007, Acórdão 181/2007; TCPT 02.05.2007, Acórdão 273/2007. 107 TCPT 22.11.1994, Acórdão 606/1994 (English summary available); TCPT 16.05.2000, Acórdão 278/2000; TCPT 391/2012 (n 105). 108 TCPT 11.04.2000, Acórdão 240/2000. 109 TCPT 03.03.2015, Acórdão 141/2015 (English summary available). 110 CJEU 11.11.2014 C-333/13 (Dano) ECLI:EU:C:2014:2358. 111 Guerra Martins/Prata Roque, ‘Judicial Dialogue’ (n 96) 314 ff.

650  Ana Maria Guerra Martins However, the reduction of social rights protection must respect the principles of equality, proportionality, and legal certainty comprised in the rule of law. Insofar as the Court considered that these principles were violated, it did not hesitate to declare the national legislation unconstitutional, despite of the government’s argument that the provisions in question were crucial to fight the economic and financial crisis.112 Initially, austerity measures were restricted to fiscal issues. However, during the course of the financial and public debt crisis, and specifically after the request for external assistance and the signature of the Financial Assistance Programme (FAP) for Portugal113 in May 2011,114 they were extended to several matters such as cuts in staff, in wages and other benefits for the public sector, cuts in pensions for the public and private sector, ‘flexibility’ of the labour market, and reduction of unemployment benefits for private sector. It must be pointed out that, as a rule, these cuts did not touch the lower levels of wages, pensions and benefits. Some of these measures were integrated into the State Budget Laws of 2011, 2012, 2013 and 2014. All of them were challenged before the Constitutional Court. Other measures were formally independent from the State Budget Laws, but closely connected to them. A third group of measures was completely autonomous of the State Budget Laws. The Constitutional Court, in several proceedings, declared the unconstitutionality of some of these measures. All of them were enacted in order to fulfil obligations which stemmed from EU law and other closely related legal constraints, such as the Memoranda of Understanding (MoU) in the course of the restructuring of the Portuguese public debt.115 In all of these cases the Constitutional Court demonstrated full awareness of the Portuguese international and EU obligations and took them into consideration in all decisions. It is true that those obligations required Portugal to adopt the specified austerity measures they set out, as one of the conditions for the phased fulfilment of the convened financing obligations. However, these obligations leave a certain margin of appreciation to the Portuguese legislator. For example, the Memoranda imposed the goal to reduce a certain percentage of the public expenditure concerning civil servants or health services. However, they did not explicitly indicate the concrete measures that the government had to take in order to achieve this goal. Cutting wages, pensions or social benefits, or intervening in the public private partnerships are certainly appropriate means to achieve the delineated objectives. In this vein, the Constitutional Court has neither questioned the detailed goals of the Programme nor the obligations of the Portuguese state to achieve them. Moreover, it accepted

112 Opposing the ‘case law of the crisis’: Gonçalo de Almeida Ribeiro and Luís Pereira Coutinho, O Tribunal Constitucional e a crise – Ensaios críticos (Coimbra, Almedina, 2014). Replying to criticism Jorge Reis Novais, Em Defesa do Tribunal Constitucional – Resposta aos críticos (Coimbra, Almedina, 2014). 113 This Programme comprises a set of legal instruments. On the one hand, the Portuguese Government and the IMF signed a Technical MoU and a Memorandum of Economic and Financial Policies, which set out the terms and conditions governing the provision of financial assistance to Portugal by the IMF. On the other hand, the Portuguese Government and the EU signed a MoU on Specific Economic Policy Conditionality. These Memoranda require the Portuguese State to adopt the measures they set out, as one of the conditions for the phased fulfilment of the financing contracts entered into force by the same parties. 114 The FAP for Portugal and all the documents related to are available on the website of Banco de Portugal (www.bportugal.pt). 115 For closer scrutiny, compare Ana Maria Guerra Martins, ‘Constitutional Judge, Social Rights and Public Debt Crisis: The Portuguese Constitutional Case Law, 22 Maastricht Journal of European and Comparative Law (2015) 678. Some authors sustained that the TCPT should have controlled the conformity with the CPR of the Memoranda, as a guardian of the Constitution. Cf José de Melo Alexandrino, ‘Jurisprudência da Crise. Das Questões Prévias às Perplexidades’, in G A Ribeiro and L P Coutinho (eds), O Tribunal Constitucional e a crise – ensaios críticos (Coimbra, Almedina, 2014) 62, 63. On the legal nature of the Memoranda cf Francisco Pereira Coutinho, ‘A natureza jurídica dos memorandos da “troika”’, XIII Themis (2013) no 24/25, 147.

Portugal  651 the control of public debt as a relevant public interest that justified some restrictive measures.116 The Court, however, controlled the constitutionality of the specific measures that were taken, given that the international law obligations did not impose any discriminatory measures. In the ‘case law of the crisis’, the Constitutional Court actually expressly acknowledged the constraints derived from European commitments – particularly those arising out of the TFEU, the TSCG and the excessive deficit procedure.117 Accordingly, the Constitutional Court took due account of the relevant international law, which, unless it forms part of jus cogens, does not prevail over the CPR, and of pertinent EU law, which is likely to prevail over constitutional law, when the conditions set out in Article 8(4) CPR and EU law are fulfilled. However, this issue did not appear in the cases that the Constitutional Court had to decide.118 Consequently, it is fair to say that the Constitutional Court did not scrutinise obligations resulting from Portugal’s EU membership. The Constitutional Court also did not scrutinise obligations flowing from relevant international law obligations against the yardstick of the CPR, simply because it was not necessary. Declaring Portuguese legislation constitutional was done on the premise that these international obligations would allow for measures which respect the CPR requirements. Thus, ultra vires control of international obligations was never really at stake.

VII.  Concluding Remarks The present study leads us to the following conclusions: • The CPR was adopted on 2 April 1976 and has undergone seven constitutional revisions. Portugal is a sovereign Republic, based on the dignity of the human person and the will of the people. The structure of the State is unitary with autonomous regions. Furthermore, Portugal is a democratic state based on the rule of law. The political system is based on a separation of effective – positive and negative – powers between president, parliament and government, possessing identical democratic legitimacy. The judicial review is assured by courts and the control of constitutionality and legality is multi-shaped. The last final arbiter belongs to the Constitutional Court. • The CPR is the outcome of several political commitments, which are sometimes hardly reconcilable or even contradictory. The divergences between the wording of the CPR and the living constitution are numerous. Therefore the Constitutional Court has always played an active and influential role in order to concretise the constitutional norms and principles. • Portugal’s accession to the EC and the subsequent amendments of the treaties led to several constitutional revisions. Currently, the CPR contains specific provisions on EU membership, such as the ‘EU clause’, the reception of primary and secondary EU law clause, the clause on primacy of EU law over Portuguese law, including constitutional law. However, the ‘golden rule’ is not included in the CPR. • Concerning international law, the CPR comprises a principle of ‘friendship to international law’, foreseeing the reception of international customary law and treaties. The hierarchical

116 Eg TCPT 27.10.2010, Acórdão 399/2010 (English summary available); 21.09.2011, Acórdão 396/2011; 05.07.2012 Acórdão 353/2012 (English summary available); 05.04.2013 Acórdão 187/2013 (English summary available); etc. 117 Eg TCPT 14.08.2014 Acórdão 574/2014 (English summary available). 118 For a more detailed study see Guerra Martins, ‘Constitutional Judge’ (n 115).

652  Ana Maria Guerra Martins

•

•

•

•

•

position of international law within the domestic legal order is subject to debates in the Portuguese doctrine, but everyone agrees that peremptory norms of international law prevail over national law, including constitutional law. Further transfer of powers to the EU through treaty amendments is constitutionally limited. Such limitations are the principle of reciprocity, the fundamental principles of a democratic state based on the rule of law, and the principle of subsidiarity. Material limits regarding constitutional revision can also be considered as limits to further transfer of powers to the EU. As regards ultra vires acts, the CPR does not expressly specify the application of secondary EU law in the Portuguese legal order. Furthermore, neither the case law of Portuguese courts, nor Portuguese scholarship have discussed this issue in detail. The CPR also contains rules on the implementation of EU law, including on the participation of the AR in secondary EU law-making. This participation has increased in the last ten years as a result of the evolution of EU law. Moreover, the CPR regulates the transposition of EU legal acts into the internal legal order, which can take the form of a law, a decree-law, or a regional legislative decree. Finally, Portuguese courts, as any other EU Member State courts, should interpret and apply EU law in their decisions, since they are common judges of EU law. They also can, and in certain cases must, cooperate with the CJEU through the preliminary reference procedure under Article 267 TFEU. The judicial dialogue between Portuguese courts and the CJEU also exists outside this procedure. During the public debt crisis and the Financial Assistance Programme for Portugal (2010–15), the Constitutional Court had to control some austerity measures, such as cuts in staff, wages and other benefits, and pensions that were sometimes imposed by either EU law or international law.

References C Blanco de Morais, ‘As metamorfoses do semipresidencialismo português’, (1997) 22 Revista Jurídica 141. C Blanco de Morais, ‘A sindicabilidade do direito da Unão Europeia pelo Tribunal Constitucional Português’, in J Miranda (ed), Estudos em Homenagem ao Prof. Doutor Sérvulo Correia vol I (Coimbra, Coimbra Editora, 2011) 211. C Blanco e Morais, Curso de Direito Constitucional. Teoria da Constituição em Tempo de Crise do Estado Social, vol II.2 (Coimbra, Coimbra Editora, 2014). V Canas, ‘Sistema semi-presidencial’, in Dicionário Jurídico da Administração Pública – 1º Suplemento (Coimbra, Almedina, 1998) 472. E Correia Baptista, Direito Internacional Público, vol I, Conceito e fontes (Lisbon, Lex, 1998). JJ Gomes Canotilho, Direito Constitucional e Teoria da Constituição, 7th edn (Coimbra, Almedina, 2003) JJ Gomes Canotilho and V Moreira, Fundamentos da Constituição (Coimbra, Coimbra Editora, 1991). JJ Gomes Canotilho and V Moreira, Os poderes do Presidente da República (Coimbra, Coimbra Editora, 1991). JJ Gomes Canotilho and V Moreira, Constituição da República Portuguesa Anotada, vol II, 4th edn (Coimbra, Wolters Kuwer/Coimbra Editora, 2006). JJ Gomes Canotilho and Vital Moreira, ‘Commentary to Article 8’, in JJ Gomes Canotilho and V Moreira (eds), CRP – Constituição da República Portuguesa – anotada, vol 1 (Coimbra, Coimbra Editora, 2014) 258. A Gonçalves Pereira and F de Quadros, Manual de Direito Internacional Público, 3rd edn (Coimbra, Almedina, 1993).

Portugal  653 AM Guerra Martins, ‘La Constitution portugaise: Une Constitution inflationniste’, in G Amato et al (eds), The Constitutional Revision in Today’s Europe (London, Esperia Publications, 2002) 129. AM Guerra Martins, ‘The Portuguese Semi-Presidential System – About Law In the Books and Law In Action’, (2006) 2 European Constitutional Law Review, 81. AM Guerra Martins, ‘A Portuguese Perspective of the Accession on the European Union to the European Convention of Human Rights’, in I Iliopoulos-Strangas, V Pereira da Silva and M Potacs (eds), The Accession of the European Union to the ECHR (Baden-Baden, Nomos, 2013) 201. AM Guerra Martins, ‘Constitutional Judge, Social Rights and Public Debt Crisis: The Portuguese Constitutional Case Law’, (2015) 22 Maastricht Journal of European and Comparative Law 678. AM Guerra Martins, Manual de Direito da União Europeia, 2nd edn (Lisbon, Almedina, 2017). AM Guerra Martins and MP Roque, ‘Chapter 18 – Universality and Binding Effect of Human Rights from a Portuguese Perspective’, in R Arnold (ed), The Universalism of Human Rights (Dordrecht, Springer, 2013) 297. AM Guerra Martins and MP Roque, ‘Judicial Dialogue in a Multilevel Constitutional Network – the Role of the Portuguese Constitutional Court’, in M Andenas and D Fairgrieve (eds), Courts and Comparative Law (Oxford, OUP, 2015) 300. MI Jalles, ‘Primado do Direito Comunitário sobre o Direito nacional dos Estados membros’, Documentação e Direito Comparado Separata No 4, Lisbon 1980. MI Jalles, Implications juridico-constitutionnelles de l’adhésion aux Communautés européennes – Le cas du Portugal (Brussels, Bruylant, 1981). P Lomba, ‘Commentary to Article 108’, in P Otero (ed), Comentário à Constituição Portuguesa (Coimbra, Almedina, 2008). F Lucas Pires, Teoria da Constituição de 1976. A transição dualista (Coimbra, Edition of the author, 1988). R Medeiros, A Constituição Portuguesa num Contexto Global (Lisbon, UCP, 2015). J Melo Alexandrino, ‘Jurisprudência da Crise. Das Questões Prévias às Perplexidades’, in G A Ribeiro and L P Coutinho (eds), O Tribunal Constitucional e a crise – ensaios críticos (Coimbra, Almedina, 2014) 62. J Miranda, ‘O Tratado de Maastricht e a Constituição Portuguesa’, in P de Pitta e Cunha (ed), A União Europeia na encruzilhada (Coimbra, Almedina, 1996) 45. J Miranda, ‘Decisões políticas: aprovação, abstenção e rejeição no momento constituinte de 1976’, (2000) 46 Studia Juridica. Universidade de Coimbra. Boletim da Faculdade de Direito 177. J Miranda, ‘Acabar com o frenesim constitucional’, in AAFDL (ed), Nos 25 anos da Constituição da República Portuguesa de 1976 (Lisbon, AAFDL, 2001) 5. J Miranda, Curso de Direito Internacional Público, 4th edn (Cascais, Principia, 2009). J Miranda, ‘Artigo 8’, in J Miranda and R Medeiros, Constituição Portuguesa Anotada, vol I, 2nd edn (Coimbra, Coimbra Editora, 2010). J Miranda, Manual de Direito Constitucional, vol III, 6th edn (Coimbra, Coimbra Editora, 2010). J Miranda, Manual de Direito Constitucional, vol II, 7th edn (Coimbra, Coimbra Editora, 2013). J Miranda, Manual de Direito Constitucional, vol VI, 4th edn (Coimbra, Coimbra Editora, 2013). J Miranda, Manual de Direito Constitucional, vol I.2, 10th edn (Coimbra, Coimbra Editora, 2014). J Miranda, Da Revolução à Constituição (Cascais, Principia, 2015). J Miranda and R Medeiros, Constituição Portuguesa Anotada, vol III (Coimbra, Coimbra Editora, 2007). A Moreira, ‘O regime: presidencialismo do Primeiro-Ministro’, in M Baptista Coelho (ed), Portugal – O Sistema Político e Constitucional 1974/87 (Lisbon, ICSUL, 1989) 31. V Moreira, ‘A CRP e a União Europeia’, in Estudos em homenagem ao Conselheiro Presidente Rui Moura Ramos, vol I (Coimbra, Almedina, 2016) 869. P Otero, O poder de substituição em Direito Administrativo, vol II (Lisboa, Lex, 1995). P Otero, Direito Constitucional Português. Identidade Constitucional, vol I (Coimbra, Almedina, 2010). P Otero, Direito Constitucional Português. Organização do Poder Político, vol II (Coimbra, Almedina, 2010). F Pereira Coutinho, ‘A natureza jurídica dos memorandos da “troika”’, (2013) XIII (24/25) Themis 147. F Pereira Coutinho, Os tribunais nacionais na ordem jurídica da União Europeia – o caso português (Coimbra, Coimbra Editora, 2013).

654  Ana Maria Guerra Martins C Queiroz, ‘A classificação das formas de governo. Em particular, o sistema de governo “semi-presidencial”’, in J Miranda (coord), Estudos em homenagem ao Prof. Doutor Armando Marques Guedes (Lisbon, Almedina, 2004) 321. M Rebelo de Sousa, O Sistema de Governo Português, 4th edn (Lisbon, AAFDL, 1992). J Reis Novais, Semipresidencialismo. Teoria do sistema de governos semipresidencial, vol I (Coimbra, Almedina, 2007). J Reis Novais, Em Defesa do Tribunal Constitucional – Resposta aos críticos (Coimbra, Almedina, 2014). GA Ribeiro and LP Coutinho, O Tribunal Constitucional e a crise – Ensaios críticos (Coimbra, Almedina, 2014). C Sarmento e Castro and F Vicente Silva, ‘The Portuguese Constitutional Court’, in VfGH (ed), XVIe Congrès de la Conférence des Cours Constitutionnelles Européennes, Vol II: La coopération entre les Cours constitutionnelles en Europe: situation actuelle et perspectives (Vienna, Verlag Österreich, 2014) 763. A Sousa Pinheiro, ‘Commentary to Article 109’, Pedro Lomba, ‘Commentary to Article 108’, in P Otero (ed), Comentário à Constituição Portuguesa (Coimbra, Almedina, 2008). A Sousa Pinheiro and MJ de Brito Fernandes, Comentário à IV Revisão Constitucional (Lisboa, AAFDL, 1999). A Sousa Pinheiro and P Lomba, ‘Commentary to Article 111’, in P Otero (ed), Comentário à Constituição Portuguesa (Coimbra, Almedina, 2008). A Vitorino, ‘A adesão de Portugal às Comunidades Europeias’, (1984) 3 Estudos de Direito Público 9.

23 Romania MIHAELA VRABIE

I.  Main Characteristics of the National Constitutional System The Romanian revolution of December 1989 created a favourable historical context for the adoption of the current Constitution of Romania (henceforth CRO, or the Constitution). Following the revolution, all power structures of the dictatorial communist regime in Romania were removed and a transition to a democratic regime based on the rule of law initiated. The abolition of the communist regime led also to the removal of all the constitutional provisions which were in force during the communist regime. The historical moment of December 1989 called for a new Constitution that would provide the legal framework to allow the evolution of Romania towards the rule of law, democracy and fundamental values. The pre-communist regime Constitutions of 1866 and 1923 were closely related to the monarchy. As such, they could not be used as a valid source of inspiration, because a monarchy was not a realistic goal after the fall of the dictatorial communist regime. Thus, the post-communist Constitution of Romania broke away from the pre-communist constitutional traditions of the country. The current Constitution1 was adopted in 1991, after a national referendum which took place on 8 December 1991. The project of the 1991 Constitution was adopted by the Parliament elected in May 1990, which played, first, the role of a constituent assembly and subsequently, the role of a legislative body.2 The 1991 Constitution symbolises the rebirth of democracy in Romania after almost 50 years of a dictatorial communist regime. The main traits of the national constitutional system and the most important constitutional principles, which were newly included in the current Constitution, are: the republican form of government, the separation of powers, the rule of law, political pluralism, the bicameral structure of Parliament, democracy and freedom, human dignity, inviolability and inalienability of fundamental rights, and the principle of consulting the people through a national referendum regarding constitutional amendments and measures of significant importance for the national interest.3 The 1991 Constitution also created new institutions, such as the Constitutional Court of Romania (CCRO), which is entrusted with the control of the constitutionality of the laws.4 1 The Constitution of Romania is available in English at www.presidency.ro/en/the-constitution-of-romania. 2 Ștefan Deaconu, Drept constituțional [Constitutional law] (Bucharest, CH Beck, 2011) 134. 3 Ioan Muraru and Elena Simina Tănăsescu, Drept constituțional și instituții politice [Constitutional law and political institution], 14th edn, vol I (Bucharest, CH Beck, 2011) 99. 4 Deaconu, Drept constituțional (n 2) 135.

656  Mihaela Vrabie According to Article 1 CRO, Romania is a sovereign, independent, unitary and democratic state, governed by the rule of law, having justice and fundamental rights as supreme values, based on the principle of separation of powers. The principle of separation of powers was included following the 2003 amendment of the Constitution. The amended Article 1(4) of the Constitution clearly provides for the separation of powers into legislative, executive, and judicial. However, the principle of separation of powers must be read in direct connection with the requirement of balance of powers, which must be achieved in the framework of a constitutional democracy.5 Article 2 of the Constitution enshrines the theory of national sovereignty. It provides that national sovereignty belongs to the Romanian people, who shall exercise it by means of their representative bodies resulting from free, periodical and fair elections, as well as through referenda. The 1991 Constitution opted for a combination between the legal regime of national sovereignty, expressed by the concept of representative democracy, which implies the delegation of the exercise of the sovereignty to the Parliament as supreme representative body of the Romanian people, and the legal regime of popular sovereignty, expressed by the concept of direct democracy, which implies the direct manifestation of power by the people through referenda.6 The constitutional system established by the 1991 Constitution, as amended in 2003, creates a semi-presidential political framework whereby the responsibilities of the Parliament, of the President and of the Government are connected and interrelated. The legislative power – the Parliament – is bicameral, formed by the Chamber of Deputies and the Senate, which are directly elected for four-year terms. According to Article 61 of the Constitution, the Parliament is the supreme representative body of the Romanian people and the sole legislative authority of the country. This provision establishes the competence of the Parliament to adopt laws with superior legal force over the regulatory acts issued by the executive power, having a high degree of generality. Thus, a distinction is made between the legislative activity, which belongs only to the Parliament, and the rest of the normative activity of the state, which falls within the competence of the executive power.7 The executive power of Romania is dualist, divided between the President and the Government. The President is directly elected by the people for a five-year mandate and the Prime Minister is designated by the President. The Constitution enshrines a set of mechanisms for parliamentary control over the Government. The Prime Minister and the Government can be invested by the President only after receiving the Parliament’s vote of confidence regarding the governmental programme and regarding the composition of the Government.8 In a similar way, the Parliament can withdraw its confidence in the Government by adopting a censorship motion, with a majority vote of the deputies and the senators.9 If the censorship motion is adopted, the Prime Minister and the Government are removed from office. Articles 80 and 92 of the Constitution establish the role and the competences of the President of Romania to be the following: (i) representing the Romanian state; (ii) safeguarding national independence, unity and territorial integrity of the country; (iii) supervising the observance of the Constitution and the proper functioning of the public authorities; (iv) mediating between the powers in the state, as well as between the state and the society; (v) being the commander-in-chief of the military forces.

5 Art 1(4) CRO: ‘The State shall be organised based on the principle of the separation and balance of powers – legislative, executive, and judicial – within the framework of constitutional democracy.’ 6 Ioan Muraru, Elena Simina Tănăsescu, Constituția României – Comentariu pe articole [The Romanian Constitution annotated] (Bucharest, CH Beck, 2019) 523. 7 Ibid, 525. 8 Art 85 CRO. 9 Art 113 CRO.

Romania  657 Chapter VI CRO, concerning judicial authority, includes courts of law, public ministry and the Superior Council of Magistracy, although only the courts of law are part of the judiciary.10 Article 126(1) CRO provides that justice shall be administered by the High Court of Cassation and Justice, and the other courts of law set up by the law. The judiciary comprises a hierarchically ordered system of courts, having different subject-matter jurisdictions and procedures established by law, with the High Court of Cassation and Justice as a supreme court. Article 133 CRO provides for a specific body to secure the judiciary’s independence, the Superior Council of Magistracy. The Council is composed of 19 members, of which nine shall be judges and five shall be public prosecutors. The Superior Council of Magistracy proposes to the President of Romania appointments of judges and public prosecutors. The Council is also in charge of inspections and deciding on the disciplinary liability of judges and public prosecutors. The most important constitutional principles that govern judicial power and the administration of justice in Romania are: legality,11 impartiality and equality,12 presumption of innocence13 and independence of judges.14 The CCRO is entrusted with interpretation of the Constitution and the constitutional control of the national laws.15 It has the power to adjudicate on the constitutionality of laws adopted by the Parliament, before their promulgation. It can do so following an objection of unconstitutionality raised by the President of Romania, by one of the Presidents of the two parliamentary chambers, by the Government, by the High Court of Cassation and Justice, by the Romanian Ombudsman, or by a group of at least 50 deputies or at least 25 senators, as well as ex officio, on constitutional amendments. It may also decide on the constitutionality of laws and government ordinances after they have entered into force. It can take such decisions following an objection of unconstitutionality raised before a court of law or commercial arbitration tribunals, as well as ex officio, by the courts, or raised directly by the Romanian Ombudsman. It adjudicates on the constitutionality of treaties or other international agreements (upon referral by one of the Presidents of the two chambers of parliament, a group of at least 50 deputies or at least 25 senators) and on the constitutionality of the standing orders of Parliament.16 Also, the CCRO controls the system of separation of powers between the executive, legislative and judicial powers. Unfortunately, individual direct complaints regarding violation of fundamental rights cannot be brought before the CCRO. These actions can only be brought before the ordinary courts which are part of the judiciary. The Constitution enshrines a broad catalogue of fundamental rights in Title II, including civil, political, economic and social rights. They are included in Chapters I and II of Title II. Article 20 of the Constitution recognises the importance of international human rights law and provides that constitutional provisions concerning the citizens’ rights and liberties need to be interpreted and enforced according to the Universal Declaration of Human Rights and other human rights treaties ratified by Romania. In practice, the CCRO has frequently referred to Article 20 CRO, to the European Convention on Human Rights (ECHR) and to the corresponding case law of the European Court of Human Rights (ECtHR) as sources of interpretation for the constitutional rights. 10 Ioan Muraru, Elena Simina Tănăsescu, Drept constituțional și instituții politice [Constitutional law and political institutions], 13th edn, vol II, (Bucharest, CH Beck, 2009) 283. 11 Art 124(1) CRO. 12 Art 124(2) CRO. 13 Art 23(11) CRO. 14 Art 124(3) CRO. 15 Articles 142–47 CRO and Law no 47/1992, republished in Monitorul Oficial [Official Journal of Romania] no 807/03.12.2010. 16 Art 146 CRO.

658  Mihaela Vrabie Finally, the Constitution is not very flexible when it comes to amendments. It outlines a rigid procedure, requiring a broad consensus. According to Article 150, amendment of the Constitution may be initiated by the President of Romania on a proposal of the Government, or by at least one quarter of the deputies or the senators, as well as by at least 500,000 citizens with the right to vote. Then, the project of the amendment must be adopted by the Chamber of Deputies and the Senate, by a majority of at least two-thirds of the members of each chamber. A constitutional amendment shall be final only after approval by a referendum.17 The Constitution has been amended only once so far, in 2003, regarding accession to the European Union (EU) and to the North Atlantic Treaty Organisation (NATO). The 2003 amendments entrenched the autonomy of certain key institutions – the CCRO, the Superior Council of Magistracy and the Ombudsman acquired more prominent roles within the constitutional system.18 Since this last amendment of the Constitution, there have been four different constitutional revision initiatives (in 2011, 2014, 2016 and 2019).19 However none of them was adopted.

II.  Constitutional Culture Romanian constitutional history has undergone many significant changes and different stages. This is due to different historical influences that have marked Romanian constitutional culture. During the historical period corresponding to the forming of the modern Romanian state (1860–1940), the most important influences came from the Constitution of Belgium of 1831 and from the legal system of France, while during the communist regime (1948–89), the Romanian constitutional system was influenced by the socialist constitutions.20 The current Constitution was a turning point for Romania because it marked the abolition of the dictatorial communist regime and laid down the legal foundations of a democratic state in which fundamental rights are guaranteed. According to Article 1(5) CRO the observance of the Constitution and its supremacy is mandatory. The CCRO is entrusted with a very important role. According to Article 142(1) CRO, it guarantees the supremacy of the Constitution. It is a political and jurisdictional public authority that is independent from the judiciary and independent from any other public authority.21 It is composed of nine judges appointed for a nine-year term of office: three appointed by the Chamber of Deputies, three by the Senate, and three by the President of Romania.22 The CCRO judges elect by secret ballot the Court President for a three-year term.23 The 2003 amendment of the Constitution increased the jurisdictional powers of the CCRO, introducing, inter alia, the possibility of solving constitutional questions at the request of the President of Romania, of the presidents of the two chambers of Parliament, of the Prime Minister, or of the President of the Superior Council of Magistracy,24 or the possibility of the Romanian

17 Art 151 CRO. 18 Bogdan Iancu, ‘Romania – The Vagaries of International Grafts on Unsettled Constitutions’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (The Hague, TMC Asser Press, 2019) 1050. 19 Available at: www.cdep.ro/pls/proiecte/upl_pck2015.lista?cam=2 and www.senat.ro/legis/lista.aspx#ListaDocumente. 20 Deaconu, Drept constituțional (n 2) 111, 123. 21 The CCRO is not included in Ch VI (‘Judicial authority’), Title III CRO, governing the judiciary system, but in a separate section (Title V CRO,‘Constitutional Court’). See Muraru/Tănăsescu, Drept constituțional, 13/vol II (n 10) 261. 22 Art 142(2)–(3) CRO. 23 Art 142(4) CRO. 24 Art 146(e) CRO.

Romania  659 Ombudsman initiating constitutional review by raising an objection of unconstitutionality directly before the CCRO.25 The constitutional framework and the jurisdictional powers of the CCRO provide the legal foundation for mutual control between the different state powers, in order to guarantee the rule of law and respect for the fundamental rights of the citizens. In Romania, constitutional conflicts between the Parliament and the Government,26 the President and the Parliament, or between the President and the Prime Minister,27 have led even to changes in government. Furthermore, between 2017 and 2019, many constitutional conflicts have occurred even between the judicial power, on one hand, and the legislative or executive powers, on the other. Eventually, the CCRO has resolved most of these constitutional conflicts. In these conflicts it was called to act as a ‘constitutional arbiter’ to rule on 20 claims concerning constitutional conflicts from 2014 to 2019,28 13 of which concerned conflicts between the judicial power and the legislative/executive powers. The most critical recent developments regarding the significance of the Constitution in the legal system and the independence of the CCRO relate to certain controversial decisions of the latter. These decisions had a major impact on the activity of the judiciary.29 One of the most recent such decisions of the Constitutional Court was rendered in July 2019.30 As a result of this decision, many judgments delivered by the High Court of Cassation and Justice in criminal matters (corruption offences) will be set aside and the cases will be tried de novo. Some of the constitutional judges in that case issued a separate opinion and considered that the Constitutional Court intervened arbitrarily and without jurisdiction in the activity of the judiciary, thus breaching the principle of separation of powers and affecting judicial independence.31 Another intriguing aspect that can be observed in the recent activity of the Constitutional Court is the inconsistency in interpreting and applying the provisions of the Constitution. For example, in decision no 2/201232 the Constitutional Court decided that as an EU member, Romania must apply the mechanism for cooperation and verification (‘MCV’)33 and follow the recommendations established in this legal framework. However, in March 2019, with decision no 137/2019,34 the same court ruled that the MCV reports are not binding, having the legal status of mere recommendations. These recent decisions of the Constitutional Court have raised questions

25 Art 146(d) CRO. 26 Decisions of the Constitutional Court of Romania are available at www.ccr.ro. 27 A recent example: CCRO 18.09.2019, Decision no 504, Monitorul Oficial no 801/03.10.2019. 28 CCRO decisions are available at www.ccr.ro. 29 CCRO 16.02.2016, Decision no 51 (Monitorul Oficial no 190/14.03.2016) restricted the possibility to carry out technical surveillance measures in criminal investigations by authorities outside the prosecution or police services. CCRO 18.01.2018, Decision no 22 (Monitorul Oficial no 177/26.02.2018) led to the cancellation of evidence ordered by prosecutors but enforced by the security services in criminal investigations. CCRO 07.11.2018, Decision no 685 (Monitorul Oficial no 1021/29.11.2018) confirmed the existence of a constitutional conflict between parliament and the High Court of Cassation and Justice, generated by the fact that only four of the five members of the panels of five judges were appointed by lot. 30 CCRO 03.07.2019, Decision no 417, Monitorul Oficial no 825/10.10.2019. Following this decision, the High Court of Cassation and Justice decided to submit questions to the CJEU for a preliminary ruling, questioning the decision of the Constitutional Court under EU law. See also preliminary references submitted by the Romanian High Court of Cassation and Justice to CJEU in Cases C-357/19, C-547/19. 31 See the dissenting opinion of constitutional judges Stanciu and Tănăsescu, from decision no 417 (n 30). 32 CCRO 11.01.2012, Decision no 2/11.01.2012, Monitorul Oficial no 131/23.02.2012. 33 The MCV is to address specific benchmarks in the areas of judicial reform and the fight against corruption and it was established by European Commission Decision no 2006/928/EC of 13 December 2006, on the accession of Romania to the European Union in 2007. 34 CCRO 13.03.2019, Decision no 137, Monitorul Oficial no 295/17.04.2019.

660  Mihaela Vrabie regarding its credibility and regarding the effectiveness of the constitutional system in ensuring the supremacy of the Constitution, the rule of law, the separation of powers and the independence of the judiciary.35 These recent developments reopened the constitutional debate around the independence of the constitutional judges.

III.  Constitutional Foundations of EU Membership The Accession Treaty of Romania to the EU was signed on 25 April 2005 and was ratified unanimously by the Romanian Parliament on 17 May 2005, by Law no 157/2005.36 The ratification was not preceded by a referendum, because the 2003 amendment allowed ratification of the Accession Treaty to the EU by Parliament alone. Public opinion was largely in favour of the accession of Romania to the EU. Thus, organising a referendum on the accession to the EU was not considered at that time.37 The constitutional foundations for EU integration were included in the Constitution by Law no 429/200338 on the amendment of the Constitution, which was adopted in order to prepare the accession of Romania to the EU. The 2003 amendment of the Constitution introduced Title VI, ‘Euro-Atlantic Integration’. It consists of Article 148 (Integration into the European Union), which lays down the constitutional foundations of the Romanian EU membership, and Article 149 (Accession to the North Atlantic Treaty). Article 148 CRO provides that: (1) Romania’s accession to the constituent treaties of the European Union, with a view to transferring certain powers to Community institutions, as well as to exercising in common with the other member states the abilities stipulated in such treaties, shall be carried out by means of a law adopted in the joint sitting of the Chamber of Deputies and the Senate, with a majority of two thirds of the number of deputies and senators. (2) As a result of the accession, the provisions of the constituent treaties of the European Union, as well as the other mandatory Community provisions shall take precedence over the contrary provisions of national law, in compliance with the provisions of the accession act. (3) The provisions of paragraphs (1) and (2) shall also apply accordingly for the accession to the acts amending the constituent treaties of the European Union. (4) The Parliament, the President of Romania, the Government, and the judicial authority shall guarantee that the obligations resulting from the accession act and from the provisions of paragraph (2) are implemented. (5) The Government shall send to the two Chambers of the Parliament the projects of the mandatory acts before they are submitted to the European Union institutions for approval.

Article 148(2) CRO establishes the primacy of EU law over national law, but in an original view. Only ‘the constitutive treaties’ and the other binding provisions of EU law have primacy over national law, and only in relation to conflicting provisions of national law. The ‘national law’ used by Article 148(2) of the Constitution, that must be set aside where it conflicts with EU

35 See the dissenting opinion of constitutional judges Stanciu and Tănăsescu, from decision no 417 (n 30). 36 Monitorul Oficial no 465/01.06.2005. The legislative procedure is available at ww.cdep. ro/pls/proiecte/upl_pck. proiect?idp=6308. 37 President Traian Basescu, declaration of April 2005 (www.9am.ro/stiri-revista-presei/Politica/9602/BasescuReferendumul-privind-ratificarea-Tratatului-de-arderare-la-UE-pierdere-de-timp.html). 38 Monitorul Oficial no 758/29.10.2003.

Romania  661 law, includes the larger category of all normative acts issued by the state authorities, not only the laws adopted by parliament. Regardless of the nuances added by the Constitution to the principle of primacy of EU law over national law, its application is, according to Article 148(4) CRO, incumbent on all state authorities, even the judiciary. Even though the Constitution pays special attention to the primacy of EU law, this does not mean that the other effects of EU law are not recognised in Romanian law.39 In the absence of any conditions resulting from the Act of Accession regarding the primacy of EU law, the reference made by Article 148(2) of the CRO to the Act of Accession is without relevance. Another novelty introduced by Article 148(1), following the 2003 amendment, is the provision on the ratification of treaties involving the transfer of state powers to international organisations or regarding the joint exercise with other states of specific powers deriving from state sovereignty. This possibility was not recognised before 2003, when the Constitution was amended in the view of future EU membership. Before 2003, the Constitution included only general provisions regarding the ratification of treaties (Article 11), or regarding the interpretation and application of human rights treaties (Article 20), provisions which were not repealed by the 2003 amendment. According to Article 11 CRO, (i) Romania is bound by the obligations deriving from the treaties it is a party to and (ii) treaties ratified by Parliament, are part of national law. The same provision provides that, if a treaty conflicts with the Constitution, its ratification shall only take place after an amendment of the Constitution that resolves these conflicts. Thus, since 2003, the Constitution has featured two different ratification procedures: a general one, governed by Article 11, for treaties that are part of international law, and a special one, governed by Article 148, valid only for EU integration and for the ratification of EU Treaties. The two procedures differ only in relation to the parliamentary majority required for the ratification and neither of them includes a referendum as a compulsory condition for the ratification of treaties. Article 90 CRO enables the President of Romania, after consultation with Parliament, to call for a referendum on matters of national interest. Also, according to Article 151 CRO, a referendum is necessary for the amendment of the Constitution. The general ratification procedure for treaties which are not part of EU law, established by Article 11 CRO, follows the common legislative procedure, valid for organic or ordinary laws, depending on the normative content of the treaty which needs to be ratified.40 In this context, it is worth mentioning that there are three types of laws recognised by the Romanian legal system: constitutional laws, organic laws and ordinary laws. The amendment of the Constitution is made by constitutional laws, which are adopted by the Chamber of Deputies and the Senate by a majority of at least two-thirds of the members of each chamber of the Parliament and are subject to subsequent approval by referendum.41 The organic laws are adopted by an absolute majority – a majority vote of the members of each chamber of parliament. Ordinary laws are passed by a simple majority – a majority vote of the members present in each chamber of parliament.42 Under the special ratification procedure of Article 148(1) CRO, valid only in the field of EU integration, the ratification of the Accession Treaty of Romania to the EU is realised through a law adopted in the joint sitting of the Chamber of Deputies and the Senate, with a majority of two-thirds of the total number of deputies and senators.

39 Muraru/Tănăsescu, 40 Ibid,

1332. 41 Art 151 CRO. 42 Art 76 CRO.

Constituția României 2019 (n 6) 1339 f.

662  Mihaela Vrabie The law ratifying the EU Accession Treaty is atypical in the Romanian legal system. It is not a constitutional law and differs also from the organic and ordinary laws, because in terms of normative content, it is a law ratifying a treaty and not a law amending the Constitution. In terms of the required parliamentary majority for its ratification, the numerical majority is different, even though the majority required by Article 148(1) appears the same as the majority required by Article 151, which regulates the legislative procedure for the adoption of the laws amending the Constitution. The difference is that Article 148(1) requires a two-thirds majority of the total number of deputies and senators, while Article 151 refers to a two-thirds majority of the number of deputies and senators taken separately, as distinct assemblies.43 It follows, after comparing the numbers resulting from the calculation of the majorities required by the Constitution for the adoption of different types of laws (Article 76), that the law ratifying the EU Accession Treaty requires more favourable votes than the organic or ordinary laws. This leads to the conclusion that the law ratifying the Accession Treaty of Romania to the EU has greater legal force than organic or ordinary laws, and that the transfer of state powers to international organisations or regarding the joint exercise with other states of specific powers deriving from state sovereignty might have constitutional relevance.44 According to Article 148(3) CRO, the special ratification procedure, enshrined in Article 148(1) for the ratification of the EU Accession Treaty, applies also for the ratification of the treaties amending the constituent treaties of the EU45 and applies in general, for the participation of Romania in the EU, as an EU member. Thus, from a constitutional perspective, treaties are governed by Article 11 CRO, while the EU Accession Treaty and the ratification of treaties amending the EU founding treaties are governed by the special provisions of Article 148 CRO. This was also acknowledged by the CCRO in Decision no 127/06.03.2019. That decision held that the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union46 (TSCG) is not formally part of EU law. It ruled that way because the TSCG is an intergovernmental instrument concluded pursuant to Article 20 TEU as a form of enhanced co-operation within the EU’s non-exclusive competences. Thus, in the Court’s view, only Article 11 CRO is applicable in relation to the mentioned treaty, and not Article 148.47 The main difference between the rules applicable to EU law and international law in the Romanian legal system is that binding EU law instruments prevail over conflicting national law (according to Article 148 CRO), unlike most treaties. In compliance with Article 20 CRO, only fundamental human rights treaties that are ratified by Parliament have precedence over conflicting national law, unless the Constitution or other national law contain higher standards of human rights protection. Another important difference between EU law and international law in the Romanian legal system is that binding EU law instruments can be used by the CCRO in the context of constitutionality review of laws, as a legal ground for constitutional review. However, this is possible only if the EU law provision in question is sufficiently clear, precise and unconditional and has sufficient constitutional relevance.48 This use of treaties is impossible as they are hierarchically situated under the constitutional provisions.49

43 Muraru/Tănăsescu,

Constituția României 2019 (n 6) 1333. 1334. practice, ratification of the Treaty of Lisbon was realised in compliance with Art 148(3) CRO. 46 Ratified by Law no 83/2012, Monitorul Oficial no 410/20.06.2012. 47 CCRO 06.03.2019, Decision no 127, Monitorul Oficial no 189/08.03.2019, paras 85, 86 and 87. 48 CCRO 18.05.2011, Decision no 668, Monitorul Oficial no 487/08.07.2011; CCRO Decision no 127 (n 48). 49 Muraru/Tănăsescu, Drept constituțional 14/vol I (n 3) 33. 44 Ibid, 45 In

Romania  663 The Romanian legal system gives equal footing to treaties and national law (treaties ratified by Parliament are part of national law, according to Article 11), with the exception of human rights treaties, as explained above.50 Regarding the judicial control of treaties, the Constitution (Article 146) provides a mechanism only for the ex ante constitutional control of treaties by the Constitutional Court. Such constitutional review can be triggered by one of the Presidents of the two Chambers, by at least 50 deputies or at least 25 senators. If the CCRO decides that the treaty is unconstitutional, the treaty may only be ratified after the Constitution is amended accordingly. This is because only treaties that comply with the Constitution can be ratified by the Parliament.51 This type of judicial review has not been used, until now, in relation to EU Treaties. The interaction between national law and EU law is governed by Article 148(2) CRO, which enshrines the primacy of binding EU law over conflicting national law. The CCRO has acknowledged the primacy of EU law over national legislation on the grounds of Article 148(2) CRO based on Romania’s obligations resulting from its EU membership. However, the CCRO also declared that it lacks jurisdiction to review the compatibility between national law and EU law and that the national courts are entrusted to apply EU law according to the principle of primacy of EU law enshrined in Article 148(2) CRO.52 Furthermore, the CCRO acknowledged, on the grounds of Article 148(2) and (4) CRO, the constitutional relevance of EU law, under certain cumulative conditions. According to the CCRO, it may use binding EU law in the context of constitutionality control of laws, as a legal ground for review. However, it can do so only if the EU law provision in question is sufficiently clear, precise and unconditional and has sufficient constitutional relevance, so that its normative content supports the possible unconstitutionality of the national law.53 In the practice of the CCRO, there are few decisions in which the Court accepted the use of provisions of EU law, as interposed norms that can be integrated in the standard of review for the control of constitutionality of national legislation. This mechanism was used by the Court in Decision no 534/2018. In that case the Court found that Article 21(1) TFEU and Article 7(2) of Directive 2004/38 can be used as legal grounds for the constitutionality control, based on Article 148(2) and (4) CRO. It found that way because these EU law provisions have a precise and unequivocal meaning, clearly established by the CJEU, as well as constitutional relevance, as they concern a fundamental right – the right to private life.54 In this case, it must be pointed out that before delivering this judgment, the CCRO decided, for the first time, to submit a request for a preliminary ruling to the CJEU.55 Therefore, Article 148 CRO is the legal basis for both the transfer of powers to the EU and the acknowledgment of the primacy of EU law in the Romanian legal system. Also, Article 148 CRO is the legal foundation for using the provisions of EU law in the context of the Constitutional Court’s constitutionality control of laws. The current Constitution of Romania does not contain other specific provisions on EU membership or EU integration, except for the general constitutional

50 Art 20 CRO. 51 Articles 11(3) and 147(3) CRO. 52 CCRO 25.02.2010, Decision no 137/25.02.2010, Monitorul Oficial no 182/22.03.2010; CCRO Decision no 668 (n 49); CCRO 21.06.2011, Decision no 802 Monitorul Oficial no 603/26.08.2011; CCRO 15.12.2015, Decision no 887, Monitorul Oficial no 191/15.03.2016, para 75; CCRO 26.06.2019, Decision no 414, Monitorul Oficial no 922/15.11.2019, para 116. 53 CCRO, Decision no 668 (n 53); CCRO no 127 (n 48); CCRO 10.01.2011, Decision no 950, Monitorul Oficial no 20/10.01.2013; CCRO 24.10.2018, Decision no 64, Monitorul Oficial no 366/26.04.2018. 54 CCRO 18.07.2018, Decision no 534, Monitorul Oficial no 842/03.10.2018. The same mechanism was used also in CCRO no 64/14.02.2018 (n 54). 55 CJEU 05.06.2018 C-673/16 (Coman and Others) ECLI:EU:C:2018:835.

664  Mihaela Vrabie clause related to EU membership (Article 148) and the constitutional clause related to the possible replacement of the national currency with the euro (Article 137).

IV.  Constitutional Limits to EU Integration The current provisions of the Constitution do not set any explicit limits to European integration. As mentioned above (section III), according to Article 148 CRO, the special ratification procedure for the ratification of the EU Accession Treaty applies also to the ratification of treaties amending the founding treaties of the EU56 and applies in general to the ratification of treaties involving the transfer of state powers to the EU or regarding the joint exercise with the EU of specific powers deriving from state sovereignty.57 It establishes that the ratification will be realised by a law adopted in joint session of the Chamber of Deputies and the Senate, with a two-thirdsmajority of the total number of deputies and senators. Therefore, considering the current provisions of the Constitution, there are no constitutional limits to measures strengthening EU integration or to the further transfer of powers to the EU, at least as long as such measures will be based on binding obligations undertaken by Romania as an EU Member State or under international law. Regarding further transfer of powers to the EU through treaties, it is worth repeating that, procedurally, the CCRO has the power of ex ante constitutionality review of treaties under Article 146(b) CRO. Even though the Court can block the ratification of a treaty agreed upon by the Government,58 in practice, this mechanism has not been used since 2003, when it was written into the Constitution. The coming into force of the Treaty of Lisbon passed without any significant constitutional or political debate in Romania, and its ratification process proceeded without delay or problems. According to Article 147(3) CRO, once the unconstitutionality of a treaty or an international agreement has been declared by the CCRO, that treaty or international agreement cannot be ratified by Parliament, because only international agreements that are not contrary to the provisions of the Constitution can be ratified. As a solution, Article 11(3) CRO mentions only the amendment of the Constitution and the ratification of the international agreement after the contradictions have been solved. Hence, only ratified international treaties will become applicable as part of national law. However, the CCRO does not have the power to initiate ex ante constitutionality review of treaties ex officio and systematically. Rather, it can only act at the request of one of the presidents of the two chambers of Parliament, a group of at least 50 deputies or at least 25 senators.59 This condition, combined with the fact that after the ratification, the constitutionality of treaties can no longer be challenged, leads to inefficiency in the constitutionality control of treaties, set forth by the Constitution.60 The CCRO is yet to fully develop a strong concept of national constitutional identity, as a concept opposing EU integration, and it is unlikely that in the future such a concept may be

56 In practice, ratification of the Lisbon Treaty was realised in compliance with Art 148(3) CRO. 57 Muraru/Tănăsescu, Constituția României 2019 (n 6) 1332 ff. 58 Articles 11(3) and 146(b) CRO. 59 Art 146(b) CRO. 60 Ioan Muraru and Elena Simina Tănăsescu, Constituția României – Comentariu pe articole [The Romanian Constitution annotated] (Bucharest, CH Beck, 2008) 121.

Romania  665 developed as a counter-argument to further EU integration. This is especially so considering that in Romania public opinion is still largely in favour of the EU, as it was at the time of the EU accession. However, there are some CCRO decisions where the concept of national constitutional identity is mentioned as a limit to the implementation of obligations resulting from Romania’s EU membership and even as a limit to the primacy of EU law. The most recent example is Decision no 390/2021,61 which concerned the constitutionality control of the Law no 304/2004 regarding the judicial organisation and of the Emergency Ordinance of the Government no 90/2018 regarding the department for the investigation of offences committed within the judicial system. In this case, the unconstitutionality was claimed, also, in relation to the reports issued by the European Commission in the context of the MCV and thus in relation to Articles 148(2) and (4) CRO. The national court which sent the case to the Constitutional Court for constitutionality control also sent a request for a preliminary ruling to the CJEU under Article 267 TFEU. In its decision, the CJEU held clearly that (i) Decision 2006/928 is binding in its entirety on Romania and the benchmarks in the Annex to Decision 2006/928 are intended to ensure that Romania complies with the value of the rule of law, set out in Article 2 TEU, and are binding on it, in the sense that Romania is required to take due account of the reports drawn up by the Commission on the basis of that decision, and in particular the recommendations made in those reports; (ii) Article 2 and the second subparagraph of Article 19(1) TEU and Decision 2006/928 must be interpreted as precluding national legislation providing for the creation of a specialised section of the Public Prosecutor’s Office with exclusive competence to conduct investigations into offences committed by judges and prosecutors, where the creation of such a section is not justified by objective and verifiable requirements relating to the sound administration of justice, and is not accompanied by specific guarantees such as, first, to prevent any risk of that section being used as an instrument of political control over the activity of those judges and prosecutors likely to undermine their independence and, secondly, to ensure that exclusive competence may be exercised in respect of those judges and prosecutors in full compliance with the requirements arising from Articles 47 and 48 of the Charter of Fundamental Rights of the European Union; (iii) The principle of the primacy of EU law must be interpreted as precluding legislation of a Member State having constitutional status, as interpreted by the constitutional court of that Member State, according to which a lower court is not permitted to disapply of its own motion a national provision falling within the scope of Decision 2006/928, which it considers, in the light of a judgment of the Court, to be contrary to that decision or to the second subparagraph of Article 19(1) TEU.62

Even though the wording of the CJEU was very clear, the CCRO held in its reasoning that CJEU does not consider the reports issued by the European Commission pursuant to Decision no 2006/928/EC to be legally binding on the authorities of the Romanian State. In conclusion, following the interpretation of the CJEU, the Court held that Decision no 2006/928/EC and the reports issued in the context of the MCV have no constitutional relevance and therefore these provisions cannot be used in the constitutional control of the national legislation. Also, the Court held that the reports issued in the context of the MCV are not legally binding on the authorities of the Romanian state (even if it held Decision no 2006/928/EC legally binding) and that the decision of the CJEU delivered in Case C-355/19 is insufficient reason to change its previous respective case law.



61 CCRO 62 CJEU

08.06.2021, Decision no 390, Monitorul Oficial no 612/22.06.2021. 18.05.2021 C-83/19 (Asociaţia ‘Forumul Judecătorilor din România’) ECLI:EU:C:2021:393.

666  Mihaela Vrabie Regarding the relationship between EU law and national law and the national constitutional identity, the CCRO recognised in the same decision, that according to Articles 148(2) and (4) of CRO, all national state bodies are, in principle, obliged to implement and enforce EU law, and that this also applies to the Court itself, which must ensure the primacy of EU law. But this primacy of EU law, according to the Court, must not be perceived in the sense of removing or disregarding the national constitutional identity, enshrined in Articles 11(3) and 152 CRO (viewed as a guarantee of a core identity of the Constitution), which must not be relativised in the process of EU integration. In the same context, the CCRO held that according to Articles 148(2) and (4) CRO Romania cannot adopt national legislation contrary to the obligations to which it has committed itself as a Member State, but these obligations know a constitutional limit, based on the concept of ‘national constitutional identity’. Also, the Constitutional Court held that although Article 148(2) CRO establishes the primacy of legally binding EU law over contrary national law, this provision does not establish the primacy of legally binding EU law over the Romanian Constitution, which means that, according to the Court, a national court does not have the power to review the conformity of a provision of a national law with the provisions of EU law, if that provision of a national law was previously found to be constitutional by the CCRO. According to this decision, the national courts of Romania should refrain from finding any contradiction between EU law and national law, if the national law in question has already been declared constitutional by the CCRO. This could have serious effects on the application of EU law in Romania. By this recent decision, the Constitutional Court has positioned the concept of national constitutional identity, as a concept opposing EU law and opposing, especially, the primacy of EU law over the contrary national legislation. Another recent example where the Constitutional Court mentions national constitutional identity as a limit to the implementation of obligations resulting from Romania’s EU membership is Decision no 414/2019.63 It concerned the constitutionality control of the amendment of Article 30 of Law no 312/2004 regarding the Statute of the National Bank of Romania (BNR). That law limited BNR’s power of maintaining international reserves of gold, eliminating the possibility of depositing gold abroad. This amendment was adopted against the negative opinion expressed by the BNR and without consulting the European Central Bank (ECB). In this case, the unconstitutionality was claimed also in relation to EU law, and the Constitutional Court was called on to use the provisions of EU law as a legal ground for the constitutional control. The EU law provisions allegedly claimed to be breached were Article 127(4), Article 129(4) and Article 282(5) TFEU, as well as Article 2(1) of the Council Decision 98/415/EC.64 In its reasoning, the CCRO held that, since joining the EU, EU law has been binding on Romania, according to the conditions established by the Accession Treaty and the founding EU treaties. More than that, the Court acknowledged that, in the fields in which the EU enjoys exclusive competence, regardless of the treaties it has concluded, the implementation of the obligations resulting from the treaties is subject to EU law. Therefore, the Constitutional Court declared that, according to Articles 148(2) and (4) CRO, Romania applies the obligations resulting from the Accession Treaty in good faith, accepting not to interfere with the exercise of exclusive EU competences, and, based on the conformity clause included in Article 148 CRO, Romania cannot adopt legislative acts contrary to the obligations undertaken as an EU member. Still, after recognising

63 CCRO, Decision no 414 (n 53). 64 29.06.1998, Council Decision 98/415/EC on the consultation of the European Central Bank by national authorities regarding draft legislative provisions, [1998] OJ L 189/42.

Romania  667 the binding legal force of EU law, the Court included in its reasoning a limitation. In particular, it declared that ‘all of the above, of course, know a constitutional limit, expressed in what the Court has described as national constitutional identity’.65 In Decision no 683/2012, the CCRO also refers to the national constitutional identity, but does not define this concept. It places the national constitutional identity in the larger context of the transfer of powers to the EU, declaring that ‘the essence of the EU is the transfer of powers from the Member States to the EU in order to achieve their common objectives, but without prejudicing, in the end, through this transfer of competences, the national constitutional identity’.66 It is worth mentioning that in the same decision, in the reasoning concerning the national constitutional identity, the CCRO also referred to the Lisbon Treaty decision of the German Constitutional Court.67 The conclusion of the Romanian Constitutional Court in this decision was that Member States maintain competences that are inherent in preserving their constitutional identity, and the transfer of powers, as well as the rethinking, strengthening or establishing of new orientations within the powers already transferred is depending on the constitutional margin of appreciation of the Member States.68

The constitutional doctrine describes the identity of the Constitution of Romania as a modern, rational, contemporary one. This is because the Constitution includes a democratic state organisation based on the rule of law and its supreme values are justice, free development of human personality, pluralism, citizens’ fundamental rights and freedoms. Therefore, in the constitutional doctrine it was affirmed that the Constitution should not be interpreted in a restrictive sense, because such an attitude would be contrary to the national constitutional identity.69 The CCRO has not claimed the power to review the constitutionality of EU law or to review EU legislation against the yardstick of EU law. The Court has acknowledged that it lacks jurisdiction to review the compatibility of national law with EU law and that the national ordinary courts are entrusted to correctly apply EU law, according to the principle of primacy of EU law enshrined in Article 148(2) CRO.70 Only in cases of a conflict between national law and a provision of EU law with constitutional relevance that is sufficiently clear, precise and unconditional, has the CCRO claimed the power to review the compatibility between national law and EU law, in the context of constitutionality review, using the EU law provision as legal ground for constitutional control of national law, based on Article 148(2) and (4) CRO.71 Provisions of treaties (outside EU law) need to be ratified by parliament according to Article 11 CRO to become part of Romania’s national law.72 Law no 590/2003 on treaties provides that, in exceptional situations, some treaties can be ratified by the Government through the procedure of the emergency ordinance, under special conditions. Treaties (outside EU law) need to be ratified 65 CCRO 26.06.2019, Decision no 414, para 116. 66 CCRO 27.06.2012, Decision no 683, Monitorul Oficial no 479/12.07.2012, Section II. 67 BVerfG 30.06.2009, 2 BvE 2/08 (Lissabon) BVerfGE 123, 267, ECLI:DE:BVerfG:2009:es20090630.2bve000208. 68 CCRO Decision no 683 (n 67). 69 Elena Simina Tănăsescu, ‘Despre identitatea constituțională și rolul integrator al Constituției’ [‘About the constitutional identity and the integrative role of the Constitution’], (2017) 5 Curierul Judiciar 243. 70 CCRO, Decision no 137 (n 53); CCRO, Decision no 668 (n 49); CCRO, Decision no 802 (n 53); CCRO, Decision no 887 (n 53); CCRO, Decision no 414 (n 53), para 116. In the same line, CCRO 26.11.2009, Decision no 1596, Monitorul Oficial no 37/18.01.2010. 71 CCRO, Decision no 668 (n 49); CCRO, Decision no 127 (n 48); CCRO, Decision no 950 (n 54); CCRO, Decision no 64 (n 54); CCRO, Decision no 534 (n 55). 72 Muraru/Tănăsescu, Constituția României (n 61) 116.

668  Mihaela Vrabie by the Parliament by a majority of the members present in each chamber of Parliament or by a majority of the members of each chamber of Parliament, depending on whether the content of the treaty to be ratified concerns a subject matter in the field of ordinary law or organic law.73

V.  Constitutional Rules and Practice on Implementing EU Law In the Romanian legal order, EU founding treaties and other provisions of EU law with binding legal force take precedence over contrary national law, according to Article 148(2) CRO. Within the meaning of this article, ‘national law’, which is superseded by contrary EU law, includes the broader category of all normative acts issued by the state authorities, not only the acts adopted by parliament.74 Although Article 148(2) CRO mainly focuses on the primacy of EU law over conflicting national law, the Romanian legal system acknowledges also other specific effects of EU law, like direct effect or direct applicability. The provisions of Articles 148(4) and (5) CRO75 enshrine institutional guarantees to ensure the fulfilment of the obligations resulting from Romania’s EU membership. These obligations refer to the effective and active participation of Romania in the EU, to the EU institutional and legal order, as well as to the conscious and full assumption of the obligations that result from the entire legal order of the EU. The institutional guarantees cover all three powers in the state – legislative, executive and judicial – due to the fact that participation in the EU implies not only the implementation or transposition into national law of EU law (result of the activity of the Parliament or of the Government) but also the application of EU law, realised by the administrative and judicial bodies. Therefore, all the public authorities of the Romanian state are bound to contribute, within the limits of the powers assigned to them by the Constitution, to the efficient observance and implementation of the provisions of the EU normative system.76 The essential meaning of Article 148(4) CRO is to establish a clear obligation at the constitutional level (thus, not easily amended), for the Parliament, for the President of Romania, for the Government and for the judicial authority to implement EU law and, furthermore, to guarantee that the other state authorities, bodies or institutions are implementing EU law. According to Article 148(5) CRO, the Government must inform the two chambers of Parliament about the projects of normative acts before they are sent to the EU institutions. The Constitution lacks any special procedural provisions governing the implementation of the EU legislation. Therefore, the subsequent national legislation aiming to implement secondary EU legislation or to transpose EU directives, will be adopted according to the general rules for the adoption of internal legislation, established by Articles 73–78 CRO (for Acts of Parliament) or by Article 108 CRO (for acts of government). Article 73 CRO on the categories of laws that can be adopted by the Parliament, establishes three types of laws. First, the Parliament can adopt constitutional laws, which concern

73 Muraru/Tănăsescu, Constituția României 2019 (n 6) 1431. 74 Ibid, 1339. 75 CRO, Art 148(4): ‘The Parliament, the President of Romania, the Government, and the judicial authority shall guarantee that the obligations resulting from the accession act and the provisions of paragraph (2) are implemented.’ CRO, Art 148 (5): ‘The Government shall send to the two Chambers of the Parliament the draft mandatory acts before they are submitted to the European Union institutions for approval.’ 76 Muraru/Tănăsescu, Constituția României 2019 (n 6) 1335.

Romania  669 amendment of the Constitution. Second, it can adopt organic laws, which are reserved for more important areas, eg the electoral system, the organisation and holding of a referendum, the state of siege and emergency, the state of partial or total mobilisation of the armed forces and the state of war. Third, it can adopt ordinary laws, which concern all other areas that do not fall under the objects of regulation provided for organic law. The fields of organic law are strictly provided by Constitution and are of strict interpretation, as the Constitutional Court ruled.77 Romanian courts have gradually understood their important role in the uniform application of EU law and make constant use of the primacy of EU law over contrary national law.78 Regarding the implementation of EU law, the Constitutional Court declared that, according to Article 148(4) CRO, state authorities are set to guarantee the fulfilment of the obligations resulted from the EU founding treaties and other binding EU law provisions79 and that Romania’s EU membership implies the application of EU law.80 As mentioned before, the CCRO has acknowledged the primacy of EU law, that it lacks jurisdiction to review the compatibility between national law and EU law and that the national courts are entrusted to correctly apply EU law, according to the principle of primacy of EU law enshrined by Article 148(2) CRO.81 This view was fully developed after a period of initial hesitation regarding the perception of EU law and the practice on implementing EU law, marked by jurisprudential fluctuations.82 In Decision no 668/2011, the CCRO clearly held that it lacks the power to review national law in light of the TFEU, based on Article 148 CRO. According to the Court, such jurisdiction belongs to the ordinary national courts, which, in order to reach a correct and legal conclusion, may ask (ex officio, or at the request of a party to the proceeding) a preliminary question under Article 267 TFEU to the CJEU. In the same decision the Court held that it would claim the jurisdiction to rule on the compliance of the national law with the provisions of EU law where a conflict of jurisdiction between the two courts might occur.83 In this case, the CCRO was called to review the constitutionality of Emergency Ordinance of the Government (EOG) no 50/2008,84 which established the obligation to pay a pollution tax applied to motor vehicles at the moment of their first registration in Romania. The decision of the Constitutional Court came after the CJEU delivered its decision in Tatu,85 which generated many objections of unconstitutionality raised before the Romanian national courts by those who paid the pollution tax. Article 110 TFEU must be interpreted as precluding a Member State from introducing a pollution tax levied on motor vehicles on their first registration in that Member State if that tax is arranged in such a way that it discourages the placing in circulation in that Member State of second-hand vehicles purchased in other Member States without discouraging the purchase of second-hand vehicles of the same age and condition on the domestic market.86

77 CCRO 18.05.1994, Decision no 53/1994, Monitorul Oficial no 312/09.11.1994. 78 Muraru/Tănăsescu, Constituția României 2019 (n 6) 1339. 79 CCRO 19.05.2009, Decision no 802, Monitorul Oficial no 428/23.06.2009. 80 CCRO 16.04.2003, Decision no 148, Monitorul Oficial no 428/23.06.2009. 81 CCRO, Decision no 137 (n 53); CCRO, Decision no 668 (n 49); CCRO, Decision no 802 (n 53); CCRO, Decision no 887 (n 53), para 75; CCRO, Decision no 414 (n 53), para 116; CCRO, Decision no 1596 (n 71; CCRO, Decision no 137 (n 53); CCRO 07.06.2007, Decision no 558, Monitorul Oficial no 464/10.07.2007; CCRO 18.05.2010, Decision no 676, Monitorul Oficial no 416/22.06.2010; CCRO 07.12.2010, Decision no 1574, Monitorul Oficial no 110/11.02.2011. 82 Elena Simina Tănăsescu, ‘La primauté du droit communautaire dans la jurisprudence de la Cour Constitutionelle roumaine’, (2009) 3 Analele Universității din București 120. 83 CCRO, Decision no 668 (n 49) para 3. 84 Monitorul Oficial no 327/25.04.2008. 85 CJEU 07.04.2011 C-402/09 (Tatu) ECLI:EU:C:2011:219. 86 Ibidem.

670  Mihaela Vrabie In the Nisipeanu case,87 the CJEU endorsed the same reasoning and gave a similar decision. Since the CJEU clearly interpreted Article 110 TFEU and the Government and the legislative power refused to amend EOG no 50/2008, which violated EU law, the CCRO was called to review the constitutionality of EOG no 50/2008, taking into account the relevant decisions of the CJEU. Although the CJEU offered all the elements needed for the correct interpretation of EU law in relation to EOG no 50/2008, the CCRO refused to review the constitutionality of the measure based on Article 110 TFEU, as a legal ground for constitutional control.88 The same position of the Court was clearly expressed in decision no 498/2012, where it was declared that the Court lacks jurisdiction to review the conformity of a provision of national law with the TFEU, based on Article 148 CRO.89 More nuances were added to this rule by the CCRP in decision no 137/2010. In that case it held that the task of applying the binding Community regulations based on the principle of primacy with regard to the provisions of the national legislation, belongs to the ordinary national courts. It is a question of law enforcement, not constitutionality. The Court finds that, in the relation between the Community law and the national legislation (with the exception of the Constitution), it is possible to speak only of primacy in the application of Community law over the national law, which is a matter that falls within the jurisdiction of the ordinary courts. Moreover, the Court holds that, if the contrary view is accepted, in the sense that the Constitutional Court may establish the constitutionality or unconstitutionality of a provision of national law related to the provisions of a Community act, it would obviously breach the jurisdiction of the Court of Justice of the European Union, since the CJEU has the power to interpret the EU treaties (art. 267 TFEU).90

However, it seems that this solution was not entirely original, since it was inspired by a decision of 19 December 2006 of the Constitutional Tribunal of Poland.91 In another case, the CCRO held that lack of compatibility between national law and EU law will not lead to invalidation of the national law provision within the national legal order.92 In the same line, it was considered that unconstitutionality of a national law provision cannot be ascertained based only on its non-compliance with EU law, because the CCRO can only declare the unconstitutionality of national law based on the contradictions between the national law and the Constitution.93 Regarding the practice of national ordinary courts in implementing EU law, by Decision no 1609/2010,94 the CCRO endorsed the possibility of judicial review of the final court decisions delivered in administrative litigation, based on Article 21 of Law no 554/2004 on administrative proceedings.95 That review aims at removing the violation of the principle of the primacy of EU law. This decision of the CCRO was further endorsed and developed by the High Court of Cassation and Justice (panel of judges for interpreting Romanian law) in Decision no

87 CJEU 07.07.2011 C-263/10 (Nisipeanu) ECLI:EU:C:2011:466. 88 CCRO Decision no 668 (n 49) para 3. 89 CCRO 10.05.2012, Decision no 498 /10.05.2012, Monitorul Oficial 428/28.06.2012. 90 CCRO 25.02.2010, Decision no 137, Monitorul Oficial no 182/22.03.2010. 91 Dragoș Călin, Dialogul dintre tribunalele constituționale și Curtea de Justiție a Uniunii Europene [Dialogue between the Constitutional Couorts and the CJEU] (Bucharest, Editura Universitară, 2018) 101. 92 CCRO 17.012007, Decision no 59, Monitorul Oficial no 98/08.02.2007. 93 Károly Benke, ‘Competența instanțelor constituționale în raport cu aceea a Curții de Justiție a Comunităților Europene în lumina jurisprudenței acestora’ [‘Jurisdiction of the Constitutional Courts vis-à-vis the CJEU in the light of their case-law’], (2008) 11 Dreptul 138. 94 CCRO 09.12.2010, Decision no 1609 Monitorul Oficial no 70/27.01.2011. 95 Monitorul Oficial no 1154/07.12.2004.

Romania  671 45/12.12.2016,96 which concerned a preliminary ruling (which is mandatory for all courts of law) on the interpretation of the provisions of Article 21(2) of Law no 554/2004.97 The first decision where the CCRO considered the possibility of using EU law provisions as standards of constitutional review of national law was Decision no 668/2011.98 In this decision, the Court ruled that the use of EU law provisions as interposed norms within the control of constitutionality of the national law implies, according to Article 148 (2) and (4) of the Constitution, a cumulative conditionality: the EU law provision must be sufficiently clear, precise and unconditional by itself or its meaning to have been clearly, precisely and unequivocally established by the Court of Justice of the European Union and, on the other hand, the EU rule must have a certain level of constitutional relevance, so that its normative content supports the possible violation by the national law of the Constitution – the only direct reference norm within the constitutionality control. In such a hypothesis, the Constitutional Court’s constitutionality review is distinct from the mere application and interpretation of the law, which are the responsibilities of courts and administrative authorities.99

The same view was maintained by the CCRO in subsequent decisions.100 Therefore, the Court created a test, based on which it selects the situations in which it can exceptionally review the constitutionality of national law by using EU law as a legal ground. The problem with this test is that it lacks any objective criteria for the evaluation of the level of constitutional relevance of the EU law instrument in question. In this context, it is worth mentioning that in the end, in Decision no 668/2011, the Court settled the case by declaring that, although the EU law provision in question was sufficiently clear, precise and unconditional, it lacked constitutional relevance. The Court did not explain why the EU law lacked constitutional relevance. What is important to point out is that, according to the CCRO, even when it finds that all the conditions are met for the use of EU law provisions within the control of constitutionality, EU law provisions can only serve as ‘interposed norms’ that can be integrated in the standard of review for the control of compatibility of national legislation with Article 148 CRO. Hence, the ‘reference norm’ used in the constitutionality review is Article 148, the only ‘direct’ norm that can be used in this type of control, while EU law provisions are ‘interposed norms’. Another interesting result of this interpretation of the CCRO is that EU law provisions can have different effects, depending on the judicial proceedings in which they are invoked. When EU law provisions are invoked in judicial proceedings before the ordinary national courts, the conflicting national law must be set aside by the national judge (based on the primacy of EU law). However, when EU law provisions are invoked in the context of the constitutionality control, the Court can decide to trigger Article 148 CRO and declare the unconstitutionality of the national legislation which is incompatible with EU law. In any case, the Court reserved the exclusive right to decide when EU law provisions can be used as ‘interposed norms’ within the control of constitutionality of national law.

96 Monitorul Oficial no 386/23.05.2017. 97 Mihaela Vrabie, ‘Revizuirea hotărârilor definitive ale instanţelor de contencios administrativ în cazul nerespectării hotărârilor Curţii de Justiţie a Uniunii Europene. Privire critică asupra termenului de revizuire stabilit de Înalta Curte de Casaţie și Justiţie prin Decizia obligatorie nr. 45/2016 [‘Judicial review of final judgments delivered by national courts in administrative litigation in case of non-compliance with the interpretation of EU law given by the CJEU. Critical view on the procedural term for judicial review established by the High Court of Cassation and Justice by mandatory Decision no 45/2016’], (2018) 2 Revista Română de Drept European 49. 98 CCRO, Decision no 668 (n 49). 99 Ibid, para 3. 100 CCRO 07.07.2011, Decision no 921, Monitorul Oficial no 673/21.09.2011; CCRO 21.05.2014, Decision no 283, Monitorul Oficial no 454/20.06.2014; CCRO 05.11.2015, Decision no 762, Monitorul Oficial no 86/05.02.2016.

672  Mihaela Vrabie After the Charter of Fundamental Rights of the European Union (Charter) acquired a legally binding status with the entry into force of the Treaty of Lisbon, the possibility was created of making a reference for a preliminary ruling to the CJEU on its interpretation.101 Thus, after its entry into force, an increased interaction between the CJEU and the national (constitutional) courts can be observed in the context of the preliminary reference procedure, in cases regarding the application of the Charter.102 The Charter has also strengthened the role of national courts in the application of EU law, by adding an important legally binding instrument in the field of protection of fundamental rights that must be taken into account and applied by the national courts when Member States are implementing EU law. This multiplication of the legal instruments for the protection of fundamental rights at the EU level can result in contradictions between national legislation and the provisions of Charter that must be solved by the national courts.103 The CCRO for the first time took into account, from a constitutional perspective, the judicial enforcement of the Charter in 2010. Initially, the Court ruled that the provisions of the EU Charter are applicable within the constitutionality control only if they ensure, guarantee and develop the Romanian constitutional provisions in the field of fundamental rights and only if the level of protection provided by the EU Charter is not lower than the level of protection provided by the Romanian constitutional norms in the field of human rights.104

Later, the Court ruled that ‘it is clear from the case-law of the CJEU that the EU Member States are required to respect the EU fundamental rights enshrined by EU law, when they are implementing EU law’ and declared that this rule, as enshrined in the Charter, applies equally to central authorities and to regional or local courts, as well as to public bodies when they are implementing EU law, therefore obliging the EU Member States to apply these rules as far as possible.105 In another case, the CCRO acknowledged the legal status of the Charter, declaring that these provisions contained in a legal instrument having the same legal force as the founding EU treaties, must be related to the provisions of Article 148 and not Article 20 CRO, which governs human rights treaties.106 In 2015, the Court decided to finally review the constitutionality of a provision of national law by using EU law – including a provision of the Charter – as a standard of review. In Decision no 64/2015, the CCRO ruled that Article 153(1) TFEU, Article 27 of the Charter and Articles 2 and 3 from Directive 98/59/CE, which were allegedly breached by national law, meet the conditions established by the Court for the use of EU law provisions within the control of constitutionality, as ‘interposed norms’. In other words, the Court decided that these EU law provisions were sufficiently clear, precise and unconditional and had a certain level of constitutional relevance,

101 Mihaela Vrabie, ‘The Review of National Action for Compliance with the Charter of Fundamental Rights of the European Union’, in M Sandru, G Martinico, M Banu (eds), Dealing with EU Law: Role of the National Courts in Interpretation and Application of EU Law (Bucharest, Editura Universitară, 2014) 142. 102 MihaelaVrabie, Protecția judiciară a Cartei drepturilor fundamentale a Uniunii Europene [Judicial Protection of the Charter of Fundamental Rights of the European Union] (Bucharest, Universul Juridic, 2017) 238. See also the preliminary reference made by the French Constitutional Court in CJEU C-168/13, (PPU – F.) ECLI:EU:C:2013:358 and the preliminary reference made by the Spanish Constitutional Tribunal in CJEU 26.02.2013 C-399/11 (Melloni) ECLI:EU:C:2013:107. 103 Vrabie, Protecția (n 103) 227 f. 104 CCRO 15.06.2010, Decision no 871/25.06.2010, Monitorul Oficial no 433/28.06.2010. In the same line CCRO 08.11.2011, Decision no 1479, Monitorul Oficial no 59/25.01.2012; and CCRO 13.12.2016, Decision no 745 Monitorul Oficial no 148/28.02.2017. 105 CCRO 25.01.2012, Decision no 53, Monitorul Oficial no 234/06.06.2012. 106 CCRO 2011.2012, Decision no 967/20.11.2012, Monitorul Oficial no 853/18.12.2012. In the same line, CCRO 02.07.2014, Decision no 390, Monitorul Oficial no 532/17.07.2014.

Romania  673 so that their normative content supported the violation by the national law of the Constitution. More specifically, in this case, the Court held that the EU law provisions had constitutional relevance because they protected the right of the workers to be informed and consulted, which is related to the fundamental right to the social protection of workers, enshrined by Article 41(2) CRO. Another relevant aspect in the reasoning was that the Court declared that the EU law used within the control of constitutionality did not breach the national constitutional identity. In the end, the Court found the national law unconstitutional for violating Article 148(2) CRO, which, in the Court’s opinion, implicitly includes a clause of conformity of national laws with EU law. When this clause is violated and the EU law provisions have constitutional relevance, it must be sanctioned by the CCRO.107 After this decision, it was believed that the CCRO changed its view, showing a large opening towards EU law, while still self-limited on the aspects of national constitutional identity, which is an undefined concept and interpretable.108 An interesting development regarding the constitutionality control of implementing secondary EU legislation was marked by Decision no 1258/2009109 of the CCRO. In this case, the Court constitutionally reviewed Law no 298/2008, which transposed Directive 2006/24/EC.110 It declared the national law unconstitutional, finding a violation of the right to intimate, private and family life and of the right to the secrecy of correspondence and freedom of expression enshrined in the Constitution. Although the Court applied national constitutional provisions within the constitutionality control of a national law transposing an EU directive, it did not use the opportunity to refer to the equivalent EU fundamental rights enshrined in the Charter or to submit a preliminary reference to the CJEU regarding the EU standard of protection of the fundamental rights in question. An academic opinion was expressed that in this case a preliminary reference to the CJEU was indeed necessary.111 Later on, Directive 2006/24/CE was invalidated by the CJEU in Digital Rights Ireland112 for breaching Articles 7 and 8 of the Charter. After the invalidation, the Constitutional Court was called again to review the constitutionality of another national law (Law no 82/2012) that transposed Directive 2006/24/CE. This time, the Court referred extensively to Articles 7 and 8 of the Charter, quoting also the most relevant fragments from the reasoning of the CJEU.113 On 29 November 2016, the CCRO finally decided to submit its first preliminary reference to the CJEU, in Coman.114 The case dealt with the potential conflict between Article 277 of the Romanian Civil Code, which provided that marriages between same-sex persons concluded in other countries are not recognised in Romania, and Article 21(1) TFEU and Article 7(2) of Directive 2004/38/EC.115 Submitting a preliminary reference was a turning point in the practice of 107 CCRO 24.02.2015, Decision no 64, Monitorul Oficial no 286/28.04.2015. 108 Mona Pivniceru and Károly Benke, ‘Constitutional Court of Romania: Constitutionalisation of the obligations under international treaties and European Union binding acts. Decision no 64 of 24 February 2015’, (2015) 3 ICL Journal 455, 456. 109 CCRO 08.10.2009, Decision no 1258/08.10.2009, Monitorul Oficial no 798/23.11.2009. 110 Directive 2006/24/EC of the European Parliament and of the Council of 15 March 2006 on the retention of data generated or processed in connection with the provision of publicly available electronic communications services or of public communications networks and amending Directive 2002/58/EC, [2006] OJ L 105/54. 111 Călin, Dialogul (n 92) 146. 112 CJEU 0804.2014 Joined Cases C-293/12 and C-594/12 (Digital Rights Ireland) ECLI:EU:C:2014:238. 113 Ibid. 114 CJEU C-673/16 (n 56); CCRO, Case no 78D/2016. 115 Directive 2004/38/EC of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States amending Regulation (EEC) No 1612/68 and repealing Directives 64/221/EEC, 68/360/EEC, 72/194/EEC, 73/148/EEC, 75/34/EEC, 75/35/EEC, 90/364/EEC, 90/365/EEC and 93/96/EEC, [2004] OJ L 158/77.

674  Mihaela Vrabie the Court that previously constantly had rejected requests for submitting preliminary references to the CJEU, sometimes without any reasoning116 and sometimes claiming that the preliminary reference was only for ordinary courts to make.117 The request has been made in proceedings related to the refusal of the Romanian authorities to grant a right of residence in Romania for more than three months to a third-country national based on a marriage lawfully concluded abroad between an EU citizen and his spouse who was of the same sex, a third-country national, on the ground that a marriage between people of the same sex is not recognised under Romanian law. In this context, the CCRO had doubts as to the interpretation to be given to relevant provisions of Directive 2004/38, read in the light of the Charter and of the recent case law of the CJEU. Thus, the Court asked the CJEU for an interpretation of Articles 2(2)(a), 3(1) and 7([2]) of Directive 2004/38, read in the light of Articles 7, 9, 21 and 45 of the Charter, asking whether these EU law provisions require a Member State to grant the right of residence in its territory for a period of longer than three months to the same-sex spouse of a citizen of the EU. Unsurprisingly, the CJEU held that Article 21(1) TFEU is to be interpreted as meaning that, in circumstances such as those of the main proceedings, a third-country national has the right to reside in the territory of the Member State for more than three months and that derived right of residence cannot be made subject to stricter conditions than those laid down in Article 7 of Directive 2004/38. In the end the CCRO followed the CJEU reasoning and decided to use Articles 21(1) TFEU and 7(2) of Directive 2004/38 as interposed norms for the constitutionality control, based on Articles 148(2) and (4) CRO.118 Thus, the Court found that the provisions of Article 277 of the Romanian Civil Code are constitutional only to the extent that they allow granting of the right of residence on the territory of the Romanian state, under the conditions stipulated by EU law, to the spouses – EU citizens and/or third-country nationals – of same-sex marriages, concluded in an EU Member State. Therefore, with Decision no 534/18.07.2018, the CCRO not only endorsed the primacy of EU law and of the interpretation of EU law offered in this case by the CJEU, upon its first referral to the CJEU, but also clearly acknowledged the constitutional relevance of EU law, since this case was one of the few when the Court decided to use EU law provisions as interposed norms integrated in the standard of review for the control of constitutionality of national law. In order to do this, the Constitutional Court held that Article 21(1) TFEU and Article 7(2) of Directive 2004/38 have both a precise and unequivocal meaning, clearly established by the CJEU, as well as constitutional relevance, as they concern a fundamental right, respectively the right to private life.

VI.  Resulting Relationship between EU Law and National Law The constitutional foundations of Romania’s EU membership are enshrined in Article 148 and were included in the Constitution in 2003, following a constitutional amendment that prepared Romania’s accession to the EU. Article 148 CRO is a rather complex ‘European clause’, setting forth not only the procedure for transferring sovereign powers to international organisations, but also the interaction and the 116 CCRO 25.03.2008, Decision no 392, Monitorul Oficial no 309/21.04.2008; CCRO 25.03.200, Decision no 394, Monitorul Oficial no 309/21.04.2008. 117 CCRO Decision No 668 (n 49). 118 CCRO, Decision no 534 (n 55). The same mechanism was used by the CCRO in Decision no 64 (n 54).

Romania  675 relationship between Romanian and EU law. From this perspective, Article 148(2) CRO clearly establishes the primacy of legally binding EU law over contrary national law. It must be pointed out that, within the meaning of Article 148(2), all normative acts of the state authorities, not only the laws adopted by the Parliament, fall within the category of national law that must be set aside in case of conflict with EU law. Still, the Constitutional Court declared recently that the Constitution is not included in the category of national law that must be set aside in case of conflict with EU law. Besides the primacy of EU law, the ‘European clause’ sets forth distinct institutional guarantees regarding the implementation of EU law at national level. According to Article 148(4) CRO, all three branches (legislative, executive and judicial) must guarantee the observance of the obligations resulting from EU law. The Constitutional Court acknowledged the primacy of EU law over national legislation on the grounds of Article 148(2) CRO, based on the obligations resulting from the EU membership. Furthermore, the Court declared that Article 148(2) CRO implicitly includes a clause of conformity of national law with EU law, which can lead to the unconstitutionality of national law provisions when this clause is violated and when the EU law provisions in question have constitutional relevance. Initially, the Court declared that it lacks jurisdiction to review the compatibility between national law and EU law, which is a matter that falls within the jurisdiction of the ordinary courts. Subsequently, the CCRO offered a nuanced opinion and acknowledged, on the grounds of Articles 148(2) and (4) CRO, the constitutional relevance of EU law, under certain conditions. In the Court’s view, it can use legally binding EU law as a ground for constitutional review of national laws only if the EU law provision in question is sufficiently clear, precise and unconditional and has a certain level of constitutional relevance, so that its normative content supports the possible violation of the constitutional provisions by the national law under review. Therefore, the Court created a test, based on which it selects the situations when EU law provisions can serve, exceptionally, as standards of constitutional review of national law. It is important to point out that, even when the Court finds that all these conditions are met, the EU law provisions can only serve as ‘interposed norms’ that can be integrated in the standard of review for the control of constitutionality of national legislation with Article 148 CRO. Hence, the ‘reference norm’ used in the constitutionality review is Article 148, the only ‘direct’ norm that can be used in this type of control, while EU law provisions are ‘interposed norms’. Nevertheless, the CCRO reserved the exclusive right to decide when EU law provisions can be used as ‘interposed norms’ within the control of constitutionality of national law. This interpretation of the CCRO has important consequences on the judicial enforcement of EU law on national level because it allows EU law to produce different effects. When EU law provisions are invoked in judicial proceedings before ordinary national courts, the conflicting national law must be set aside by the national judge (based on the primacy of EU law). However, when EU law provisions are invoked in the context of the constitutionality control, the Court can decide to trigger Article 148 CRO and declare the unconstitutionality of the national legislation which is incompatible with the EU law provisions. The Constitutional Court is yet to fully develop a concept of national constitutional identity as a concept opposing EU integration. However, there are some decisions of the Court where the concept of national constitutional identity, while not defined, is mentioned as a limit in the larger context of the transfer of powers to the EU and more recently, as a limit to the implementation of obligations resulting from Romania’s EU membership and even as a limit to the primacy of EU law. Since the decisions of the Constitutional Court are legally binding on all the other national courts in Romania, the decision of the Constitutional Court which establishes that a national

676  Mihaela Vrabie court does not have the power to review the conformity of a provision of a national law with the provisions of EU law, if that provision of a national law was previously found to be constitutional by the CCRO,119 can produce systemic effects on the application of EU law at national level in the future. In the field of EU fundamental rights, initially the CCRO declared that the provisions of the EU Charter are applicable within the constitutionality control only if they contribute to reinforcing the meaning given to constitutional rights and only if the level of protection provided by the Charter is not lower than the level of protection provided by the constitutional rights. After several decisions that acknowledged the legally binding status of the Charter, the Court eventually decided to review the constitutionality of a national law provision based on a provision of the Charter and even to submit a preliminary reference to the CJEU regarding the interpretation of the Charter. All in all, regarding the judicial enforcement of EU law, ordinary Romanian courts have gradually understood their important role in the uniform application of EU law and the application of the primacy of EU law over conflicting national law. Furthermore, the Constitutional Court accepted the constitutional relevance of certain EU law provisions and decided to review the constitutionality of national law by using EU law provisions – including Charter – as interposed norms that can be integrated in the standard of review.

References K Benke, ‘Competența instanțelor constituționale în raport cu aceea a Curții de Justiție a Comunităților Europene în lumina jurisprudenței acestora’ [‘Jurisdiction of the Constitutional Courts vis-à-vis the Court of Justice of the European Communities in the light of their case law’], (2008) 11 Dreptul 99. D Călin, Dialogul dintre tribunalele constituționale și Curtea de Justiție a Uniunii Europene [The Dialogue between the Constitutional Courts and the CJEU] (Bucharest, Editura Universitară, 2018). Ș Deaconu, Drept constituțional [Constitutional law] (Bucharest, C.H. Beck, 2011). B Iancu, ‘Romania – The Vagaries of International Grafts on Unsettled Constitutions’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (Hague, TMC Asser Press, 2019) 1047. I Muraru and ES Tănăsescu, Constituția României – Comentariu pe articole [The Romanian Constitution annotated] (Bucharest, C.H. Beck, 2008). I Muraru and ES Tănăsescu, Constituția României – Comentariu pe articole [The Romanian Constitution annotated] (Bucharest, C.H. Beck, 2019). I Muraru and E S Tănăsescu, Drept constituțional și instituții politice [Constitutional law and political institutions], 13th edn, volume II (Bucharest, C.H. Beck, 2009). I Muraru and E S Tănăsescu, Drept constituțional și instituții politice [Constitutional law and political institutions], 14th edn, volume I (Bucharest, C.H. Beck, 2011). M Pivniceru and K Benke, ‘Constitutional Court of Romania: Constitutionalisation of the obligations under international treaties and European Union binding acts. Decision no. 64 of 24 February 2015’, (2015) 3 ICL Journal 451. ES Tănăsescu, ‘La primauté du droit communautaire dans la jurisprudence de la Cour Constitutionelle roumaine’, (2009) 3 Analele Universității din București 110. ES Tănăsescu, ‘Despre identitatea constituțională și rolul integrator al Constituției’ [‘About the constitutional identity and the integrator role of the Constitution’], (2017) 5 Curierul Judiciar 243.



119 CCRO,

Decision no 390/2021 (n 61).

Romania  677 M Vrabie, ‘The Review of National Action for Compliance with the Charter of Fundamental Rights of the European Union’, in M Sandru, G Martinico, M Banu (eds), Dealing with EU Law: Role of the National Courts in Interpretation and Application of EU Law (Bucharest, Editura Universitara, 2014) 142. M Vrabie, Protecția judiciară a Cartei drepturilor fundamentale a Uniunii Europene [Judicial Protection of the Charter of Fundamental Rights of the European Union] (Bucharest, Editura Universul Juridic, 2017). M Vrabie, ‘Revizuirea hotărârilor definitive ale instanţelor de contencios administrativ în cazul nerespectării hotărârilor Curţii de Justiţie a Uniunii Europene. Privire critică asupra termenului de revizuire stabilit de Înalta Curte de Casaţie și Justiţie prin Decizia obligatorie nr. 45/2016’ [‘Judicial review of final judgments delivered by national courts in administrative litigation in case of non-compliance with the interpretation of EU law given by the CJEU. Critical view on the procedural term for judicial review established by the High Court of Cassation and Justice by mandatory Decision no. 45/2016)’] (2018) 2 Romanian Review of European Law 49.

678

24 Slovenia The Case of a Maturing Relationship MATEJ AVBELJ

I.  Main Characteristics of the National Constitutional System A.  The Formation of the Slovenian Independent State The Republic of Slovenia declared its independence on 25 June 1991.1 This was the result of a centuries-long struggle for national liberation2 and of the hopes and aspirations that culminated in a condensed historical period between 1987 and 1991.3 Amid favourable geo-political circumstances, marked by the fall of the Berlin Wall, which announced the collapse of the Soviet Union and the demise of the totalitarian communist regimes in Central and Eastern Europe (CEE), the people of Slovenia seized the unique historical opportunity and asserted their statehood. The national identity of Slovenians had been formed in explicit political terms at least since the Spring of Nations in 1848. In cultural terms, Slovenians self-consciously, under a common name, existed long before, at least since the Reformation. Many other cultural monuments, archaeological artefacts and other insignia prove that Slovenians have inhabited this most south-eastern part of the Alpine basin since at least the seventh century, developing their common language, traditions, and customs in limited self-governance under foreign rule.4 As a relatively small cultural community, being uniquely positioned between the Germanic north, the Romanic west, the Hungarian east, and the Slavic south-east under Ottoman rule, since the Spring of Nations Slovenians have sought refuge among the peoples with whom they felt they had most in common. This was a normal reflex of national movements throughout Europe at a time when they were in the pursuit of homogeneity, first of all ethnic, but also cultural. Following the collapse of the Austrian-Hungarian Empire in 1918, the decision was taken to join the union of South Slavic peoples in the Balkans. It was a decision taken under historical

1 The research for this chapter has been supported by the Slovenian Research Agency Grant No J5-7359. 2 Preamble to the Constitution of the Republic of Slovenia (hereinafter CRSI); for an in-depth analysis, see Lovro Šturm (ed), Komentar Ustave Republike Slovenije [Commentary on the Constitution of Slovenia] (Kranj, Evropska Pravna Fakulteta, 2002). 3 Peter Jambrek, ‘Zgodovinski spomin Nove Revije in slovenska prihodnost v Evropi’ [‘The Historical memory of Nova Revija and the Slovenian future in Europe’], (2007) 26 Nova Revija April, 60, 82. 4 For a more extensive discussion, see Peter Štih, Vasko Simoniti, and Peter Vodopivec, Slovenska zgodovina: družba, politika, kultura [Slovenian History: Society, Politics, Culture] (Ljubljana, Inštitut za novejšo zgodovino, 2008).

680  Matej Avbelj duress. After centuries-long political unions with the German-speaking majority, with whom they shared everything but language, the Slovenians joined a Slavic union with the Serbs and Croats, with whom they shared a language family but not much else. Due to the irreconcilable differences between these cultures at large, which were only exacerbated during pre- and postWWII authoritarian and totalitarian regimes, the union of South Slavs always seemed bound to collapse. This perception, as well as the reality of it, was never stronger than in the late 1980s. However, the mere perception of imminent social change is insufficient for the latter to actually take place. What is always needed is a roadmap showing a path for this social change to occur. This roadmap was drawn by a group of Slovenian intellectuals assembled around the journal Nova Revija. In its landmark issue 57 (1987), these intellectuals with disparate world views outlined a so-called Slovenian national programme in which they openly and for the first time called for an independent Slovenia.5 Only three years later, Slovenians were called to a plebiscite in which they were invited to express themselves on whether they wanted Slovenia to become an independent state. Amid a huge turnout, the outcome was overwhelmingly positive. This can be explained by the dormant, even subconscious, but eventually overwhelming desire of the Slovenian people to return to where they had always felt they belonged: Europe. The desire to create a normal European state was put in the cradle of the new Slovenian democracy. The chosen means to achieve this paramount objective was full membership of the European Union (EU) and active participation in the North Atlantic community symbolised by NATO. This explains why Slovenians voted so eagerly in favour of joining both international organisations only 12 years after asserting their country’s sovereignty and independence. A European Slovenia, as a normal European state, was, thus, mandated by a double plebiscite majority. A commitment to European integration is thus embedded in the ethos of the Slovenian Constitution and, thus, as a transcendental order binds this and any future Slovenian political constellation.6

B.  The Constitutional Features of the Slovenian State The Constitution defines Slovenia as a democratic republic, based on the principles of the rule of law and the welfare state.7 Recently, the Constitutional Court read into the provision of Article 1 of the Constitution a guarantee of the protection of equal human dignity.8 Accordingly, Slovenian constitutional democracy should be understood as a foundational democracy whose foundation and paramount normative objective is the protection of equal human dignity.9 The protection of human rights plays a central role in the Slovenian Constitution. Its Article 5 proclaims that Slovenia shall protect human rights and fundamental freedoms within its ­territory.10 Chapter II of the Constitution is devoted exclusively to these human rights. By putting the provisions on human rights before all the other constitutional norms and principles, including

5 (1987) V Nova Revija, 57,. 6 The preceding discussion draws directly on Matej Avbelj, ‘Crises and Perspectives in Building a European Nation – The Case of Slovenia’, in P Jambrek (ed), Nation’s Transitions: Social and Legal Issues of Slovenia’s Transitions: 1945–2015 (Kranj, Graduate School of Government and European Studies, 2014) 395. 7 CRSI Arts 1 and 2. 8 Constitutional Court [hereinafter USRS] 26.09.2011, Case U-I-109/10, ECLI:SI:USRS:2011:U.I.109.10. 9 See Matej Avbelj, ‘Zgodovinska odločitev ustavnega sodišča’ [‘Historic Decisions of the Constitutional Court’] (2011) No 39-40 Pravna praksa, 17. 10 CRSI Art 5/5.

Slovenia  681 those regulating the organisation and functioning of the state, the Constitution grants a special status to human rights and confirms their paramount constitutional importance. Article 15/5 of the Constitution stipulates the highest standard of human rights protection. In this context a special role is played by the European Convention on the Protection of Fundamental Rights and Freedoms (ECHR),11 which is directly applicable in Slovenia12 and, thus, forms part of a body of mandatory legal rules for all state organs.13 Accordingly, no human right or fundamental freedom regulated by legal acts in Slovenia may be restricted on the grounds that a particular right is not recognised by the Constitution or that it is recognised to a lesser extent.14 The Constitution of the Republic of Slovenia puts in place a system of separation of powers. According to Article 3, power is vested with the people; the citizens exercise it directly through elections, consistent with the principle of the separation of legislative, executive and judicial powers.15 By contrast with the former communist regime, the judiciary is now a distinctive and independent branch of government. Judges are to be independent in the performance of their judicial function and are bound exclusively by the Constitution and the statutes.16 Judges are tenured and enjoy all the privileges and immunities typical of a constitutional democracy.17 However, a remnant of the socialist past is that judges are elected by the National Assembly on the proposal of the Judicial Council and can also be dismissed only by the National Assembly in those rare cases provided for by the law.18 The highest court is the Supreme Court.19 Its role is to rule as a last instance in criminal and civil cases, commercial lawsuits, administrative review, and labour and social security disputes. It, thus, decides on ordinary and extraordinary legal remedies. It might also perform other functions provided for by law.20 The Constitution also features a Constitutional Court, which is not part of the ordinary judiciary. The Constitution defines it as the highest body of the judicial power for the protection of constitutionality, legality, human rights and fundamental freedoms.21 Its more detailed regulation can be found in the Constitutional Court Act.22 The Constitutional Court is composed of nine judges. They are proposed by the President of the Republic and, as with the Supreme Court, elected by the National Assembly. The judges are elected for nine-year non-renewable terms.23 Among other factors, it is important for the Constitutional Court’s independence that the National Assembly decides on its budget, upon the proposal of the Court itself. The Constitutional Court has important competences in relation to the executive and the legislative branch. It has jurisdiction to decide a number of actions. These are the constitutionality of laws; the conformity of laws and other regulations with ratified treaties and with the general principles of international law; legality and constitutionality of regulations; legality and constitutionality of local community regulations; legality (including regulations) and constitutionality of

11 For the overview of the role of the ECHR in Slovenia, see Matej Avbelj and Jernej Letnar Černič, ‘Slovenia’, in L Hammer and F Emmert (eds), The European Convention on Human Rights and Fundamental Freedoms in Central and Eastern Europe (The Hague, Eleven International Publishing, 2012) 527. 12 CRSI Art 8: ‘Ratified and published treaties shall be applied directly.’ 13 This is drawing on Avbelj/Letnar Černič, ‘Zgodovinska odločitev’ (n 11). 14 CRSI Art 15/5. 15 CRSI Art 3. 16 CRSI Art 125. 17 CRSI Art 129. 18 CRSI Art 130. 19 CRSI Art 127. 20 Ibid. 21 Constitutional Court Act (OJ 64/07), Art 1. 22 Ibid. 23 CRSI Art 165.

682  Matej Avbelj general acts issued for the exercise of public authority; constitutional complaints stemming from the violation of human rights and fundamental freedoms by individual acts of public authorities; competence disputes between the state and local communities and among local communities themselves; competence disputes between courts and other state authorities; competence disputes between the National Assembly, the President of the Republic, and the Government; the constitutionality of acts and activities of political parties; and other matters for which jurisdiction is vested in the Constitutional Court by the Constitution or laws. In the process of ratifying a treaty, the Constitutional Court, on the proposal of the President of the Republic, the Government, or a third of the deputies of the National Assembly, issues an opinion on the conformity of such a treaty with the Constitution. The National Assembly is bound by the opinion of the Constitutional Court. Unless otherwise provided by law, the Constitutional Court decides on an individual’s constitutional complaint only if all available legal remedies have been exhausted. The Constitutional Court decides whether to accept a constitutional complaint for adjudication on the basis of criteria and procedures provided by law.24

II.  Constitutional Culture The Constitution of the Republic of Slovenia is a living Constitution and the Constitutional Court is its main interpreter and the driving force behind it. The Court defined Slovenia as a constitutional democracy: The Constitution and the decisions of the Constitutional Court are not binding only on the National Assembly as the legislature, but also on citizens when they exercise power directly by voting on an individual law at a referendum. Slovenia is a constitutional democracy, the essence of which is that the values protected by the Constitution prevail over the democratically adopted decisions of the majority.25

The Constitutional Court played a central role in lying down and upholding the legal foundations of the Slovenian state. The first formation of the Court made a lasting contribution in that regard. It introduced all constitutional review standards typical of modern constitutional democracies. The Constitutional Court has always regarded the Constitution as a living document. Its approach to judicial decision-making has been creative, leading some critics to repeatedly accuse the Court of judicial activism.26 Undoubtedly, the Court has established itself not only as a negative but sometimes also as a positive legislator and has proven to be one of the most influential state institutions in Slovenia.27 Among the Court’s recent decisions that are bound to leave a lasting imprint on the Slovenian constitutional development, the Titova cesta28 and the Patria29 cases deserve a special mentioning. In the former case the Constitutional Court invalidated the ordinance of the municipality

24 CRSI Art 160. 25 USRS 08.07.2004, Case U-I-111/04, ECLI:SI:USRS:2004:U.I.111.04, para 26. 26 See, for example, Ivan Kristan, ‘Delitev oblasti ob ankaranski zgodbi’ [‘The Separation of Powers in the Ankaran Story’], (2011) 47 No 1 and 2 Javna uprava, 25 and 218; for an overview see also Marko Novak, Delitev oblasti; medigra prava in politike [The Separation of Powers. An Interplay of Law and Politics] (Ljubljana, Cankarjeva, 2003). 27 Matej Avbelj, ‘(Ne)demokratičnost ustavnosodnih odločb’ [‘(Un)democratic Constitutionality of Decisions’], (2003) 58 Pravnik, 11. 28 USRS U-I-109/10 (n 8). 29 USRS 20.04.2015, Case Up-879/14, ECLI:SI:USRS:2015:Up.879.14.

Slovenia  683 of Ljubljana which in 2011 renamed one of the city streets after the former Yugoslav dictator Josip Broz Tito. The city of Ljubljana rushed with this decision in the context of the discovery of another post-WWII mass grave, an abandoned mine, hiding several hundred victims executed extra-judicially by Tito’s partisans immediately after WWII.30 A former communist concentration camp intern challenged the Ljubljana ordinance for having violated her human dignity and the Court ruled in her favour. By doing so, it put the right to human dignity at the apex of the Slovenian constitutional structure, insisting that as a principle it must pervade all elements and institutional actions of the Slovenian democracy. In the notorious Patria case,31 the Constitutional Court entrenched the principle of legality in criminal law, by quashing the entire process against the former leader of the opposition charged with graft. Constitutionally, the process has been extremely problematic. It lasted for almost a decade, affecting at least three general parliamentary elections. It raised strong suspicions about its political motivations. The leader of the opposition was first indicted and later convicted on all counts and at all instances of the ordinary judiciary, including the Supreme Court. The conviction was for having accepted a promise of an unknown award, at a vaguely determined time, at an unknown place, and through an unidentified method of communication to use his influence then as a prime minister to have a military contract awarded to the Finnish arms-selling corporation ‘Patria’.32 By standing firmly for the rule of law in this case, and putting human dignity at the heart of the Slovenian constitutional order, the Constitutional Court once again proved itself as the ultimate guardian of constitutionality in Slovenia.33 However, Slovenian constitutional culture cannot be properly identified or assessed by merely scratching the surface. One has to dig deeper and look beyond the mere normative level and the formal institutional structure. Slovenian constitutional culture is marked by a great discrepancy between the law on books, where Slovenia has always been regarded as a good disciple, and the law in action, where the Slovenian practical record is sometimes extremely poor. It would not be an exaggeration to argue that Slovenia suffers from a systemic crisis in the rule of law.34 The country features a dysfunctional judiciary. The data gathered by the Council of Europe confirm that Slovenia, while boasting the highest number of judges per capita in Europe,35 has the least efficient and most expensive judiciary. As a result, popular trust in the judiciary is among the lowest in Europe.36 This finding cannot be dismissed as being established on subjective grounds alone, since it, unfortunately, finds corroboration in the country’s record before the European Court of Human Rights (ECtHR). By the end of 2017 Slovenia had been convicted 353 times before the ECtHR, of which more than 263 cases concerned a violation of the right to a speedy trial and the right to an effective remedy.37 In its landmark Lukenda pilot-judgment the ECtHR went as far as to proclaim that the Slovenian violation of the right to a trial within a reasonable time is ‘a systemic problem that has resulted from inadequate legislation and inefficiency in the

30 Mitja Ferenc, Huda Jama (Grave Pit) Coal Mine Massacre (Ljubljana, Znanstvena založba Filozofske fakultete, 2013). 31 Matej Avbelj, ‘Failed Democracy: The Slovenian Patria Case – (Non) Law in Context’, (2014) No 61/62 Dignitas, 59. 32 Ibid. 33 See also Matej Avbelj, Nepravna država. Zapisi o zadevi Patria. (2011–2015) [The Wrong State. Records on the Patria Case (2011–2015)] (Ljubljana, GV Založba, 2015). 34 For the argument that follows see Matej Avbelj, ‘Transformation in the Eye of the Beholder’, in M Bobek (ed), Central European Judges under the European Influence: The Transformative Power of the EU Revisited (Oxford, Hart, 2015) 275. 35 See the report at www.coe.int/t/dghl/cooperation/cepej/evaluation/2012/Rapport_en.pdf. 36 See the report at www.europeansocialsurvey.org/docs/findings/ESS5_toplines_issue_1_trust_in_justice.pdf, 9–10. 37 See the report at www.echr.coe.int/Documents/Stats_violation_1959_2014_ENG.pdf.

684  Matej Avbelj administration of justice. The problem continues to present a danger affecting every person seeking judicial protection of their rights.’38 Slovenia, despite its tiny population of 2 million, also belongs among the nine highest case-count states, which are those with more than 1,700 cases pending before the ECtHR,39 and also falls into a group of nine states with the highest number of applications per capita.40 This problem is exacerbated further, as the judiciary has succumbed to complacency, refusing to admit, let alone address, its systemic problems. The judiciary keeps downplaying its shortcomings, or, even more frequently, it shifts the responsibility onto someone else’s shoulders, be it onto the legislative and/or executive branch, the allegedly too litigious population, the opposition, or even individual judges who have publicly voiced criticism regarding the system’s malfunctioning.41 While all the initially broadly conceived reform attempts, including those that were a direct response to the ECtHR rulings, have, as a rule, always been diluted in Slovenia, the judges have most determinedly publicly stood up for one thing: their salaries. We can add to this that the rule of law in Slovenia is applied selectively, sheltering from the reach of justice those who are close to the formal and informal centres of power, which means that not even equality before the law is respected.42 All of this can be explained against the backdrop of Slovenia’s peculiar (non-)break-up with its communist past. In contrast to other CEE countries, Slovenia basically refused to carry out any lustration or, at least, to change its elites. The elite retention rate, the people who were in key societal positions in both pre- and post-communist times, was exceptionally high.43 The judiciary was no exception in that regard. While the legislation provided a narrowly circumscribed possibility not to re-appoint judges who partook in the trials whose rulings violated fundamental human rights and freedoms, this provision was rarely used and, according to the Constitutional Court, ought not to be regarded as an element of a ‘real’ lustration.44 This meant that, similar to the general situation in the country, the retention of the judicial elite was high too. Bugarič is, therefore, justified in claiming that Slovenian rule of law institutions, including the courts, ‘have been deeply politicised by the cadre from the old political nomenclature’.45 As a result, Slovenia combines a modern, western liberal democratic constitution, with an old communist mindset and practice, which is due to the personal continuity in the main institutions of the state still dominant in many social sub-systems, the legal system included.

38 ECtHR 06.10.2005, Application No 23032/02 (Lukenda v Slovenia) ECLI:CE:ECHR:2005:1006JUD002303202, para 93. 39 These countries are Bulgaria, Georgia, Italy, Romania, Russia, Serbia, Ukraine, United Kingdom, and Slovenia (www. echr.coe.int/Documents/Stats_analysis_2013_ENG.pdf, 7/60). 40 These countries are Bosnia and Herzegovina, Croatia, Moldova, Montenegro, North Macedonia, Romania, Serbia, Slovenia and Ukraine. 41 See, Matej Avbelj, ‘O sodnikih in vladavini prava’ [‘About Judges and the Rule of Law’], (2007) 20, Pravna praksa 26; Matej Avbelj, ‘Odnos nosilcev oblasti do prava’ [‘The Attitude of the Authorities to Law’], (2009) 33 Pravna praksa, 17. 42 This has, consequently, sparked a public debate whether Slovenia has two regimes of the rule of law: one for the more privileged class, the other for the less privileged. 43 Frane Adam and Matevž Tomšič, ‘Transition Elites: Catalysts of Social Innovation or Rent Seekers’, (2000) 32 Družboslovne razprave, 138. 44 USRS 15.07.1994, Case U-I-83/94, ECLI:SI:USRS:1994:U.I.83.94: ‘The disputed provision could represent a lustration element from the point of view of substantial demand, that the judges whose judicial office has been performed in violation of human rights as the result of submission of judicial office to politics shall not discharge judicial functions in a democratic system. But it does not imply lustration either in the sense of exclusion from an office or in the sense of exclusion from or access to any other public office, except judicial; neither does it imply lustration in the sense of its complete objectivity based on affiliation.’ 45 Bojan Bugarič, ‘Crisis of Constitutional Democracy in Post-Communist Europe: “Lands in-between” democracy and authoritarianism’, (2015) 13 ICON, 219.

Slovenia  685

III.  Constitutional Foundations of EU Membership and Closely Related Instruments Slovenia became a full member of the EU on 1 May 2004. The country’s membership was decided in a referendum, in which an overwhelming majority voted in favour of accession to the EU.46 The referendum was convened because joining the EU was anticipated to bring about far-reaching and transformative effects on the character of Slovenian statehood. By analogy with the plebiscite on independence, this decision also had to be taken by the Slovenian people. The legal basis for that was explicitly included in the Constitution.47

A.  Four EU Law Amendments to the Constitution The Constitution has been amended three times to facilitate Slovenian entry into the EU. The first amendment came already during the EU membership negotiations. The then Article 68 of the Constitution excluded aliens, including EU citizens, from acquiring immovable property in Slovenia, except in cases of inheritance under conditions of reciprocity. Following the amendment, aliens are allowed to acquire ownership rights to real estate under conditions provided for by law or a treaty ratified by the National Assembly.48 The second amendment concerned Article 47 of the Constitution. Its objective was to enable the surrender of Slovenian citizens to other Member States. The Constitution originally contained a general prohibition of extradition of Slovenian citizens to a foreign country and aliens could be extradited only in cases covered by treaties binding on Slovenia. The amended article now provides that no citizen of Slovenia may be extradited or surrendered unless such an obligation to extradite or surrender arises from an Article 3a treaty on the basis of which Slovenia has transferred the exercise of a part of its sovereign rights to an international organisation.49 Article 3a is the main constitutional amendment for the purposes of Slovenian EU membership. It grants a special status to EU law in the Slovenian constitutional order, yet without explicitly referring to it. It reads as follows: Pursuant to a treaty ratified by the National Assembly by a two-thirds majority vote of all deputies, Slovenia may transfer the exercise of part of its sovereign rights to international organisations which are based on respect for human rights and fundamental freedoms, democracy and the principles of the rule of law … Before ratifying a treaty referred to in the preceding paragraph, the National Assembly may call a referendum. A proposal shall pass at the referendum if a majority of voters who have cast valid votes vote in favour. The National Assembly is bound by the result of such referendum. If the referendum has been held, a referendum regarding the law on the ratification of the treaty concerned may not be called. Legal acts and decisions adopted within international organisations to which Slovenia has transferred the exercise of part of its sovereign rights shall be applied in Slovenia in accordance with the legal regulation of these organisations.

46 89.64% of the voters voted in favour of accession to the EU (www.dvk-rs.si/files/files/porocilo-o-referendumu-EU. pdf). 47 CRSI Art 3a. 48 CRSI Art 68. 49 CRSI Art 47.

686  Matej Avbelj In procedures for the adoption of legal acts and decisions in international organisations to which Slovenia has transferred the exercise of part of its sovereign rights, the Government shall promptly inform the National Assembly of proposals for such acts and decisions as well as of its own activities. The National Assembly may adopt positions thereon, which the Government shall take into consideration in its activities. The relationship between the National Assembly and the Government arising from this paragraph shall be regulated in detail by a law adopted by a two-thirds majority vote of deputies present.

Drafting this constitutional amendment confronted a dilemma of whether to opt for a minimalist abstract approach, blending EU law with general international law, or to endorse a more casuistic, concrete approach in the form of an explicit EU article in the Constitution.50 At the end of the day, the abstract and purportedly more neutral approach prevailed. The drafters of the amendment wanted to eschew the theoretical difficulties raised by EU law, in particular by its (contested) supranational nature. Due to the evolutionary character of integration, they were also uncertain as to which terminology to use – is it the European Communities or the EU?51 Moreover, at the time of drafting the amendment, Slovenian membership in the EU was not yet certain, which was another reason to go for a more circumspect approach to the Constitution-amending process.52 Nevertheless, the approach taken has been subject to strong critiques. It has been alleged that Slovenia’s Constitution suffers from an EU deficit.53 By subsuming the EU under international organisations, the Constitution fails to do justice to the actual legal and political character of the integration.54 Furthermore, reliance on the constitutions of some old Member States that have not been amended since the late 1950s and are, therefore, increasingly out of touch with the present state of the EU, fails to persuade either. To the contrary, most of the constitutional texts in the EU Member States have by now moved away from an abstract to a concrete constitutional approach, dedicating a special title or at least a few provisions to the EU.55 Irrespective of these drafting dilemmas, Article 3a is also important because it provides for a close parliamentary control of the national executive branch in its activities on the supranational level, especially in the decision-making process of the EU Council. The details of this parliamentary control are further set out in the Act on Cooperation between the National Assembly and the Government in EU Affairs, and will be described and explained below. Finally, in order to fulfil the requirements stemming from the Fiscal Compact and to constitutionalise the so-called golden fiscal rule, in May 2013 the Parliament amended Article 148 of the Constitution concerning the budget. The new Article 148 now requires that all revenues and expenditures for the financing of public spending must be included in the budgets of the state.56 They must be balanced in the medium-term without borrowing, or revenues must exceed expenditures. Temporary deviation from this principle is only allowed when exceptional circumstances

50 Republika Slovenija Državni Zbor, Ustavne razprave 2001–2003 [Constitutional Debates 2001–2003], Vol III.1 (Ljubljana, Državni zbor RS, 2004) 25. As a compromise also a combined approach was suggested, Šturm, Komentar (n 2) 74 f. See also, Ciril Ribičič, Mozaik ustavnih sprememb: od slovenske ustavnosti k evropski [The mosaic of constitutional reform: from Slovenian towards European constitutionality], (Ljubljana, GV Založba, 2003) 22; Ivan Kristan, ‘Sprememba Ustave RS: Ustava: mednarodni del: evropski člen’ [‘Amendment to the Constitution of the Republic of Slovenia. Constitution: International Section: European Article’], (2001) 6 Podjetje in delo, 867. 51 For an overview, see Matej Avbelj, ‘Slovensko ustavno pravo v odnosu do prava EU’ [‘Slovenian Constitutional Law in Relation to EU Law’], in I Kaučič (ed), Dvajset let Ustave Republike Slovenije: pomen ustavnosti in ustavna demokracija (Ljubljan,: Ustavno sodišče RS in Pravna fakulteta, 2012) 341. 52 Ibid. 53 Ribičič, Mozaik (n 50) 29 f. 54 Avbelj, ‘Slovensko ustavno pravo’ (n 51). 55 Monica Claes, ‘Constitutionalizing Europe at its Source: The “European Clauses” in the National Constitutions: Evolution and Typology’, (2005) 24 Yearbook of European Law, 81. 56 CRSI Art 142/1.

Slovenia  687 affect the state.57 However, actual implementation of the golden fiscal rule, now laid down in the Constitution, is left to a statute to be adopted by the National Assembly by a two-thirds majority vote of all deputies. The statute shall define the manner and the time frame of the implementation of the golden fiscal rule, the criteria for determining exceptional circumstances, and the course of action when they arise.58 The statute mandated by the Constitution should have been adopted in a year after the passing of the constitutional amendment. However, due to the difficulty of gathering the required qualified parliamentary majority, it took more than two years to pass the statute in June 2015. Slovenia was, thus, in an unconstitutional situation, in which the golden fiscal rule existed on the books, but was not operational in practice.

B.  Continued Relevance of (General) Provisions on International Law and/or Treaties Despite labelling EU law as international law and subsuming the EU under international organisations, the Slovenian Constitution has clearly separated EU law from international law. As a dualist country, Slovenia needs to build a bridge for the incorporation of international law in its legal order. This bridge is provided for in Article 8 of the Constitution, which, however, was found as an inadequate legal basis for the Slovenia’s accession to the EU. As a result and as described above a special article 3a had to be included in the Constitution to enable EU law to produce effects in Slovenia. Nevertheless, in purely formal terms, the EU founding treaties, such as the Treaty of Lisbon, remain international law treaties. This has been confirmed by the Constitutional Court, which ruled59 that in accordance with Article 8 and Article 153/II of the Constitution, statutes must be in conformity with the Treaty of Lisbon. The Court is, thus, competent to review the compliance of national laws with the Treaty of Lisbon.60 In the same vein, however, the Court is also competent to conduct upon the request of the National Assembly ex-ante constitutionality review of the Treaty of Lisbon. No such review, however, has been requested so far.61

IV.  Constitutional Limits to EU-Integration Article 3a of the Slovenian Constitution defines the material and procedural conditions for the Slovenian EU membership and subjects the latter to substantive constitutional limitations. In material terms, Slovenia may only transfer to the EU ‘the exercise of a part of its sovereign rights’. Thus, not even a part of, let alone the whole of, Slovenian sovereignty is transferrable and, therefore, remains with the Slovenian people. This provision is clearly influenced by the negative experience of Slovenia in the Yugoslav federation. The preamble to the Basic Constitutional Charter of Independence (BCC) states in an identical manner that Slovenia was already an independent state in Yugoslavia which only exercised parts of its sovereign rights through the

57 CRSI Art 148/2. 58 CRSI Art 148/3. 59 USRS 18.10.2012, Case U-I-17/11, ECLI:SI:USRS:2012:U.I.17.11. 60 Ibid. 61 See also Jadranka Sovdat, ‘The Constitutional Court of the Republic of Slovenia and European Union Law’, (2013) 13 Hrvatska i komparativna javna uprava, 900.

688  Matej Avbelj institutions of the federation.62 The same circumspect attitude once held in a Yugoslav federation is nowadays constitutionally applied against the EU. In procedural terms, the exercise of a part of its sovereign rights is further subject to three conditions. It can only happen by an international treaty, which must be ratified by the National Assembly with a qualified two-thirds majority of the representatives. Finally, a referendum may also be convened, requiring the majority of the voters who participated in the referendum vote in favour. The substantive constitutional limitations to the Slovenian membership in the EU are divided in three groups and are clearly influenced by the jurisprudence of the German Constitutional Court. For EU law to apply in Slovenia in accordance with its own rules and principles, it must respect human rights and fundamental freedoms; democracy and the principles of the rule of law; and it must be adopted strictly within the competences conferred on the EU.63 These substantive limitations have, however, so far remained on paper only. The Constitutional Court explicitly refrained from adopting a position whether there are certain instances in which the Slovenian Constitution requires the disapplication of EU law.64 Other than these three substantive constitutional limits to EU law in Slovenia, the Constitution does not contain any a priori limits to the development of the process of European integration. There are, thus, no constitutional provisions on the formal limits to (further) transfer of powers to the EU through Treaty amendments, such as ‘core competences’, ‘non-transferable’ constitutional identities, which ‘must’ remain with the national parliament etc. Unlike the constitutions of some EU Member States, the Slovenian Constitution does not contain an explicit eternal clause (Ewigkeitsklausel), stipulating the so-called irreducible epistemic core of the national constitutional order. This stands for formal and substantive elements that any legal order has to preserve intact in order to be an autonomous and independent legal order.65 However, in the Decision of the Constitutional Court Rm-1/09,66 the Court invoked the BCC as an applicable constitutional act and as such, a permanent and inexhaustible constitutional source of the statehood of the Republic of Slovenia. The BCC is a founding document of the Slovenian state that came into being on the basis of a plebiscite in the absence of a new constitution. The latter was adopted six months after independence. As the Court explained, the BCC was adopted on 25 June 1991 as the fundamental constituting state act of the Republic of Slovenia. With its adoption, the Republic of Slovenia definitively broke its ties with Yugoslavia and established itself as a sovereign state. The BCC was, admittedly, invoked in a case unrelated to European integration. However, the Court’s use of the language of ‘permanent and inexhaustible source of the statehood of the Republic of Slovenia’ signals that the Court might assert the same doctrine also against EU law, if the need arises. However, no such EU law-related case has been brought to the Court yet. As will be explained in further detail below, neither the Constitutional Court nor any other Slovenian court has claimed competence to review the constitutionality of secondary EU law.

62 ‘Considering the fact that the Republic of Slovenia has been a state under the constitutional order hitherto in force and has exercised only a part of its sovereign rights within the Socialist Federal Republic of Yugoslavia.’ 63 CRSI Art 3a. 64 USRS, 17.12.2012, Case U-II-1/12, U-II-2/12, ECLI:SI:USRS:2012:U.II.1.12. 65 Neil Walker, ‘Late Sovereignty in the European Union’, in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 3, 28. 66 USRS 18.03.2010, Decision Rm-1/09, ECLI:SI:USRS:2010:Rm.1.09 on the Constitutionality of the Arbitration Agreement between Slovenia and Croatia for the determination of the disputed sea and land borders between the two countries.

Slovenia  689 Nonetheless, the Constitutional Court views EU law as treaty-based law, serving as a point of reference (a so-called upper premise) for reviewing the validity of national statutes and by-laws. The flight to international law during the EU’s attempts to resolve the economic crisis, resulting most notably in the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union (TSCG or Fiscal Compact) and in the Treaty Establishing the European Stability Mechanism is not covered by the EU law provisions of the Slovenian Constitution. As these are international law instruments, they have the status of international law in Slovenia. In practice, this means that they are hierarchically superior to statutes, but inferior to the Constitution and they are not endowed with a constitutionally privileged status of supranational law. Nevertheless, in December 2012 the Constitutional Court banned a referendum on a statute, which was also indirectly intended to give full effect to the Fiscal Compact.67 In this case the Constitutional Court was willing to override the constitutional right to a referendum by giving precedence to a financial instrument of international law. This could be interpreted as meaning that under financial duress, such as the one Slovenia was under at the peak of the economic crisis in 2012, not only the will of the people but even the Constitution as such could bend to the demands of international law. In other words, in such exceptional circumstances, it is not the Constitution, which sets the limits to international or EU law, but it is the other way around.

V.  Constitutional Rules and/or Practice on Implementing EU Law A.  The Role of the National Assembly It has already been mentioned that Article 3a of the Constitution provides for a close parliamentary control of the national executive branch in its activities on the supranational level, especially in the decision-making process at the EU Council. The parliamentary scrutiny of EU affairs in the National Assembly is the mandate-based model of scrutiny, inspired by the Danish model.68 It is governed by Article 3a of the Constitution, the Act on Cooperation between the National Assembly and the Government regarding EU Affairs, and section 2a of Part IV of the National Assembly Rules of Procedure. The National Assembly participates in the formulation of positions of the Republic of Slovenia in EU affairs, as a plenum as well as with its two specialised committees: the Committee for European Affairs and the Committee for External Affairs.69 The Committee for European Affairs discusses draft positions within the time limits required by the work within EU institutions. The National Assembly also discusses amendments to the founding Treaties.70 At least once a year the National Assembly discusses the state of affairs in the EU and the position of the Republic of Slovenia therein, and adopts positions on the political guidelines for Slovenia’s activity in EU institutions.71 The Government promptly informs the National Assembly of EU affairs that fall

67 USRS U-II-1/12, U-II-2/12 (n 64). 68 See, eg, Claudia Hefftler, Christine Neuhold, Olivier Rozenberg and Julie Smith, The Palgrave Handbook of National Parliaments and the European Union (London, Palgrave Macmillan, 2015). 69 The following draws closely on Miro Cerar, ‘The Role of national parliaments in the European Union’, in GC Rodríguez Iglesias and L Ortiz Blanco (eds), The role of national parliaments in the European Union. Proceedings of the FIDE XXIV Congress, Vol I (Madrid, Servicio de Publicaciones de la Facultad de Derecho de la Universidad Complutense de Madrid, 2010) 411. 70 Act on Cooperation between the National Assembly and the Government regarding EU Affairs (OJ 34/04). 71 Ibid Art 5.

690  Matej Avbelj within its competence and reports on its executive decisions and related activities within EU institutions. The Government also informs the National Assembly about other documents that are relevant for the exercise of its constitutional powers and concern the political and programme aspects of the activity of the EU. The National Assembly may also discuss other EU affairs on a Government’s proposal or on its own initiative.72 The Government sets out its positions to the Committee for European Affairs prior to meetings of any of the EU Council formations. The Committee for European Affairs takes note of these positions, normally without making any amendments. In the event that the Government sends a draft position regarding a legislative proposal, an opinion on such a draft position is first delivered by the working body responsible (depending on the topic covered by the proposal). The working body responsible sends its written opinion to the Committee for EU Affairs, which takes it into account and finally votes on the position of the Republic of Slovenia, thus, providing the Government with a mandate for the negotiations.73 If the Government is unable to fully secure Slovenia’s positions, or when doing so is not feasible or would not benefit Slovenia, it may decide otherwise. In so doing, the Government immediately informs the National Assembly thereof in writing and states the circumstances that have led to such a decision.74 The Government must do so in accordance with Article 10 of the Act on Cooperation between the National Assembly and the Government regarding EU Affairs. In practice, this means that the Committee for EU Affairs discusses the matter and takes a position thereon usually at its next meeting. In such a case, the Government’s responsibility is political. The Act on Cooperation between the National Assembly and the Government in EU Affairs has been challenged before the Constitutional Court for excluding the State Council from the abovementioned procedure. The State Council is an institution representing the interest of employers, employees, civil society, and municipalities. Its constitutional status is ambiguous and disputed. While some argue that it is a full-fledged second chamber of the Slovenian Parliament, others claim that the Slovenian Parliament is unicameral, the National Assembly being its only chamber.75 The latter view also appears to find confirmation in the Constitutional Court’s jurisprudence. The Court has upheld the exclusion of the State Council from the oversight of the Government in EU affairs. It has argued that Article 3a of the Constitution neither provides for such an involvement, nor is this required by the EU Treaties and their Protocol 2, which leave this question to the national constitutional autonomy.76 The Committee for European Affairs of the National Assembly has been quite active. Besides exercising control over the Government, it has also voted in favour of issuing a yellow card to the European Commission for its failure to respect the principle of subsidiarity in its proposal for the establishment of the European Public Prosecutor.77

B.  Techniques of Implementing Secondary Legislation, Especially of Directives: Laws, Administrative Regulations, Others In 2004 when Slovenia joined the EU, the Government’s Office of Legislation issued detailed legal drafting guidelines, also to ensure coherence, legality, and effectiveness in the execution and 72 Ibid Art 4/3. 73 Ibid Art 10/1. 74 Ibid, Art 10/2. 75 For a good overview, see Dušan Štrus, Drugi dom parlamenta [The Second House of Parliament] (Ljubljana, Fakulteta za upravo, 2011). 76 USRS U-I-17/11 (n 59). 77 Decision of the Committee for European Affairs of the Slovenian National Assembly 008-09/13-7, 25.10.2013.

Slovenia  691 implementation of the EU rules.78 The guidelines were subsequently amended and consolidated in 2008. The Office identified three basic approaches to the implementation of EU directives, which can be used either independently or in combination with each other. The first approach requires that a directive is implemented in accordance with the national legal drafting rules, so that it is implemented fully and in its entirety, to ensure its full effectiveness.79 The second approach takes the form of a verbatim copying of the text of a directive in national legal acts. This approach is permitted and suitable in case of technically detailed provisions, which often contain lists, tables, equations or numbers. In such cases the provisions of directives require no or little interpretation and leave no scope for a different regulatory approach.80 Finally, in exceptional cases, again concerning more technical legal regulations, it is also permissible to implement a directive by merely including a reference to it in a statute. However, this third type of implementation is excluded when a directive is intended to confer rights or duties on individuals.81 Furthermore, the guidelines provide that in accordance with the principle of legality the Slovenian Acts implementing EU law must have a Slovenian statutory legal basis.82 EU rules cannot serve as a legal basis for the adoption of Slovenian by-laws. To facilitate the implementation of EU law in Slovenia, the Government of the Republic of Slovenia Act contains a general authorisation clause according to which the Slovenian Government can issue regulations and other acts to implement EU law.83 A similar provision can be found in the State Administration Act, which also generally and in advance authorises individual ministers to issue by-laws for the implementation of EU rules.

C.  Enforcement Through Courts As we have seen, EU law enjoys special constitutional status in Slovenia. Nevertheless, especially in the first five years of Slovenia’s EU membership, EU law-related cases before the Constitutional Court, the ordinary courts, and national and local administrative organs have remained very few. Slovenia did even worse in terms of applying EU law, as did most of the new Member States. For example, for the first five years, Slovenia occupied the last place, in both absolute and relative terms, in the number of referrals for a preliminary ruling made to the Court of Justice of the European Union (CJEU).84 Several factors could have contributed to this. The first has to do with a relatively short period of EU membership. Law must be internalised, its addressees must be socialised into it, and this takes time. However, in Slovenia this might take even longer due to the remnants of its totalitarian past. Slovenia is a post-communist state where no lustration took place. The legal actors, in particular judges, who were in office during communist times, are still in office, enjoying their permanent tenure. These people, of course, must have big difficulties in switching paradigms: from servicing the ruling communist party, to servicing the rule of law, democracy, and a strange phenomenon called EU law. This requires personal intellectual renewal, which is – with a degree of sociological realism in mind – very hard to achieve.

78 Služba Vlade Republike Slovenije za zakonodajo, Nomotehnične smernice [Government Office for Legislation, Technical Guidelines] (Ljubljana, Služba Vlade Republike Slovenije za zakonodajo, 2008). 79 Ibid, 175. 80 Ibid. 81 Ibid, 176. 82 Ibid, 178 f. 83 Government of the Republic of Slovenia Act Art 21/7. 84 http://curia.europa.eu/jcms/upload/docs/application/pdf/2015-04/en_ecj_annual_report_2014_pr1.pdf.

692  Matej Avbelj Furthermore, knowledge of EU law in Slovenia is generally still very poor. The present active generation of lawyers was educated during communist times and in the socialist legal paradigm, with little or no training in EU law. The lack of EU knowledge was openly admitted even by the group of experts charged with drafting the above-mentioned Slovenian EU constitutional clause.85 Some of the experts called for the inclusion of foreign learned advisors, but their pleas were not heard, which resulted in a strangely worded article that assimilates EU law with international law.86 In other words, this lack of knowledge among (the so-called) experts and academics was passed on to the practising lawyers, which resulted in a self-perpetuating non-relevance of EU law. Lack of knowledge leads to a lack of interest in a particular topic, and the lack of interest translates into non-relevance of a particular issue. Such has been the fate of EU law. Another reason for a difficult reception of EU law in Slovenia was also a strong theoretical and practical adherence to statutory-based legal positivism. According to this mind-set, law is (almost) synonymous with a statute. What is in a statute is the law. Other sources of law, including the Constitution, are much less relevant in practice. It has long been part of the official academic and practical doctrine in Slovenia, that a judge-made law is not formally binding, it is not of a precedential value but is merely persuasive authority. National judicial decisions, especially of ordinary courts, are, therefore, rarely taken into account. As a result, Slovenian lawyers are simply not used to think about the law in this inductive, casuistic, case law perspective. Of course, given the role of the CJEU in the evolution of EU law, this has been a great obstacle for the correct and up-to-date application of EU law in Slovenia. The next EU law-impeding factor in Slovenia might be related to the aversion to foreign law, which is paradoxically combined with a concurrent unreflective acceptance and overestimation of its authority. The first part of the phenomenon is, of course, not uniquely Slovenian. National legal actors are in general much more comfortable with the use of the national, rather than foreign law. However, again in Slovenia this problem might be more acute, exacerbated due to the specific national mind-set, characterised by the lack of self-confidence and consequent (unfounded) aggrandisement of foreign authorities. This could be nicely illustrated through the, of course publicly unstated, reasons for which the Slovenian judges were long reluctant to refer a case for a preliminary ruling to the CJEU. They were simply embarrassed to ask a question of the foreign court, as they believed that sending a preliminary reference would demonstrate their ignorance and would be, thus, ridiculed by the CJEU judges. Finally, a very slow emergence of EU law in Slovenia could be also ascribed to the country’s economic isolationism. Unlike the other CEE countries, Slovenia has had a very low share of direct foreign investment. As is well known, it has been the transnational companies which benefited most from EU law, in particular from its single market. They have been the most frequent EU law litigators. Consequently, their relative absence in Slovenia can also explain why there have been fewer EU law-related cases litigated here than in other CEE countries. In short, if there are fewer subjects interested in pursuing their rights derived from EU law, there will also be fewer EU law-related cases.

D.  EU Law Jurisprudence of the Constitutional Court In the midst of this initially largely reserved attitude towards EU law, it was again the Constitutional Court which acted as a path-breaker. Despite the small number of EU law-related cases that has so

85 RSDZ, 86 Ibid.

Ustavne razprave (n 50) 29.

Slovenia  693 far reached it, the Court has managed to lay down, albeit in a more implicit and sometimes very modest and abstract manner, the basic principles governing the relationship between national and EU law. Accordingly, the Court has ruled that the legal basis for the application of EU law in Slovenia is Article 3a of the Constitution, and not EU law itself. The latter, however, has been, in virtue of Article 3a, constitutionalised in Slovenia. It ranks as a constitutional value in and of itself.87 EU law in toto – primary and secondary rules as well as the case law of the CJEU – is, thus, binding on the Constitutional Court, which must apply it in accordance with the standards, rules, and principles that EU law prescribes.88 Furthermore, the Constitutional Court incorporated in the Slovenian Constitution the structural principles of EU law, as laid down in the Treaties and developed in the jurisprudence of the CJEU, endowing them with the force of constitutional provisions.89 As such, they bind not only the Constitutional Court, but all organs of the state.90 The Constitutional Court has, however, put a special emphasis on the EU law-related duties of the ordinary courts as they derive from the Slovenian Constitution. In its landmark decision Up-1056/11,91 it found the unjustified refusal of the Supreme Court to make a reference for a preliminary ruling unconstitutional, amounting to a violation of the right to a lawful judge. The Court found that the Supreme Court is, within the meaning of Article 267 TFEU, the court of last resort in Slovenia and is, as such, in principle, bound to make a reference to the CJEU. The latter is, pursuant to the Slovenian Constitution, conceived of as a lawful judge, to which an individual has the right to access when the conditions for its exclusive jurisdiction under Article 267 TFEU are met. It is a duty of the Supreme Court to either make a reference for a preliminary ruling, or to justify, why in its own view these conditions are not met and a reference is not necessary.92 The Constitutional Court does not control whether the national courts apply EU law correctly, but it reviews that in the application of EU law the judicial protection is ensured by a statutory court. The Constitutional Court has additionally stressed that the decisions of CJEU are a binding source of law, which has to be followed by national courts.93 The latter have to explain how they have implemented the rulings of the CJEU. Failure to do so can amount to a violation of Article 22 of the Constitution, which ensures the equal protection of rights.94 The gist of the Constitutional Court case law is that the central role in the correct application of EU law in Slovenia has to be played by the ordinary courts and the Supreme Court as the highest judicial authority in Slovenia. On this basis the Constitutional Court devised a special division of labour with the ordinary courts that plays a very important role in particular in the protection of rights stemming from the EU Charter of Fundamental Rights (CFR). By now the Constitutional Court has referred to the CFR several times, but only in passing,95 without developing any significant case law on it. Like the Treaty of Lisbon, the Charter too is part of primary EU law and is considered a treaty in the Slovenian legal order, providing a legal standard against which the Constitutional Court and ordinary courts can review national legislation.

87 USRS U-II-1/12, U-II-2/12 (n 64). 88 USRS 03.07.2014, Decision U-I-65/13, ECLI:SI:USRS:2014:U.I.65.13. 89 Ibid. 90 Ibid. 91 USRS 21.11.2013, Case Up-1056/11, ECLI:SI:USRS:2013:Up.1056.11; which was later confirmed in USRS 12.03.2015, Case Up-797/14, ECLI:SI:USRS:2015:Up.797.14. 92 Ibid. 93 USRS 21.01.2014, Case Up-150/13, ECLI:SI:USRS:2014:Up.150/13. 94 Ibid. 95 USRS 13.11.2008, Case U-I-146/07, ECLI:SI:USRS:2008:U.I.146.07.

694  Matej Avbelj In the constitutional complaint procedure the Constitutional Court has long insisted on the exhaustion of all national remedies.96 For a constitutional complaint to be admissible, the complainant has to prove that he has argued the violation of a particular human right, including those enshrined in the CFR, on all instances of the judicial procedure, including the Supreme Court. The latter is the final authority among ordinary courts to rule on human rights protection. It is, thus, anticipated that the Constitutional Court will be seized with cases involving the violation of rights contained in the Charter only in those rare and exceptional circumstances, in which the Supreme Court will have failed to discharge its tasks under EU law and, hence, also under national constitutional law.97 This, however, does not mean that EU law is considered to have an infra-constitutional status98 and that the Constitutional Court is absolved from following it. The Constitutional Court has not just explicitly declared itself to be bound by EU law,99 but it has also proven this in practice by actively engaging in judicial dialogue with the CJEU. In Decision U-I-295/13 the Constitutional Court for the first time made a reference for a preliminary ruling to the CJEU. That case concerned the constitutionality of a banking law that was adopted as a crisis-management mechanism implementing the instructions (soft law) of the European Commission on the bailing-out of previously private banks and consequent cancelling of shares and subordinate bonds. The CJEU found the applicable EU provisions valid and the Slovenian Constitutional Court, consequently, also approved the national implementing measures.100 This brief overview of the practice of the Constitutional Court demonstrates that it has adopted an explicitly EU law-friendly attitude. It has often insisted on the centrality of the respect of the principle of loyalty.101 The Court has, thus, not only declaratorily subscribed to all the structural principles of EU law, including its primacy, but has also applied them in its own practice and mandated the same in the practice of ordinary courts. In line with its rather conservative approach to non-national constitutional issues, it has refrained from rushing to conclusions on the way conflicts between constitutional law and EU law are to be resolved. With regard to national legislation and by-laws it has left no doubt that they have to comply with EU law, as they are, pursuant to the Constitution, hierarchically inferior to EU law. In contrast, the Constitutional Court has been more determinate and expansive as far as constitutionality review of national implementing acts is concerned. Already in its first EU law decision in the JATA case, the Court availed itself of the Zuckerfabrik exception and suspended the application of a national act implementing a contested EU directive.102 Since then the Constitutional Court has consistently held that it remains competent to review the constitutionality of national implementing acts.103 In case U-I-37/10104 it even invalidated the national by-law, implementing the EU Natura 2000 programme, but suspended the effects of its ruling for one year to preserve the full effectiveness of EU law in Slovenia. The Constitutional Court also did not hesitate to stress that, given the lack of EU law provisions to the contrary, the way in

96 Sovdat, ‘Constitutional Court (n 61) 904. 97 Ibidm. 98 As it was, for example, a constitutional doctrine in Spain, see Rafael-Leal Arcas, ‘The Reception of European Community Law in Spain’, (2005) 1 Hanse Law Review, 18. 99 USRS U-I-146/07 (n 95). 100 USRS 19.10.2014, Case U-I-295/13, ECLI:SI:USRS:2014:U.I.295.13; CJEU 30.09.2016, C-526/14 (Kotnik and Others) ECLI:EU:C:2016:767. 101 USRS 05.03.2009, Case Up-2012/08, ECLI:SI:USRS:2009:Up.2012.08; USRS U-II-1/12, U-II-2/12 (n 64). 102 USRS 07.02.2007, Case U-I-113/04, ECLI:SI:USRS:2007:U.I.113.04. 103 Ibid. 104 USRS 18.04.2013, Case U-I-37/10, ECLI:SI:USRS:2013:U.I.37.10.

Slovenia  695 which EU law is implemented in Slovenia, including the type of national legal act and the level of authority (national or local), remains part of the national constitutional autonomy.105 In any case, the implementation of EU law remains subject to Slovenian constitutional standards, in particular to the principle of legality.106 While the Constitutional Court remains competent to review national legislation implementing EU law, it has declined the competence to review the compliance of national laws with EU directives107 and regulations.108 These acts are not treaties within the meaning of Article 8 of the Constitution. Instead, they belong to another legal order and, therefore, cannot be used as a criterion of legality of Slovenian legal provisions.

E.  EU Law Jurisprudence of Ordinary Courts The first five years of the Slovenian membership in the EU were marked by the glaring discrepancy between EU law on the books and EU law in action. Only a handful of cases before ordinary courts related, often indirectly, to EU law. However, since 2010 the trend has been changing and ordinary courts have been applying EU law much more often. Administrative courts have decided the highest number of EU law-related cases, which concerned matters of international protection (asylum) or protection of constitutional rights. Administrative courts either ground their decisions on EU law or they at least refer to it in their reasoning. EU law is also widely present in the decisions of civil courts in the field of mutual recognition of judgments in civil matters. Other EU law-relevant cases have concerned restitution of property confiscated after WWII, where Slovenian courts have consistently ruled that these cases ratione temporis fall outside the scope of EU law.109 Cases related to value added tax, customs, European Arrest Warrant in criminal cases, and the determination of citizenship have also increasingly raised EU law issues, however, without major controversies and potential case-law developments. Slovenia was also one of the last countries among the new Member States to refer a case for a preliminary ruling to CJEU. This happened only in 2009, five years after accession. The reasons for that have been explained above, but to demonstrate what a ground-breaking occasion that was, it needs to be mentioned that the referring judge was even awarded a special law prize, which is awarded annually by the Slovenian Lawyers’ Association to an individual who has contributed most to European law’s presence in Slovenia and Slovenia’s presence in Europe. However, since then Slovenian courts, in particular the Supreme Court, have requested many more preliminary rulings. The underlying cases covered the fields of agriculture, taxes, insurance law, damages, family law and enforcement of judgments from another Member State.

VI.  Resulting Relationship between EU Law and National Law The question of the relationship between EU law and Slovenian law ultimately remains open. As shown above, the Constitutional Court has given an increasingly detailed guidance on the

105 Ibid and USRS 16.12.2009, Case U-I-265/07, ECLI:SI:USRS:2009:U.I.265.07. 106 Ibid. 107 USRS 09.02.2006, Case U-I-32/04, ECLI:SI:USRS:2006:U.I.32.04. 108 USRS 08.07.2004, Case Up-328/04 and U-I-186/04, ECLI:SI:USRS:2004:Up.328.04. 109 Matej Avbelj, ‘Slovenska sodišča in postopek za predhodno odločanje’ [‘Slovenian Courts and the Procedure of Preliminary Ruling’] (2009) 28 Pravna praksa, 9.

696  Matej Avbelj interpretation and application of the Slovenian conduit provision to the EU (ie Article 3a of the Constitution). However, it has stopped short of developing a comprehensive and definite constitutional doctrine on the relationship between EU and Slovenian law. This is yet to take shape. In accordance with the Slovenian law-making tradition, it can be anticipated that in the process of the ultimate formation of this constitutional doctrine an important role will also be played by the Slovenian academic community. However, the attitude of the Slovenian academic community towards EU law has not been uniform. A distinction can be made between three different approaches: the international law approach, the supranational approach, and the pluralist approach.110 The proponents of the international law approach tend to assimilate EU law with general public international law and treat the EU as an international organisation, albeit of a special kind.111 While this perspective has been in decline, its original strong influence can be traced in the terminology employed in 2003 when Article 3a of the Constitution was drafted. Similar to some of the constitutions of the old Member States,112 it does not contain any explicit reference to the EU. Instead it speaks of international organisations to which Slovenia can transfer the exercise of a part of its sovereign rights.113 The proponents of the supranational approach espouse and defend the legal nature of European integration as construed by the CJEU in its heyday, ie when its primacy and direct effect jurisprudence culminated in the Simmenthal case.114 They argue that it is evident that the Slovenian Constitutional Court in general accepts the autonomy, direct effect, and primacy of EU law, which is a legal order sui generis, and which is separate from international law.115 EU law, furthermore, is perceived as hierarchically superior to national law and requires all domestic legal acts to comply with it.116 This means that a domestic legal order incrementally subjects itself to EU law, clears the path for the latter’s prevalence, and is, therefore, in a gradual retreat.117 This cannot be prevented even by the safeguards contained in Article 3a of the Constitution, which bind the EU to respect human rights, democracy, and rule of law. This is because in the opinion of one of the drafters of the said article, they cannot be used to refuse the application of EU law acts, even when they would violate the Constitution.118 Moreover, the influence of the proponents of the supranational approach can be identified in the language used in the Slovenian translation of the primacy clause in the failed Constitutional Treaty as well as in the present declaration on the principle of primacy in the Treaty of Lisbon. While all other language versions speak of primacy of EU law, the Slovenian version proclaims the supremacy of EU law over the entire body of national laws.119 Among the academic lawyers, 110 This draws directly on Matej Avbelj, ‘Report on Slovenia’, in G Martinico and O Pollicino (eds), The National Judicial Treatment of the ECHR and EU Laws. A Comparative Constitutional Perspective (Groningen, Europa Law Publishing, 2010) 441. 111 Igor Kaučič, ‘Ali je ustava primeren document za to mednarodno organizacijo’ [‘Is the Constitution a suitable document for this international organisation’], Presentation at a Conference in Bovec (2005). 112 Claes, ‘Constitutionalizing’ (n 55). 113 CRSI Art 3a. 114 ECJ 09.03.1978 106/77 (Amministrazione delle Finanze dello Stato v Simmenthal SpA) ECLI:EU:C:1978:49. 115 Mirjam Škrk, ‘The Role of the Constitutional Court of the Republic of Slovenia Following Integration into the European Union’, Presentation at the PCCEU Conference, Bled, 30.09. to 02.10.2004. 116 Albin Igličar, ‘Zakonodajna suverenost v pogojih polnopravnega članstva Slovenije v EU’ [‘Legislative Sovereignty under the Conditions of Slovenia’s Full EU Membership’] (2006) 25 Pravna praksa, 9; Petja Toškan, ‘Evropski člen in ustavna zaščita temeljnih človekovih pravic’ [‘The European Article and the Constitutional Protection of Fundamental Human Rights’] (2002) 3 Pravna praksa 4. 117 Ribičič, Mozaik (n 50) 29 f. 118 Miro Cerar, ‘URS in PUE: Usklajevanje Ustave RS z Ustavo za Evropo?’ [‘The Slovenian Constitution in the TECE: Harmonization of the Constitution of the Republic of Slovenia with the Constitution for Europe?’] (2005) 10 Pravna praksa, 6. 119 Matej Avbelj, ‘Zakaj je prevod PEU napačen?’ [‘Why is the TECE Translation Wrong?’] (2007) 18 Pravna praksa, 9; see also Matej Avbelj, ‘Supremacy or Primacy of EU Law – (Why) Does it Matter?’ (2011) 17 European Law Journal, 744.

Slovenia  697 this provision has been, by some, approved as yet another confirmation of a long established fact,120 while others have been extremely critical of it.121 The pluralist perspective recognises the EU’s special nature, its sui generis character, the autonomy of its legal order, while simultaneously insisting that national legal orders shall retain their original autonomy as well. European integration, therefore, functions as a common pluralist whole that is composed of twenty eight or twenty seven national and one supranational legal order. These orders are connected in a heterarchical rather than a hierarchical manner through the relational principle of primacy, whose effectiveness depends on the two-fold conditions stemming from EU and national law.122 This perspective is still only taking ground, but is becoming increasingly popular especially, albeit not exclusively, among the younger generation of lawyers.123

References F Adam and M Tomšič, ‘Transition Elites: Catalysts of Social Innovation or Rent Seekers’, (2000) 32 Družboslovne razprave, 138. R-L Arcas, ‘The Reception of European Community Law in Spain’, (2005) 1 Hanse Law Review, 18. M Avbelj, ‘(Ne)demokratičnost ustavnosodnih odločb’ [‘(Un)democratic Constitutionality of Decisions’] (2003) 58 Pravnik, 11. M Avbelj, ‘Zakaj je prevod PEU napačen?’ [‘Why is the TECE Translation Wrong?’] (2007) 18 Pravna praksa, 13. M Avbelj, ‘O sodnikih in vladavini prava’ [‘About Judges and the Rule of Law’] (2007) 20 Pravna praksa, 26. M Avbelj, ‘Slovenska sodišča in postopek za predhodno odločanje’ [‘Slovenian Courts and the Procedure of Preliminary Ruling’] (2009) 28 Pravna praksa, 9. M Avbelj, ‘Odnos nosilcev oblasti do prava’, [‘The Attitude of the Authorities to Law’] (2009) 33 Pravna praksa, 17. M Avbelj, ‘Report on Slovenia’, in G Martinico and O Pollicino (eds), The National Judicial Treatment of the ECHR and EU Laws. A Comparative Constitutional Perspective (Groningen, Europa Law Publishing, 2010) 441. M Avbelj, ‘Zgodovinska odločitev ustavnega sodišča’ [‘Historic Decisions of the Constitutional Court’] (2011) 39–40 Pravna praksa, 17. M Avbelj, ‘Supremacy or Primacy of EU Law – (Why) Does it Matter?’, (2011) 17 European Law Journal, 744. M Avbelj, ‘Theory of European Union’, (2011) 36 European Law Review, 818. M Avbelj, ‘Slovensko ustavno pravo v odnosu do prava EU’ [‘Slovenian Constitutional Law in Relation to EU Law’], in I Kaučič (ed), Dvajset let Ustave Republike Slovenije: pomen ustavnosti in ustavna demokracija (Ljubljana, Ustavno sodišče RS in Pravna fakulteta, 2012) 341. M Avbelj, ‘Crises and Perspectives in Building a European Nation – The Case of Slovenia’, in P Jambrek (ed), Nation’s Transitions: Social and Legal Issues of Slovenia’s Transitions: 1945–2015 (Kranj, Graduate School of Government and European Studies, 2014) 395. M Avbelj, ‘Failed Democracy: The Slovenian Patria Case – (Non) Law in Context’, (2014) 61/62, Dignitas 59.

120 Cerar, ‘URS’ (n 117) 6. 121 Avbelj, ‘Supremacy’ (n 118). 122 Matej Avbelj, ‘Theory of European Union’, (2011) 36 European Law Review 818. 123 For example, Ciril Ribičič, ‘Implementing European Standards into the Case-law of the Constitutional Court’, Presentation at the PCCEU Conference, Bled, 30.09. to 02.10.2004, who noted that ‘Hierarchy and primacy of one set of courts over another is not the right solution, at least insofar as the protection of human rights is considered. Dialogue, cooperation and synergy are necessary instead. It would be therefore wrong or at least premature to close the door for a potential intervention by the national constitutional courts when the values on the basis of which the states have entered the Union would get under threat.’ See also Boštjan Zalar, ‘Prve izkušnje sodišča in sodnikovi pogledi na uporabo prava Evropske unije’ [‘The Court’s First Experience and the Judge’s Views on the Application of European Union Law’] (2005) 6 Pravna praksa, 15.

698  Matej Avbelj M Avbelj, Nepravna država. Zapisi o zadevi Patria (2011–2015) [The Wrong State. Records on the Patria Case (2011–2015)] (Ljubljana, GV Založba, 2015). M Avbelj, ‘Transformation in the Eye of the Beholder’, in M Bobek (ed), Central European Judges under the European Influence: The Transformative Power of the EU Revisited (Oxford, Hart, 2015) 275. M Avbelj and J Letnar Černič, ‘Slovenia’, in L Hammer and F Emmert (eds), The European Convention on Human Rights and Fundamental Freedoms in Central and Eastern Europe (The Hague, Eleven International Publishing, 2012) 527. B Bugarič, ‘Crisis of Constitutional Democracy in Post-Communist Europe: “Lands in-between” democracy and authoritarianism’, (2015) 13 ICON, 219. M Cerar, ‘URS in PUE: Usklajevanje Ustave RS z Ustavo za Evropo?’ [‘The Slovenian Constitution in the TECE: Harmonization of the Constitution of the Republic of Slovenia with the Constitution for Europe?’] (2005) 10 Pravna praksa, 6. M Cerar, ‘The Role of national parliaments in the European Union’, in GC Rodríguez Iglesias and L Ortiz Blanco (eds), The role of national parliaments in the European Union. Proceedings of the FIDE XXIV Congress, Vol I (Madrid, Servicio de Publicaciones de la Facultad de Derecho de la Universidad Complutense de Madrid, 2010). M Claes, ‘Constitutionalizing Europe at its Source: The “European Clauses” in the National Constitutions: Evolution and Typology’, (2005) 24 Yearbook of European Law, 81. M Ferenc, Huda Jama (Grave Pit) Coal Mine Massacre (Ljubljana, Znanstvena založba Filozofske fakultete, 2013). C Hefftler, C Neuhold, O Rozenberg and J Smith, The Palgrave Handbook of National Parliaments and the European Union (London, Palgrave Macmillan, 2015). A Igličar, ‘Zakonodajna suverenost v pogojih polnopravnega članstva Slovenije v EU’] ‘Legislative Sovereignty under the Conditions of Slovenia’s Full EU Membership’], (2006) 25 Pravna praksa, 9. P Jambrek, ‘Zgodovinski spomin Nove Revije in slovenska prihodnost v Evropi’ [‘The Historical memory of Nova Revija and the Slovenian future in Europe’], (2007) 26 Nova Revija, 60. I Kristan, ‘Sprememba Ustave RS: Ustava: mednarodni del: evropski člen’ [‘Amendment to the Constitution of the Republic of Slovenia. Constitution: International Section: European Article’] (2001) 6 Podjetje in delo, 867. I Kristan, ‘Delitev oblasti ob ankaranski zgodbi’ [‘The Separation of Powers in the Ankaran Story’], (2011) 47 Javna uprava 25 and 218. M Novak, Delitev oblasti; medigra prava in politike [The Separation of Powers. An Interplay of Law and Politics] (Ljubljana, Cankarjeva, 2003). Republika Slovenija Državni Zbor, Ustavne razprave 2001–2003 [Constitutional Debates 2001–2003], Vol III.1 (Ljubljana, Državni zbor RS, 2004). C Ribičič, Mozaik ustavnih sprememb: od slovenske ustavnosti k evropski [The mosaic of constitutional reform: from Slovenian towards European constitutionality] (Ljubljana, GV Založba, 2003). Služba Vlade Republike Slovenije za zakonodajo, Nomotehnične smernice [Government Office for Legislation, Technical Guidelines] (Ljubljana, Služba Vlade Republike Slovenije za zakonodajo, 2008). J Sovdat, ‘The Constitutional Court of the Republic of Slovenia and European Union Law’, (2013) 13 Hrvatska i komparativna javna uprava, 900. P Štih, V Simoniti, and P Vodopivec, Slovenska zgodovina: družba, politika, kultura [Slovenian History: Society, Politics, Culture] (Ljubljana, Inštitut za novejšo zgodovino, 2008). D Štrus, Drugi dom parlamenta [The Second House of Parliament] (Ljubljana, Fakulteta za upravo, 2011). L Šturm (ed), Komentar Ustave Republike Slovenije [Commentary on the Constitution of Slovenia] (Kranj, Evropska Pravna Fakulteta, 2002). P Toškan, ‘Evropski člen in ustavna zaščita temeljnih človekovih pravic’ [‘The European Article and the Constitutional Protection of Fundamental Human Rights’], (2002) 3 Pravna praksa, 4. N Walker, ‘Late Sovereignty in the European Union’, in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 3. B Zalar, ‘Prve izkušnje sodišča in sodnikovi pogledi na uporabo prava Evropske unije’ [‘The Court’s First Experience and the Judge’s Views on the Application of European Union Law’], (2005) 6 Pravna praksa, 15.

25 Slovakia From Constitutional Enthusiasm to Constitutional Blindness ZUZANA VIKARSKÁ AND KAMIL BARANÍK

I.  Main Characteristics of the Slovak Constitutional System The Slovak Republic (‘SR’, ‘Slovakia’) is a country situated in Central Europe, with approximately 5.4 million inhabitants and an area of slightly less than 50,000 sq km. As a post-communist country, Slovakia is a very young constitutional democracy that is still striving to achieve many of the democratic values that are taken for granted in western societies. Slovakia also belongs to the group of countries that joined the European Union (EU) in the historically biggest accession wave in 2004, which still makes it a ‘new’ EU Member State, even though more than 15 years have lapsed since this accession.1 The Slovak Republic became an independent state on 1 January 1993, after the amicable dissolution of the Czech and Slovak Federative Republic2 and with the entry into force of the Constitution of the Slovak Republic.3 This constitutional document, adopted in September 1992, is the cornerstone of the Slovak constitutional system. The constitutional order comprises the Constitution as well as its subsequent constitutional amendments and a number of self-standing constitutional acts. Furthermore, it stands at the top of the legal hierarchy and serves as the ‘referential framework’ for constitutional review exercised by the Slovak Constitutional Court (hereinafter SKCC). 1 These circumstances, taken together, form the necessary background for a good understanding of this chapter: in a smaller jurisdiction like Slovakia, few cases have been litigated and little legal scholarship has been produced on the specific matter of EU membership and its constitutional consequences. Furthermore, the authors have observed a general unwillingness of Slovak courts (and, unfortunately, also of academic authors) to enter into discussions or confrontations with the European legal order. In more detail, see for example Michal Bobek and Zdeněk Kühn, ‘Europe Yet to Come: The Application of EU Law in Slovakia’, in A Lazowski (ed), The Application of EU Law in the New Member States – Brave New World (The Hague/Cambridge, TMC Asser Press/Cambridge University Press, 2010) 373. See also Zuzana Vikarská and Michal Bobek, ‘Slovakia: Between Euro-Optimism and Euro-Concerns’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (The Hague, T.M.C. Asser Press, 2019) 835. 2 For a general overview of this process in English, see eg Eric Stein, ‘Out of the Ashes of a Federation, Two New Constitutions’, (1997) 45 American Journal of Comparative Law 45. 3 Constitutional Act No 460/1992 Coll, Constitution of the Slovak Republic (hereinafter CSK). A number of English translations of the CSK can be found online; the authors have primarily worked with the translation published on the website of the Slovak President, available at www.prezident.sk/upload-files/46422.pdf. It is interesting to note that some parts of the CSK entered into force on 1 October 1992 and that some of these were in direct contradiction with the (then still effective) federal constitution, eg appointment of judges in Art 145 CSK, which contradicted Art 101(3) of the previous constitution.

700  Zuzana Vikarská and Kamil Baraník

A.  Structure of the Constitution The Slovak Constitution is the foundation of the Slovak legal system, a document at the top of the legal hierarchy. It consists of 171 articles4 and can be said to have two main foci of interest: (a) the institutional arrangement of the state, including the separation of powers into three main branches, and (b) the system of protection of fundamental rights and freedoms. Historically, it is interesting to note that the Slovak Constitution has comprised a number of systemic requirements formulated during the Velvet Revolution of 1989, including the separation of powers, the parliamentary form of government, the rule of law, and – last but not least – the protection of fundamental rights and protection of minorities. The Constitution is divided into a preamble and nine chapters. The first chapter of the Constitution establishes the fundamental constitutional principles. These include democracy, sovereignty, rule of law and respect for international obligations (Article 1), limited state power originating from the citizens (Article 2), state territory and boundaries (Article 3), protection of national wealth (Article 4), citizenship (Article 5), language (Article 6), modalities of entry into international obligations and the position of international treaties in the national legal order (Article 7), state symbols (Articles 8 and 9), and the capital of the republic (Article 10). The second chapter includes a full catalogue of rights (Articles 12–54). The chapter first sets out basic principles concerning fundamental rights (Article 12), the concept of binding legal obligations and limitations of rights (Article 13), and then proceeds to the following categories of rights: fundamental human rights and freedoms (Articles 14–25), political rights (Articles 26–32), rights of national minorities and ethnical groups (Articles 33–34), economic, social and cultural rights (Articles 35–43), environmental and cultural heritage rights (Articles 44–45), procedural rights (Articles 46–50) and concluding provisions (Articles 51–54) which, interestingly, also include the right to asylum (Article 53).5 The third chapter addresses the economy of the country (Articles 55–59) and the institution of the Supreme Audit Office (Articles 60–63). In a comparative perspective, it is quite uncommon that the Constitution contains a provision on the protection of economic competition (Article 55(2)). The fourth chapter sets out the principles of territorial self-administration (Articles 64–71). The fifth, sixth and seventh chapters govern the three major powers of the state respectively: the legislature (Chapter V; Articles 72–100), the executive branch (Chapter VI; Articles 101–23), and the judiciary (Chapter VII; Articles 124–48). The Constitution does not explicitly recognise the principle of separation of powers, but the constitutional structure clearly stems from it. As for the legislature, the National Council of the Slovak Republic (NCSR) is composed of a single chamber, consisting of 150 Members of Parliament (MPs). The Constitution sets the quorum at ‘more than half of all MPs’ in Article 84(1). A simple majority of the MPs present is therefore required for decisions in general (Article 84(2)), as well as for adoption of (ordinary) statutes, while an absolute majority (ie 76 MPs) is required to approve the most important (ie ‘presidential’) treaties, as well as to overrule the presidential veto (Article 84(3)), and to express the vote of non-confidence to the government (Article 88(2)). A three-fifths majority (ie 90 MPs) 4 The articles are numbered from 1 to 156, but through amendments, 20 further articles were added (Arts 7a, 55a, 64a, 81a, 88a, 125a–b, 127a, 129a, 141a–b, 145a, 151a, 154a–g) and two were removed (Arts 11 and 132). 5 This peculiar placement of the right to asylum is similar in the Czech Constitution. See (Czech) Constitutional Act No 2/1993 Coll, Charter of Fundamental Rights and Freedoms, which also includes the right to asylum in its final chapter titled ‘General Provisions’.

Slovakia  701 is required to adopt constitutional statutes, to impeach the President, to approve special categories of international treaties, and in other cases foreseen in Article 84(4). Chapter V further includes a sub-chapter on referendum (Articles 93–100). As for the executive branch, Slovakia is a parliamentary republic with a directly elected President. Chapter VI is divided into two sub-chapters, one addressing the President (Articles 101–07) and the other one devoted to the Government (Articles 108–23). Chapter VII establishes the judicial branch and consists of two sub-chapters: the first subchapter addressing the SKCC (Articles 124–40) and the second sub-chapter dealing with ordinary courts (Articles 141–48). The second sub-chapter also includes two provisions dealing with the Judicial Council (Article 141a–b).6 The judiciary consists of three tiers of ordinary courts (54 district courts, eight regional courts and the Special Criminal Court, and the Supreme Court and newly established Supreme Administrative Court7), while the SKCC stands outside the system of ordinary courts. The judicial proceedings operate at two instances; usually a district court hears the case at first instance and a regional court hears the appeal. In certain cases, a regional court (or the Special Criminal Court) will be competent to decide a case at first instance, and the Supreme Court will hear the appeal. Furthermore, the Supreme Court hears extraordinary appeals, which are only admissible under certain circumstances that are strictly specified in the statutes. The SKCC does not serve as an appellate court; it only deals with the questions arising from ‘the protection of constitutionality’.8 The final constitutional provisions address the Prosecutor’s Office (Chapter VIII; Articles 149–51), the Public Defender of Rights (Article 151a), and transitional and derogatory provisions (Chapter IX; Articles 152–56).

B.  The Constitutional Court As outlined above, the SKCC is addressed in Chapter VII, sub-chapter I of the Constitution. The SKCC is defined in Article 124 as ‘an independent judicial body charged with the protection of constitutionality.’ It operates outside the judicial structure of ordinary courts and does not directly hear appeals against cases decided by ordinary courts.9 It only deals with questions of constitutional nature. The SKCC can be described as ‘a specialised and concentrated constitutional court with the power of ex post judicial review.’10 The SKCC was established to serve as the last check in constitutional disputes.11 It wields powers over 21 different types of proceedings. The SKCC hears cases in plenary sessions, or in one of its four three-member Senates. The most prominent of the SKCC’s powers is the power of

6 Together with many other provisions, Art 141a was inserted into the SKC by the Euro Amendment which came into effect on 1 July 2001 and is discussed below in section III.A. Art 141b was inserted much later, by a constitutional amendment No 422/2020 Coll of 9 December 2020, effective from 1 January 2021. 7 The Supreme Administrative Court was established by Act No 422/2020 Coll., effective from 1 January 2021, and started to function as of 1 August 2021. 8 See eg SKCC 14.03.2001, Decision I. ÚS 13/01 (SKCC decisions are available in Slovak at http://merit.slv. cz/./). A very narrow selection of English translations of SKCC decisions is available via the Court’s website at www.ustavnysud.sk/o-ustavnom-sude-slovenskej-republiky. 9 There are some exceptions from this rule – in a material sense, it hears appeals against decisions of the Supreme Court to suspend and/or to dissolve a political party, as foreseen in Art 129(4) SKC. 10 Lucia Berdisová, Constitutional Law (Trnava, Trnavská univerzita v Trnave. Právnická fakulta, 2013) 77. 11 For more details about the powers of constitutional judiciary in the Central European region see Kamil Baraník ‘Why Have Constitutional Courts Been so Important for Democracy in Central Europe (… And So Hated by Those in Power)?’ (2018) 11 Juridiskā zinātne 77, 84–87.

702  Zuzana Vikarská and Kamil Baraník abstract constitutional review. The Court acts as the ‘Kelsenian negative legislator’ (Article 125) that reviews the parliamentary legislation in light of the Constitution and its material principles, as well as in light of international treaties.12 In other words, the SKCC may declare a statute invalid not only for its conflict with the Constitution, but also for its conflict with a directly applicable international treaty, which is very interesting in comparative perspective.13 Second, the SKCC is the ultimate interpreter of the Constitution in interpretative disputes (Article 128). Both constitutional interpretation and review are vested in the Plenary and enjoy generally binding effect. The Senates conduct ‘concrete’ review (linked to a real, not a hypothetical case) in cases of alleged violation of individuals’ constitutional rights (Article 127). The SKCC has 13 judges, appointed for non-renewable terms of 12 years (Article 134). With three exceptions, the current SKCC judges were appointed in 2019–2020 and the current court is often referred to as ‘the Fourth Constitutional Court’.14 The Czechoslovak constitutional judiciary dates back to 1921 when the first Czechoslovak Constitutional Court was established, with its competence limited to abstract constitutional review.15 Its foundations were based on a Kelsenian vision where a specialised judicial institution decides on issues of constitutionality by which it acts as a guardian of the Constitution. This first Czechoslovak Constitutional Court was rather lethargic: between 1923 and 1929 it did not decide any cases and the judges only met once a year at a plenary meeting which was more of a social event than a work meeting. This period in the court’s existence is often referred to as ‘Advent Constitutional Court’, since these yearly meetings took place in the advent period before Christmas.16 This ‘First Constitutional Court’ ceased to exist upon the expiration of the 10-year mandate of the first judges in 1931, due to the lack of political consensus on the nomination of new judges. The plenary unanimously agreed in 1931 that the judges were to remain in their function until the appointment of new judges, but the Court was still not operational.17 On 11 July 1937, President Edvard Beneš appointed new members but their letters of appointment were never delivered. The Court was only re-established on 10 May 1938.18 During World War II, constitutional review in the Slovak State was assigned to the Constitutional Senate.19 This body, however, never delivered any decisions. Between 1948 and 1989, the protection of constitutionality was left to the legislature and no constitutional court was established (despite the wording of the so-called ‘Federalisation’ Amendment of 1968).20 After the ‘Velvet Revolution’, the constitutional judiciary was reborn and in 1992 the Constitutional Court of the Czech and Slovak Federative Republic (CSFR) was established and endowed with the competence of both abstract and concrete judicial review.21 From a theoretical perspective, interestingly, legal norms that are declared unconstitutional are not abrogated immediately; the SKCC merely declares their unconstitutionality. As a result, 12 See Art 125(1) a) CSK. 13 See Kamil Baraník and Marián Giba, ‘Medzinárodné zmluvy a Ústavný súd Slovenskej republiky v treťom funkčnom období’ [‘International Treaties and the Constitutional Court of the Slovak Republic in the Third Term of Office’], in L Orosz et al (eds), Tretie funkčné obdobie Ústavného súdu Slovenskej republiky – VII. ústavné dni (Košice, Univerzita Pavla Jozefa Šafárika, 2019) 241. 14 For a list of all the serving judges of the fourth constitutional court, see www.ustavnysud.sk/sudcovia. 15 See Art III of the first Czechoslovak Constitution, Act No 121/1920 Coll, which laid the legal basis for the first Czechoslovak Constitutional Court, as well as Act No 162/1920 Coll, on the Constitutional Court. 16 See Tomáš Langášek, Ústavní soud Československé republiky a jeho osudy v letech 1920–1948 [The Constitutional Court of the Czechoslovak Republic and Its Destiny in the Period 1920–1948] at www.usoud.cz/ustavni-soud-csr-1920-1948/. 17 Ibid. 18 Ibid. 19 See Art 98 of Constitutional Act No 185/1939 Coll, the Constitution of the Slovak State. 20 See Arts 86 to 101 of Constitutional Act No 143/1968 Coll, on the Czechoslovak Federation. 21 See Constitutional Act No 91/1991 Coll, on the Constitutional Court of the CSFR, as well as Act No 491/1991 Coll, on the organisation of the Constitutional Court of the CSFR.

Slovakia  703 such legal norms lose their legal effects (ie these norms are inapplicable), although they still remain part of the legal system. The competent institution is then under a legal obligation to bring them into conformity with the Constitution within six months. If that does not happen, the unconstitutional legal norms lose not only their effect, but also their validity and, thus, are repudiated from the legal system.22

1.  Appointment of Judges to the Constitutional Court The appointment of judges to the SKCC can be divided into four stages. First, candidates are proposed to the NCSR by the so-called competent entities.23 Second, the Constitutional Committee of the NCSR verifies whether the candidates comply with the constitutional prerequisites for the position of a constitutional judge, and subsequently this Committee conducts hearings with the proposed candidates. These interviews are not as detailed as in some other countries with comparable judicial positions.24 This less rigid practice, however, has changed significantly in the most recent hearings in 2019. Third, the NCSR votes on the candidates and approves twice as many candidates as are to be finally appointed. Last, the President of the Republic selects which candidates are to be appointed to the SKCC. This relatively straightforward and hitherto unchallenged SKCC appointment process became source of a serious tension between the NCSR and the President in recent years. The dispute over the interpretation of relevant constitutional provisions (whether the President is obliged to accept or is allowed to refuse nominated candidates for the SKCC’s judgeships) left the Court incomplete for well over three years (July 2014–December 2017). The issue took off when the newly elected President Andrej Kiska was presented with six nominated candidates in April and May 2014 and was expected to elevate three of them to the SKCC. However, the President decided to appoint only one of them, and refused to appoint any of the remaining five candidates. The unsuccessful candidates brought constitutional complaints, three of which were decided on 17 March 2015.25 These actions unleashed another political power struggle over the SKCC’s vacancies, the constitutional competencies of the President and thereby connected nomination powers of respective branches government. The issue was also discussed seriously at the international level.26 The nomination saga was temporarily settled, on 6 December 2017, when one of the SKCC’s Senates, in an extremely contested decision, held that the President had infringed the rights of

22 See Art 125(3) of the Constitution. 23 These are defined in Act No 314/2018 Coll, on the organisation of the Constitutional Court, and they include, inter alia, members of the NCSR, the Government, the President of the Constitutional Court, the President of the Supreme Court, the General Prosecutor, lawyers’ organisations, and scientific institutions. 24 Berdisová, Constitutional Law (n 10) 79. 25 See SKCC 17.03.2015 Decision III. ÚS 571/2014. Interestingly, there are significant discrepancies between the oral verdict delivered on the day of the decision, and the written decision that was published a couple of weeks later. The oral verdict was more lenient and allowed the President a certain amount of discretion in extraordinary cases, but the written decision of the SKCC establishes a very straightforward obligation of the President always to appoint half of the candidates nominated by the NCSR. The President has reacted to these discrepancies by a letter addressed to the SKCC, and the SKCC has published a response to this letter, stating that the oral and the written decision do not differ significantly and that in cases of doubt, the written decision has precedence over the oral verdict. 26 CDL-AD (2017) 001 Slovak Republic – Opinion on questions relating to the appointment of judges, adopted by the Venice Commission at its 110th Plenary Session (March 10–11, 2017) 7); CDL-AD (2014) 015 Opinion on the procedure for appointing judges to the Constitutional Court in times of the presidential transition in the Slovak Republic, adopted by the Venice Commission at its 99th Plenary Session (June 13–14, 2014), accessible at www.venice.coe.int/webforms/ documents/.

704  Zuzana Vikarská and Kamil Baraník the rejected candidates who brought the constitutional complaint. The President backtracked, and ultimately decided to appoint two additional judges for the remaining vacancies at the SKCC on 14 December 2017.27 On 16 February 2019, the tenure of nine out of 13 of the SKCC’s judges expired. Consequently, only four judges remained in function and the SKCC effectively became paralysed.28 Despite the fact that the expiration of the judicial mandate of a majority of judges had been long foreseen, the NCSR did not duly select 18 candidates, as required by the Constitution. After lengthy obstructions by the parliamentary majority, as well as numerous disputes about the procedural aspects of the vote (eg a public versus secret ballot), the NCSR finally nominated six candidates on 3 April 2019 and two on 4 April 2019. On 17 April 2019, the President appointed three new judges out of the eight nominees. This unusual approach, which did not follow the conventional practice from previous appointments (ie to wait until the full number of nominees is presented to the President), was defended on the grounds of an emergency measure that enabled the SKCC’s plenary session to operate. Moreover, this nomination involved the appointment of the SKCC President. This was strictly necessary, since the President of the Court was required to administer the oath of the newly elected President of the Republic. The day of the inauguration was approaching quickly and, therefore, the outgoing President appointed the President of the SKCC to prevent putting the constitutionality of the process of inauguration at risk. Six more judges were appointed on 10 October 2019 and with the last judicial appointment of 30 September 2020, the ‘fourth Constitutional Court’ became complete again. This heated issue could be perceived as part of an ongoing constitutional dispute between the respective branches of government over their competencies and postures within the Slovak constitutional system. Various appointment power disputes29 were based on the use of different methods of constitutional interpretation. The SKCC has constantly proven incapable of following a coherent line of argumentation and twisted its deliberations when governing coalitions changed. Therefore, the SKCC contributed heavily to its recent reputational demise by missing several looming opportunities to cement itself as a neutral law-abiding institution.30

2.  The Constitutional Court and Its Role under Article 267 TFEU The SKCC referred its first question to the Court of Justice of the EU (CJEU) on 23 January 2019, ie almost 15 years after the accession of the Slovak Republic to the European Union. The SKCC has, however, always stated very openly that it considers itself a ‘referring court’ in the sense of Article 267(3) TFEU. In 2008, the SKCC declared that ‘it is for the national court to ensure fair protection of the rights and legitimate interests of the parties’.31 According to the SKCC, one such guarantee in the Slovak legal system presupposes that national courts will be able to refer 27 SKCC 06.12.2017, Decision I. ÚS 575/2016. For a dedicated discussion on the results and ramifications of this decision see I-CONnect Symposium: The Slovak Constitutional Court Appointments Case that revealed quite a significant chasm even among participating academics; Šimon Drugda, ‘Introduction’; Šimon Drugda, ‘Intermezzo to the Constitutional Conflict in Slovakia: A Case Critique’; Marek Domin, ‘The President’s Appointments’; Kamil Baraník, ‘Perplexities of the Appointment Process Resolved by Means of “Fire and Fury”’; Tomáš Ľalík, ‘Born is the King: The Day When Effective Judicial Review Arrived’ (all best accessible via www.iconnectblog.com/2018/01/symposium-slovak-appointments-case-lalik/). 28 Pursuant to Art 131 CSK, the plenary session adopts decisions by a majority of all its members, ie by seven judges. Therefore, as of 16 February 2019, the SKCC was not able to decide plenary cases. 29 Over the Prosecutor General (SKCC 24.10.2012, Decision PL. ÚS 4/2012); the Vice-President of the National Bank of the Slovak Republic (SKCC 23.09.2006, Decision PL. ÚS 14/06); and over the Members of the Government (SKCC 02.06.1993, Decision I. ÚS 39/93). 30 Eg in I. ÚS 575/2016 (n 27) the respective Senate implicitly declared itself politically biased (64–66). 31 SKCC 03.07.2008, Decision IV. ÚS 206/08, part II.

Slovakia  705 a question dealing with the interpretation of EU law to the CJEU. In that regard, however, the SKCC pointed out that parties to the proceedings are in principle not entitled to such a referral and it is for a respective court to decide. In this regard, the case law of the SKCC is in line with the case law of the European Court on Human Rights (ECtHR). That is, although parties do not have a right per se to have their case referred for a preliminary ruling, their procedural rights are breached if the court’s decision not to refer the case lacks sufficient reasoning.32 On the other hand, if a national court decides at last instance, it may be obliged to refer a relevant question to the Court of Justice, as follows from Article 267(3) TFEU. Interestingly, the German Federal Constitutional Court has developed a doctrine of a ‘right to a lawful judge’ in situations where a national court of last instance fails to refer a question for a preliminary ruling, despite being obliged to do so. In such cases, the ‘right to a lawful judge’ is also the EU judge in Luxembourg. The SKCC has followed this doctrine in its own case law: in the scope of EU law, the lawful judge in a matter considered by a national court is not only the judge determined by the case assignment schedule … but also the EU judge, since his/her duty to rule on questions of interpretation of EU law is at the same time his/her entitlement, which cannot be circumvented, except for situations foreseen by EU law which follow from the case law of the Court of Justice, for example in matters where the CILFIT formula applies.33

Furthermore, the SKCC also added that the failure to refer may trigger the state’s liability for breach of EU law, as has been established in the famous Köbler case.34 Last but not least, in an obiter dictum, the SKCC also declared that ‘in the exercise of its competences, the Constitutional Court may reach a situation in which it will be under an obligation to refer a question to the Court of Justice’.35 This ‘situation’ occurred in a judicial review case initiated by the Slovak President: in January 2019, the SKCC turned to the Court of Justice with a question on interpretation of the internal market electricity directive36 and the CJEU issued its judgment in June 2020.37

II.  Constitutional Culture Despite the fact that the Slovak Constitution – unlike its Czech counterpart – appears to avoid any explicit reference to the common Czechoslovak history,38 the Slovak constitutional culture can be said to be based on the Czechoslovak constitutional tradition, namely on the constitutional norms and culture that the two nations shared from 1918 to 1993. During its constitutional history, Czechoslovakia adopted several federal constitutions (the post-WWI Constitution of 1920, the post-WWII Constitution of 1948, the working-class-centred socialist Constitution of 1960, and the federalisation legislation adopted between 1968 and 1970).

32 See ECtHR 20.09.2001, Application Nos 3989/07 and 38353/07 (Ullens de Schooten and Rezabek v Belgium) ECLI:CE:ECHR:2011:0920JUD000398907; 10.04.2012, Application No 4832/04 (Vergauwen and others v Belgium) ECLI:CE:ECHR:2012:0410DEC000483204; 08.04.2014, Application No 17120/09 (Dhahbi v Italy) ECLI:CE:ECHR:2014: 0408JUD001712009; or 21.07.2015, Application No 38369/09 (Schipani v Italy) ECLI:CE:ECHR:2015:0721JUD003836909. 33 SKCC IV. ÚS 206/08 (n 31) part II. 34 ECJ 30.09.2003 C-224/01 (Köbler) ECLI:EU:C:2003:513. 35 SKCC IV. ÚS 206/08 (n 31) part II in fine. 36 Directive 2009/72/EC of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54/EC, [2009] OJ L 211/55. 37 ECJ 11.06.2020 C-378/19 (Prezident Slovenskej republiky) ECLI:EU:C:2020:462. 38 The CSK Preamble refers all the way back to the ‘spiritual heritage of Cyril and Methodius and the historical legacy of Great Moravia’, but not to Czechoslovak constitutional history of the twentieth century.

706  Zuzana Vikarská and Kamil Baraník Formally, the dissolution of the CSFR followed from bilateral political arrangements, namely from the agreement between the Prime Ministers of the two parts of the federation. However, when examined in more detail, the termination of the common state can be seen as a tale of Slovak emancipation – a unilateral secession of Slovakia by way of drafting the Slovak Constitution and invoking the right of nations to self-determination.39 In July 1992, the Slovak National Council40 adopted a political document titled ‘Declaration on the Sovereignty of the Slovak Republic’ and declared the independence of Slovakia from the CSFR.41 The document reads as follows: We, the democratically elected Slovak National Council, solemnly declare that the thousand years’ struggle of the Slovak nation for independence has been fulfilled. In this historical moment, we declare the natural right of the Slovak nation for self-determination, as embodied by all international agreements and treaties about the right of nations to self-determination. Recognising the right of nations to self-determination, we declare that we also want to freely create the way and form of national and state life while respecting the rights of everybody, all citizens, nations, national minorities, ethnic groups, and the democratic and humanist legacy of Europe and of the world. By this declaration, the Slovak National Council declares sovereignty of the Slovak Republic as the basis for a sovereign state of the Slovak nation.

In reaction to this declaration, political and legal measures were adopted, eventually leading to the dissolution of the CSFR. In September 1992, the Slovak National Council adopted the Constitution of the Slovak Republic. This constitutional document ‘was prepared in a hasty and secretive manner and gave rise to bitter political divisions in the Parliament’.42 Yet, it passed by the required (three-fifths) majority43 and entered into force on 1 October 1992 (with the exception of some provisions that entered into force on 1 January 1993). As of 1 January 1993, two new sovereign states appeared on the map of the world: the Czech Republic and the Slovak Republic. Sovereignty, however, surely did not mean isolation. Immediately in 1993, Slovakia became a member of various international bodies, including the United Nations, the Organisation for Security and Cooperation in Europe, and the Council of Europe. Since the beginning of its independent existence, Slovakia has entered into hundreds of bilateral and multilateral treaties from which it draws both political and legal consequences.44 Within the national legal order, it is the role of constitutional law to define the conditions for the validity of norms of international law (as well as EU law) and to lay down the rules for the relationship between international, European and domestic law.

39 The right of nations to self-determination is even explicitly mentioned in the preamble to the Slovak Constitution. 40 The Slovak National Council was the Parliament of the Slovak part of the CSFR and its competences were regulated by Art 142(2) of the Constitutional Act on the Czechoslovak Federation (n 21). Naturally, these competences did not include the adoption of a constitution for a sovereign state. 41 It is, however, necessary to underline the political, rather than legal, nature of this document. Legally, the independence of Slovakia only came into effect with the dissolution of the federation, ie on 1 January 1993. 42 Radoslav Procházka, Mission Accomplished: On Founding Constitutional Adjudication in Central Europe (Budapest, Central European University Press, 2002) 68. 43 Of 134 MPs present, 114 voted in favour of the proposal, 16 voted against and four refrained from voting. For a stenographic record of the voting in Slovak see the webpage of the Slovak National Council at www.nrsr.sk/dl/Browser/ Document?documentId=71565. 44 In this context, Stephen Krasner mentions an interesting paradox: when states act as sovereign international actors and enter into international relations with other states, they in fact limit their own Westphalian sovereignty, ie the power to decide independently about their domestic (inter-state) matters, since international relations bring international obligations which have to be reflected also in inter-state legal and political system. Inside this system, it is up to the constitutional legal order to determine how international obligations are reflected in the national legal order. See Stephen Krasner, Sovereignty: Organized Hypocrisy (Princeton NJ, Princeton University Press, 1999) 4.

Slovakia  707

III.  Constitutional Foundations of EU Membership and Closely Related Instruments The Slovak Republic applied for EU membership in 1995 and acceded on 1 May 2004, in the largest accession wave, together with nine other states. To make the accession possible, the Slovak constitutional system was redesigned significantly, since the original wording of the Constitution did not provide a possibility to accede to a supranational organisation such as the EU.45

A.  The Euro Amendment and Subsequent EU-related Amendments In the context of EU accession, the Slovak Constitution was significantly amended by Constitutional Act No 90/2001 (the so-called ‘Euro Amendment’),46 adopted by 90 MPs (the minimum number of MPs required for a constitutional amendment), with 57 MPs voting against the proposal. This constitutional change paved the way towards accession to the EU, as well as to NATO. In other words, the Euro Amendment prepared a fertile ground for a more open approach to international law and its impact in the Slovak legal order. The Euro Amendment was a direct amendment of the Constitution. It inserted a number of key articles relevant to Slovak membership in the EU, most importantly Articles 1(2), 7, 120(2), 125(1), 125a, 144, and 154c. The original Article 11, which had laid down the primacy of human rights treaties in case they provided for a more extensive human rights protection, was repealed. These issues are now governed by Articles 7(5) and 154c, as will be explained below. Most importantly, the new Article 1(2) brought a change of paradigm: it created a new constitutionally friendly perception of international law, international treaties and international obligations. By inserting this provision into the very first chapter of the Constitution titled ‘Fundamental Provisions’, the Euro Amendment in fact inserted a new fundamental constitutional principle to the Constitution.47 The provision of Article 1(2) adds a rather monist flavour to the Slovak legal order.48 It lays down a traditional principle of international law according to which a state cannot invoke a provision of its national legal order to excuse its non-performance of an international obligation.49 In relation to this provision, the SKCC has ruled the following: Article 1(2) of the Constitution concerns all international obligations of the Slovak Republic, irrespective of their content, and lays down an obligation to fulfil them. Every international obligation of the Slovak 45 There are, however, views to the contrary. Procházka has argued that an amendment was in fact not necessary. See Radoslav Procházka and Juraj Čorba, Právo Európskej únie [European Union Law] (Žilina, Poradca podnikateľa, 2006) 99. 46 Constitutional Act No 90/2001 Coll, which amends the Constitution of the Slovak Republic, adopted on 23 February 2001, effective from 1 July 2001. It is interesting to note the similarity of this constitutional amendment and the first draft version of the Czech Euro Amendment which was declined in the very early stages of the legislative process in the Chamber of Deputies of the Czech Parliament on 8 June 1999. This Czech draft is available online (in Czech) at the webpage of the Chamber of Deputies (www.psp.cz), third election period, document 208-0. See Vít Schorm, ‘Osud vládního návrhu novely Ústavy v oblasti mezinárodních vztahů, Evropské unie a Severoatlantické aliance’ [‘The Fate of the Government‘s Draft Amendment to the Constitution in the Area of International Relations, the European Union and the North Atlantic Alliance’], in B Dančak, V Šimiček (eds), Aktuálnost změn Ústavy ČR (Brno, Masarykova univerzita, 1999) 67. 47 Kamil Baraník, ‘Interakcia medzinárodného, supranacionálneho a ústavného práva pri ochrane ľudských práv v SR’ [‘Interaction of International, Supranational and Constitutional Law in the Protection of Human Rights in the Slovak Republic’] (2017) 69 Justičná Revue 1118. 48 This is similar to other post-communist constitutions; see Jiří Malenovský, ‘K nové doktríně Ústavního soudu ČR v otázce vztahů českého, komunitárního a mezinárodního práva’ [‘On the New Doctrine of the Constitutional Court of the Czech Republic in the Relationship between Czech, Community and International Law’] (2006) 14 Právní rozhledy 782. 49 See Art 27 of the Vienna Convention on the Law of Treaties (VCLT), published in the Slovak National Gazette as ordonnance No 15/1988 Coll This convention entered into force in the Czech and Slovak Socialist Republic on 28 August 1987.

708  Zuzana Vikarská and Kamil Baraník Republic has its own content comprised in an international treaty, international custom, or another source of international law. Consequently, national legal (statutory) regulation can become contrary to such an international obligation and the Constitutional Court is entitled to decide about such a conflict also in the proceedings concerning the validity of legal norms.50

The travaux préparatoires to Article 1(2) label this provision as ‘a new constitutional principle’ and state: The proposed provision must be interpreted not only as an avowal of the Slovak Republic to the general rules of international law, to international agreements by which it is bound and to its other international obligations, but also as a formulation of a constitutional commitment for the National Council of the Slovak Republic to create the legal order of the Slovak Republic in accordance with the abovementioned rules, international treaties and other international commitments of the Slovak Republic.

This commitment can also be found at the statutory level, where it is declared that the Parliament should aim to adopt such laws that can ‘become a functional component of a well-balanced, wellarranged and stable legal order of the Slovak Republic, compatible with European Union law and with the international legal obligations of the Slovak Republic’.51 Furthermore, Article 1(2) also serves as a principle of interpretation according to which the Constitution shall be interpreted in a way that ensures conformity with international and European law. According to the SKCC, the Constitution must be interpreted and applied ‘in a way that respects the international legal obligations of the Slovak Republic’, not in a way that would limit or exclude their implementation.52 Thus, the bulk of changes relating to the Slovak accession to the EU were carried out by the Euro Amendment in 2001. As far as subsequent constitutional changes relating to EU membership are concerned, in 2004 the Slovak legislator sought to regulate the relationships between the Government and the Parliament when participating in the EU legislative process, as well as the incompatibility of serving as an MP of the Slovak Parliament and the European Parliament. Both these provisions were originally included in the draft of the Constitutional Act No 140/2004. Back then, however, these two proposals did not secure sufficient support in the NCSR and were removed from this draft constitutional act. The incompatibility provision was nonetheless later included in Constitutional Act No 323/2004,53 and the relationship between the Government and the Parliament was later provided for expressly by the Constitutional Act No 397/2004 on the cooperation of the NCSR and the Government of the SR in EU matters.54 The latter constitutional amendment was different from the previously mentioned ones in that it did not directly alter the constitutional text, but rather exists as a self-standing piece of constitutional legislation.

B.  Legal Framework of the Slovak Accession to the EU The provision of Article 7(2) of the Slovak Constitution provided a legal basis for the accession of the SR to the EU. Pursuant to this article, The Slovak Republic may, by an international treaty ratified and promulgated in a manner laid down by law, or on the basis of such treaty, transfer the exercise of a part of its rights to the European Communities and European Union. 50 SKCC 21.10.2003, Decision PL. ÚS 44/03. 51 See Art 2 of Act No 400/2015 Coll, on the Creation of Legal Instruments and on the Collection of Statutes of the Slovak Republic. 52 See SKCC 30.05.2001, Decision PL. ÚS 17/00; see also SKCC 26.01.2011, Decision PL. ÚS 3/09. 53 Adopted on 14 May 2004, with effect from 1 June 2004. 54 Adopted on 24 June 2004, with effect from 1 August 2004.

Slovakia  709 It follows from this wording that there are four formal conditions for accession to the EC/EU: (1) conclusion of a treaty, (2) consent of the NCSR, pursuant to Article 84(4) of the Constitution, (3) ratification by the President, and (4) promulgation in the Collection of Laws of the SR. On the other hand, no material conditions were required in order to accede to the EU.55 The Accession Treaty was signed in Athens on 16 April 2003. On 1 July 2003, the NCSR gave its consent to the Treaty by its Decision No 365 and it classified the Treaty as falling under Article 7(5) of the Constitution, thus having primacy over domestic legislation. On 26 August 2003, the President ratified the Accession Treaty, and on 9 October 2003 it was registered with the Italian Government. On 14 April 2004, a Declaration of the Ministry of Foreign Affairs No 185/2004 Coll was published in the Collection of Laws of the SR and on 1 May 2004 the Treaty entered into force. The only major Treaty amendment that has happened since the Slovak accession was the Lisbon Treaty, signed in Lisbon on 13 December 2007. The NCSR approved the Lisbon Treaty on 10 April 2008 by its Decision No 809 and again classified the Treaty as falling under Article 7(5) of the Constitution, i.e. having primacy over the domestic legislation. The Treaty was then ratified by the President on 12 May 2008, published in the Collection on 28 November 2009 under No 486/2009 Coll, and entered into force only three days later, on 1 December 2009.

C.  Effects of International and European Law in the Slovak Legal Order The Slovak legal order recognises the primacy of international and European legal norms over the (ordinary) laws of the SR, but the Constitution remains silent on the relationship with norms of constitutional status.56 This is not unusual; constitutions rarely bow down and give priority to external sources of legal norms. From the perspective of Hans Kelsen, this would be not only undesirable, but also impossible: if a constitution explicitly identified a hierarchically superior source of law, it would confirm its subordination and, thus, would cease to be an ultimate source of validity of all legal norms in a given legal system. In this scenario, EU law would replace such constitution as highest level of positive law in that system.57 However, Juraj Čorba argues that the importance of this question should not be overrated, since it is probable that ‘in the application of the Constitution, the Slovak Constitutional Court, as well as the general courts, will lean towards euro-conform interpretation’.58 Similarly, Alexander Bröstl states that ‘the conflict of EU law and national constitutional law does not amount to Huntington’s clash of civilisations’, since the two normative systems are built on the same values and reflect the same paradigms.59 He further adds: ‘If there is something like a European legal culture and European way of looking at the world, then these must exist both at the EC/EU level and at the national level, and a potential conflict cannot be irresolvable, if these two levels cooperate together and respect each other.’60

55 By material conditions, we mean an expression of the objectives that are to be achieved through accession. The provision of Art 7(3) CSK, which is the legal basis for accession to NATO, includes such a material condition: ‘the aim of maintaining peace, security and democratic order’. See eg Procházka and Čorba (n 45) 98; or Berdisová, Constitutional Law (n 10) 16. 56 Alexander Bröstl, Ústavné právo Slovenskej republiky [Constitutional Law of the Slovak Republic] (Plzeň, Aleš Čeněk, 2013) 97: ‘Let us emphasise that the Constitution of the SR establishes primacy of legally binding acts of the EU over the laws of the SR, but not over itself.’ 57 See H Kelsen, The Pure Theory of Law (Clark NJ, The Lawbook Exchange, 2005) 193 ff and 221–24. 58 Procházka and Čorba (n 45). 59 Bröstl, Ústavné právo (n 56) 101. 60 Ibid.

710  Zuzana Vikarská and Kamil Baraník

1.  International Treaties Before the Euro Amendment of 2001, the effects of international law in domestic legal order were governed only by Article 11 of the Constitution, which read: International treaties on human rights and fundamental freedoms, which have been ratified by and promulgated in a manner laid down by law, shall have primacy over the laws of the Slovak Republic, if they guarantee a greater scope of constitutional rights and freedoms.

This was the only provision governing the relationship between international law and domestic law and it only covered treaties that guaranteed ‘a greater scope of constitutional rights’ than the Constitution.61 Academics have named this concept the ‘principle of alternative primacy of international law’ or the ‘conditional primacy of international treaties’.62 Before the Euro Amendment, no constitutional provision governed the general position of international law in the domestic legal order.63 The SKCC had an opportunity to cast more light on the interpretation of Article 11 in 1999, when deciding about a constitutional complaint of a Belarussian citizen who was criminally prosecuted in the SR and who claimed that his rights under Article 6 of the ECHR and under Article 4 of Protocol 7 to the ECHR had been violated by the Regional Court in Košice. The claimant did not invoke any of the provisions of the Slovak Constitution, and the SKCC therefore had to clarify whether it was competent to hear a constitutional complaint that only raised points under provisions of a human rights treaty. The SKCC started its (very brief) reasoning by asserting that the Constitution did not offer an unambiguous solution to the problem of the relationship between human rights treaties and the domestic legal order. It then proceeded to argue that the problem could be resolved with the help of Article 11 of the Constitution and Act No 1/1993 Coll, on the Collection of Laws of the Slovak Republic, according to which international treaties form part of the national legal order when certain conditions are met. The SKCC further stated that it is competent to decide constitutional complaints brought by natural or legal persons who claim a violation of their rights (Article 130(3) of the Constitution) and that the rights allegedly violated must be duly characterised and identified, as required by law.64 According to the SKCC, this legal provision requires a due characterisation and identification of not just any right, but a particular right guaranteed by the Slovak Constitution. To support this argument, the SKCC referred to some academic works, and then concluded: International human rights treaties have a special place in the hierarchy of legal sources in the SR. Upon fulfilment of the conditions stipulated in Article 11 of the Constitution, they have primacy over the laws of the SR. Although they have primacy over (ordinary) laws, they do not have primacy over the Constitution. They cannot be considered to be part of constitutional law. The protection of rights and freedoms therein embedded thus does not amount to a protection of constitutionality. The Constitutional Court is an institution protecting constitutionality. Therefore, the Constitutional Court is not competent to deal with the protection of rights and freedoms embedded in international instruments, if the claimant does not at the same time claim a violation of a right or freedom guaranteed by the Slovak Constitution.65 61 This formulation has also created some uncertainty: how exactly is it to be determined, which source of law guarantees ‘a greater scope of protection’? 62 Ján Klučka, ‘Medzinárodné práva a právny poriadok Slovenskej republiky’ [‘International Law and the Legal Order of the Slovak Republic’] (1993) 76 Právny obzor 295. See also Boris Balog, ‘Medzinárodné zmluvy v ústavnom systéme Slovenskej republiky’ [‘International Treaties in the Constitutional System of the Slovak Republic’], in Dny práva 2009 (Brno, Masaryk Univerzita, 2009) 2 f. 63 See Balog, ‘Medzinárodné zmluvy’ (n 62) 2. 64 Act No 38/1993 Coll, on the organisation of the Constitutional Court of the Slovak Republic, § 20. This act was later replaced by Act No 314/2018 (n 23). 65 SKCC 16.12.1999, Decision II. ÚS 91/99.

Slovakia  711 Thus, the SKCC dismissed the constitutional complaint. The SKCC also clearly identified the position of human rights treaties in the hierarchy of legal norms: they stood above ordinary laws, but below the norms of constitutional legal force. When Article 11 of the Constitution was repealed in 2001, its content was relocated to Article 7(5), which inserted into the national legal order all ratified and published international treaties and to Article 154c, which deals with treaties ratified and promulgated before the entry into force of the Euro Amendment. Article 7(5) has established the primacy of three types of international treaties: (1) treaties on human rights and fundamental freedoms; (2) self-executing treaties; and (3) treaties with direct effect; on condition that they have been ‘ratified and promulgated in a manner laid down by law.’ It follows from Article 86(d) in fine of the Constitution that it is up to the NCSR to decide whether a treaty falls into one of these three categories. If it does, it is awarded a ‘primacy clause’ and thereby also a hierarchically superior position above national statutes and other legal norms. Article 7(5) also precludes a situation where a national statute would conflict with a previously ratified treaty.66 Furthermore, treaties were added to the referential framework for the constitutionality review of legal instruments in proceedings before the SKCC, as follows from Article  125(1)(a). The SKCC cannot constitutionally review treaties ratified and published in accordance with the law. Treaties that were ratified and promulgated before 1 July 2001 (ie before the Euro Amendment entered into force) are governed by Article 154c of the Constitution. According to it, they form part of the Slovak legal order and enjoy primacy over (ordinary) laws if one of two conditions is fulfilled. They must provide either (a) ‘a greater scope of constitutional rights and freedoms’ (paragraph 1), such as the ECHR, or (b) ‘if so laid down by law’ (paragraph 2), such as the United Nations Convention on Contracts for the International Sale of Goods (No 160/1991 Coll). Thereby, the Constitution categorises the legal force of Slovak international obligations in the domestic legal order by reference to the moment of their adoption. This can be seen as paradoxical considering the rule of international law entrenched in Article 1(2).67

2.  European Union Law Next to the general provisions on the effects of international treaties in the Slovak legal order, the Constitution also contains a specific provision on the effects of EU law. The first clause of Article 7(2) of the Constitution serves as a legal basis for the transfer of competences to the EC/ EU, while the second sentence stipulates that ‘legally binding acts of the European Communities and European Union shall have primacy over the laws of the Slovak Republic’. The notion of ‘legally binding acts’ is a concept of Slovak constitutional law that does not correspond to any established concept of EU law. There is some disagreement in the academic writings about the exact meaning of this notion. According to Ján Drgonec, ‘legally binding acts’ comprise primary EU law and those parts of secondary EU law that are directly applicable, ie regulations and decisions.68 On the other hand, according to Martina Jánošíková, the notion of ‘legally binding acts of the EC/EU’ only refers to legal acts that have been ‘adopted by the EC/EU’, while primary EU law derives its effects from Article 7(5), rather than Article 7(2) of the Constitution.69 66 Unlike in Germany or in the USA. 67 Baraník, ‘Interakcia’ (n 47) 1127. 68 Ján Drgonec, ‘Ústavné a úniové právo: začneme hľadať rozumnú rovnováhu?’ [‘Constitutional and EU Law: are we Going to Find a Reasonable Balance?’] (2013) 65 Justičná revue 1463. 69 Martina Jánošíková, ‘Ústava Slovenskej republiky a členstvo Slovenskej republiky v Európskej únii’ [‘The Constitution of the Slovak Republic and the Slovak Membership in European Union’], (2014) Acta Universitatis Carolinae Iuridica 249.

712  Zuzana Vikarská and Kamil Baraník The Constitutional Court seems to have offered contradicting solutions to this dispute. In its decision on health insurance companies (discussed below), the SKCC stated that although the notion of ‘legally binding acts’ used in the second sentence of Art. 7(2) of the Constitution could give rise to problems when defining its exact scope, it can undoubtedly be concluded that the Treaty on the Functioning of the European Union is such a legally binding act.70

According to the SKCC, the TFEU falls into a ‘special subcategory of international treaties’ through which the SR has transferred competences to the EU and which are thus governed by Article 7(2) of the Constitution.71 However, treaties falling within this special subcategory still qualify as ‘international treaties to which a consent was given by the NCSR and which were ratified and promulgated in a manner laid down by law’, which means that they still belong to the referential framework of constitutional review, as is set out in Article 125 of the Constitution. However, in a more recent case on the constitutionality of data retention measures (analysed below), the SKCC stated that ‘the position of the EU founding treaties (i.e. TEU and TFEU) in the legal order of the Slovak Republic is governed by Article 1(2) and Article 7(5) of the Constitution.’72 At this stage, it is unclear whether the position of the SKCC has changed with an exchange of judges, or whether this difference amounts to an ongoing inconsistency between different chambers of the SKCC. As for the academic writings, Alexander Bröstl concludes that the disagreement about primary law falling under Articles 7(2) or 7(5) is in the end irrelevant, due to the case law of the Court of Justice.73 The effects of the abovementioned Article 7(5) can also be criticised from the perspective of EU law. The NCSR awarded the ‘primacy clause’ to the founding EU Treaties. This means that the legal instruments of primary EU law were formally given hierarchical priority over national laws. However, with regard to the effects and mechanisms of EU law, the award of such a ‘primacy clause’ is inadequate and clearly unnecessary.74 A mechanism where it is up to the national parliament to award primacy to norms of EU law has the potential of jeopardising the effectiveness of EU law as a whole, which is unacceptable from the EU perspective. Unlike other treaties that can benefit from the regime of Article 7(5), the norms of EU law are endowed with direct effect and primacy already by virtue of EU law itself. The Slovak Republic obviously cannot cherry-pick individual provisions that deserve to be given primacy over national legal norms.75 Rather, as an EU Member State, the Slovak Republic is obliged to respect all EU law provisions, irrespective of the national Parliament’s approval or disapproval. For these reasons, the model where an EU Member State attempts to define the effects of EU law in the national legal order seems to be inappropriate and contrary to the long-established case law of the Court of Justice.76 According to that case law, the position of EU law cannot be derived from national constitutions, but rather from the specificities of the supranational legal order. It is probably superfluous to point out that this case law (including the judgments in Van Gend, Costa/ENEL, or Internationale Handelsgesellschaft)77 predates the Slovak Euro Amendment by many decades. 70 SKCC PL. ÚS 3/09 (n 52) part V.3.4. 71 Ibid. 72 See SKCC 29.04.2015, Decision PL. ÚS 10/2014 (Data Retention) para 69. 73 Bröstl, Ústavné právo (n 52) 97. 74 Ján Drgonec, Ústava Slovenskej republiky – komentár [Constitution of the Slovak Republik – Commentary], 3rd edn (Bratislava, Heuréka, 2012) 69. 75 Yet again, this argument is supported by the VCLT. 76 Drgonec, ‘Ústavné a úniové právo’ (n 68) 1473. 77 ECJ 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66; ECJ 07.03.1985 32/84 (Van Gend & Loos v Inspecteur der Invoerrechten en Accijnzen) ECLI:EU:C:1985:104; ECJ 17.12.1970 11/70 (Internationale Handelsgesellschaft mbH v Einfuhr- und Vorratsstelle für Getreide und Futtermittel) ECLI:EU:C:1970:114.

Slovakia  713 Therefore, the argument that the norms of primary and secondary EU law derive their effects from the Slovak constitutional order cannot be accepted, be it from Articles 7(2), 7(5), or 125(1) of the Constitution, as these provisions are ill-fitted for this purpose. The independence of the EU legal order presumes that national constitutions remain silent about the effects of EU law, since they lack the competence to determine it.78 If a national constitution insists on acknowledging the legal effects of EU legal instruments in the national legal order, an ‘inter-system solution’ could be suggested, whereby the constitution would contain a neutral claim that the effects of EU law in that national legal order are determined exclusively by EU law itself.79 Such a solution would explicitly open the legal order to the effects of EU law without jeopardising its legal effects.

D.  European Integration Outside the EU Legal Order After 2001, all treaties, including those concerning highly important EU issues, have been domestically categorised by Article 7(4) and (5) of the Constitution. These constitutional provisions provide a flexible way of dealing with various issues of international law through the means of either direct application (the self-executing provisions of treaties), or transposition of international obligations by national legislation.80 Officially, thus, there is no special constitutional treatment of highly sophisticated instruments outside of the EU legal order such as the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union (TSCG), or the European Stability Mechanism (ESM). Politically, however, these EU issues have been routinely granted special treatment of high significance.81 Article 7(4) and (5) of the Constitution allow the NCSR to adopt specific national legislation that can indicate domestic legal expectations flowing from international obligations.82

IV.  Constitutional Limits to EU Integration A.  Transfer of ‘Rights’ vs Joining a ‘State Union’ 1.  Transfer of ‘Rights’ Pursuant to Article 7(2) The Slovak Republic made use of the abovementioned provision of Article 7(2) of the Constitution. This accession was preceded by a (facultative) referendum pursuant to Article 93(2) of the Constitution.83 Similarly, Article 7(3) of the Constitution served as the legal basis of the accession 78 Jiří Malenovský, ‘Půlstoletí od rozsudku Costa/ENEL: Je unijní právo (stále) zvláštním právním řádem?’ [‘Half Century from the Costa/ENEL decision: Is EU Law (Still) a Special Legal Order?’], (2015) 154 Právník 193, 203. 79 Ibid, 204. See for example the Hungarian Constitution of 2011, Art E[3], or Polish Constitution of 1997, Art 91(3). 80 For details see section III.C.1. 81 See the development around the adoption of the European Stability Mechanism (so called ‘Euroval no 2’) in late 2011, which was not approved by the then coalition parties in the NCSR (11.10.2011). Ultimately, this refusal led to the snap elections in 2012 and the approval of the very same mechanism in ‘the second round’ by the newly elected NCSR on 22.06.2012. 82 That was the case of the Constitutional Act No 493/2011 Coll on Budgetary Responsibility adopted on the basis of requirement of the TSCG. The international obligation was enacted in form of a Constitutional Act, which established direct obligation of a highest domestic legal value to attain balanced national budget. Ultimately, the constitutional incorporation of international obligation provides an undisputed hierarchical supremacy over all other national legal provisions. 83 Historically, the referendum on the accession of the SR to the EU has been the only legally valid referendum in Slovak constitutional history. The turnout amounted to 52.15%; 92.46% of the voters were in favour of the accession and 6.2% voted against it.

714  Zuzana Vikarská and Kamil Baraník of the Slovak Republic to NATO. This important geopolitical and legal decision was endorsed by the NCSR in April 2003 and the accession itself came into force less than a year later, on 29 March 2004. The wording of Article 7(2) is interesting in that it refers to a ‘transfer of the exercise of a part of its rights’. First, the Constitution does not allow for a transfer of the competences themselves, but merely for a transfer of their exercise.84 This is quite standard in constitutional texts. Second, the Constitution does not refer to ‘powers’ or ‘competences’, but rather to ‘rights’, which is rather unusual. It is also interesting to point out that according to one of the former drafts of Article 7(2), Slovakia could ‘transfer a part of the exercise of its sovereignty’. However, this draft wording was later changed to ‘rights’. Ján Azud has argued that the reference to ‘sovereignty’ was more appropriate, and that Article 7(2) should not refer only to the EC/EU, but rather to ‘supranational bodies’ in general.85

2.  Joining a ‘State Union’ Pursuant to Article 7(1) The provision of Article 7(1) of the Constitution deals with a more significant change in sovereignty, namely the entering into and withdrawing from a ‘state union’ with other states. Such a decision would have to be made in the form of a constitutional law, confirmed by a referendum. This provision had appeared in the original version of the Constitution in 1992, unlike other paragraphs of Article 7, which were only inserted by the Euro Amendment in 2001. The dissolution of the Czechoslovak federation in 1992 materialised without any direct involvement of Czechoslovak citizens in this crucial decision. Against these historical events, the provision of Article 7(1) was inserted into the new Slovak Constitution in order to require the most rigorous constitutional procedure for any such constitutional decisions in the future. It would require the approval by a constitutional majority in the NCSR (ie by a minimum of 90 MPs) and the support by the citizens in an obligatory referendum pursuant to Article 93(1) of the Constitution. According to the travaux préparatoires, ‘state union’ (štátny zväzok) designates an entity which stems from several states and which can be perceived as an actor in the international sphere. Such a state union should have its own constitution, own citizenship, own state symbols and own governing institutions. Already at the stage of the accession talks, some Slovak public figures voiced their concerns about the EU possibly being a state union in the sense of Article 7(1) of the Constitution. The Ministry of Foreign Affairs classified the Accession Treaty as an international treaty under Article 7(2) and (5), whereas the Ministry of Justice suggested that the situation actually fell within Article 7(1). The Government sided with the former approach and classified the Accession Treaty under Article 7(2) and (5).86 A majority of academic writings also claimed that the Union could not be defined as a ‘state union’. Viliam Karas recommended holding a

84 This is interesting in comparison with the Czech solution where academics have debated the difference between a transfer of competences and the transfer of powers. See eg J Kysela ‘K dalším důsledkům přijetí tzv. euronovely Ústavy ČR’ [‘On the other Consequences of the Euro-amendment to the Constitution of the Czech Republic’], (2002) 10 Právní rozhledy 525; and the reaction to this argument Michal Bobek, Petr Bříza and Jan Komárek, Vnitrostátní aplikace práva Evropské unie [National Application of European Union Law] (Prague, C. H. Beck, 2011) 429–36. The Slovak formulation is therefore more sensitive to Slovak sovereignty, since the CSK does not mention a transfer of powers/competences, but merely a transfer of their exercise. That means that although these powers/competences are exercised by another subject, technically they still belong to the Slovak Republic. 85 Ján Azud, ‘K otázke vplyvu vstupu do EÚ na suverenitu SR v kontexte čl. 7 Ústavy Slovenskej republiky’ [‘On the Issue of the Impact of EU Accession on the Sovereignty of the Slovak Republic in the Context of Art. 7 of the Constitution of the Slovak Republic’] (2003) 86 Právny obzor 596. 86 See Government Ordonnance No 289 of 20 April 2005 (No 7794/2005).

Slovakia  715 referendum about Slovak accession to the EU, but for historical rather than legal reasons. In 2003 he wrote: ‘It is premature to consider the European Union to be a state union, since it is not endowed with legal personality and cannot act independently in international relations.’87 Therefore, the referendum organised in May 2003 was truly a facultative referendum pursuant to Article 93(2) of the Constitution, not an obligatory one, required by Articles 7(1) and 93(1).88 The SKCC dealt with the interpretation of Article 7(1) of the Constitution in 2008 when it considered whether Slovak membership in the EU could fall within the scope of Article 7(1) rather than Article 7(2) of the Constitution. This case arose from a constitutional complaint of a number of natural persons (associated with the Conservative Institute of Milan Rastislav Štefánik) who claimed that the EU Constitutional Treaty (TECE) would transform the Union into a federation, and its ratification by the SR should, therefore, be subject to an obligatory referendum. The claimants argued that by not organising a referendum, the NCSR had breached their constitutional right to participate in the governance of public affairs. The SKCC disagreed with their claims, stating that even after the changes brought about by the Constitutional Treaty, the Union would not become a ‘state’ in the sense of Article 7(1) of the Constitution, and that the EU Treaties, including the EU Constitutional Treaty, were correctly subjected to the regime of Article 7(2). According to the SKCC, the developments in the European Union undoubtedly proceed in the direction of a future state, i.e. a state union, but the Constitutional Court believes that at present, it is impossible to predict, at which concrete moment this transformation will take place. The evidence in these proceedings revealed that the European Union shows already at present a whole range of features and functions, which can be seen … as features of a state union.

Having said that, the SKCC concluded that all the future acts and steps that the SR will take in the framework of its relationship with the … European Union, must be governed by the provision of Art. 7(2) of the Constitution. This would not be different even if the act were such as to change the qualitative parameters of the cooperation between the members of the given organisation, which is the case of the [Constitutional] Treaty.89

This conclusion is rather surprising, since it postulates either that the EU can never in the future become a ‘state union’ in the meaning of Article 7(1) of the Constitution, or that even if it one day were to become such ‘state union’, the Constitution would nevertheless treat it, irrespective of the qualitative change, as a mere ‘international organisation’ within the meaning of Article 7(2) and without the need for a referendum. Both statements appear to be highly speculative and ought to be treated as a mere obiter, hardly apt to fix the situation immutably for the future. In fact, we consider this approach of the SKCC to be rather peculiar. Unlike other constitutional courts, such as those in Germany, Italy or the Czech Republic, which carefully protect their own competence to decide whether the Union acts within its delimited area of action (the so-called ‘ultra vires review’) and whether it respects the constitutional identities of the member states (the 87 Viliam Karas, ‘Právna povaha Európskych spoločenstiev a Európskej únie a ostatná novela Ústavy SR’ [‘The Legal Nature of the European Communities and the European Union and Other Amendments to the Constitution of the SR’], (2003) 55 Justičná revue 373. Similarly see Radoslav Procházka. ‘Referendum a ratifikácia zmluvy o pristúpení k EÚ’ [‘Referendum and Ratification of the Accession Treaty’], (2003) 55 Justičná revue 119. 88 For more clarifications, as well as for thought-provoking issues of the EU withdrawal process initiated by a referendum, see Marek Domin, ‘Krátke zamyslenie nad potenciálnym referendom o vystúpení Slovenskej republiky z Európskej únie’ [‘A Brief Reflection on the Potential Referendum on the Withdrawal of the Slovak Republic from the European Union’], (2017) 69 Justičná Revue 1042. 89 SKCC 27.02.2008, Decision II. ÚS 171/05.

716  Zuzana Vikarská and Kamil Baraník so-called ‘identity review’), the SKCC seems to have explicitly given up such a competence and to have left the decision about the nature of the EU and the limits of its action to the EU itself and its institutions. Subsequently, the SKCC concluded that irrespective of the character of the EU, neither the accession of the SR to the EU nor any other act initiated either by the EU or by the SR could lead to a situation that would coincide with the aim or the content of Art. 7(1) of the Constitution.90

This part of the SKCC’s reasoning is truly surprising: the Constitution guarantees for the Slovak citizens in Article 7(1) of the Constitution that joining a state union requires the adoption of a constitutional law and the declaration of a referendum, while the SKCC declares that in relation to the EU, Article 7(1) simply cannot be applicable, no matter what the future brings. Does that mean that even if the EU became a federation, or even if it transformed into a unitary state denying the national identities of the individual European nation-states, this state entity would never amount to a state union in the sense of Article 7(1), which would require the adoption of a constitutional statute and a referendum? Such an interpretation of Article 7(1) is hard to defend. We therefore believe that the SKCC has made two mistakes in the analysed judgment. First, it should not have let the decision about the EU’s character to the EU itself and second, it should not have completely excluded the applicability of Article 7(1) in relation to the EU, especially not after it claimed that the Union undoubtedly proceeds in the direction of a future state. We put forward that the consequences of this constitutional doctrine adopted by the SKCC are truly unfavourable for the protection of Slovak sovereignty in the future. In other words, if the EU ever endangers the sovereignty of the Slovak Republic, the SKCC will not be equipped with any instruments, rhetorical or other, to protect that sovereignty (unless it changes its own case law). The analysed judgment can therefore be seen as a missed opportunity. Both the claimants and the NCSR expected the SKCC to set out the criteria for determination of the EU’s constitutional nature, not only in the context of the TECE, but also for the future. The SKCC, however, did not manage to offer a persuasive solution to the national perception of the ever-deeper European integration. Trusting the EU is certainly not problematic as such. However, the SKCC has set its trust at a complete maximum (ie the constitutional blindness), which simply seems unusual, especially in comparison with other constitutional courts in the region.

B.  Scrutiny of Secondary Legislation 1.  Unconstitutionality of the Anti-Discrimination Act (PL. ÚS 8/04)91 One of the first major rulings of the SKCC containing a significant EU dimension was the case in which the SKCC declared § 8(8) of the Anti-Discrimination Act to be unconstitutional.92 This provision did not directly implement EU law, but was a voluntary extension, foreseen by Article 5 of the Equal Treatment Directive.93 The wording of § 8(8) of the Act was as follows: ‘In order to establish equality of opportunities in practice, and to comply with the principle of 90 Ibid. 91 SKCC 18.10.2005, Decision PL. ÚS 8/04. 92 Act No 365/2004 Coll, on equal treatment in some areas and on protection against discrimination. 93 Council Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment between persons irrespective of racial or ethnic origin, [2000] OJ L 180/22.

Slovakia  717 equal treatment, it shall be possible to adopt special positive measures to eliminate inequalities based on racial origin or ethnic origin.’ The Slovak Government believed this to be in conflict with a number of constitutional provisions, and thus decided to challenge the provision before the SKCC. The NCSR in fact encouraged the SKCC to refer a question for a preliminary ruling to the ECJ. Without asking the CJEU for advice, the SKCC partially agreed with the Government’s argument and declared § 8(8) to be in conflict with the principle of the ‘rule of law’, and the principle of equal treatment, embedded respectively in Articles 1(1) and 12 of the Constitution. The main objection that the SKCC raised against the challenged provision was its broad and vague formulation, which failed to clearly formulate the aims of the measure and to emphasise that positive measures must be limited in time. The provision was thus found to be in breach of the ‘rule of law’ principle, as guaranteed by Article 1(1), due to violation of legal certainty.94 Although the Act challenged by the Government had its origin in an EU Directive, the euroanalysis presented by the SKCC was rather limited. The SKCC referred to only two cases decided by the ECJ95 and stated that even the doctrine of the Court of Justice shows that derogations from the general principle of equality must be limited to what is inevitable and proportionate in order to reach material equality through positive measures. Therefore, such positive measures must be interpreted restrictively.96

This very limited Euro analysis shows that at the time, the SKCC was still not ready to enter into a more detailed discussion of the relationship between domestic and international/European law.97

2.  Unconstitutionality of the Act on Health Insurance Companies (PL. ÚS 3/09)98 In October 2008, a number of MPs challenged the constitutionality of two provisions of the law on Health Insurance Companies.99 These two provisions, namely § 15(6) and § 86d, were introduced by an amendment, and they very much changed the legal status and economic position of health insurance companies. The claimants argued that these provisions breached the Constitution (the provisions on the ‘rule of law’ principle and the prohibition of retroactivity, the right to own property, the right to conduct business, and the provision on the protection of competition), the ECHR (Article 14, and Article 1 of Protocol No 1), and primary EU law (Articles 18, 49, 54 and 63 TFEU). The SKCC reviewed the statute’s constitutionality in light of property rights. Ultimately, it upheld the constitutional challenge and declared the provisions unconstitutional and in breach of Article 1 of Protocol No 1 to the ECHR. The Court did not assess the compatibility of the challenged provisions with primary EU law, but articulated interesting ideas relating to the position of EU law in the Slovak legal system. According to the SKCC, this was the first case, in which it was confronted with a request to rule on the conformity of a piece of legislation with primary EU law. Therefore, it used the 94 SKCC PL. ÚS 8/04 (n 91) 22. 95 ECJ 15.05.1984 222/84 (Johnston v Chief Constable of the Royal Ulster Constabulary) ECLI:EU:C:1986:206; and ECJ 11.01.2000 C-285/98 (Kreil) ECLI:EU:C:2000:2. 96 SKCC PL. ÚS 8/04 (n 91). 97 See Zuzana Antošová, ‘Ústavný súd SR po vstupe do Európskej Únie’ [‘The Constitutional Court of the Slovak Republic After EU Accession’], (2013) 5 Historia et Theoria Iuris 8, 13. 98 SKCC PL. ÚS 3/09 (n 52). 99 Act No 581/2004 Coll.

718  Zuzana Vikarská and Kamil Baraník opportunity and articulated some general principles in this field.100 First, the SKCC quoted parts of the Court of Justice rulings in Costa101 and Simmenthal,102 explaining the principle of primacy of EU law over domestic law and its practical effects for ordinary courts. If a general court is confronted with a conflict of a domestic legal norm and a provision of EU law, the judge is under a duty to give full effect to [provisions of EU law], if necessary, refusing of its own motion to apply any conflicting provision of national legislation, even if adopted subsequently, and it is not necessary for the court to request or await the prior setting aside of such provision by legislative or other constitutional means.103

The SKCC added that a general court is also entitled to suspend the proceedings and refer a question to the Court of Justice, if necessary.104 The proceedings at stake, however, did not originate before ordinary courts. Rather, they were initiated directly before the SKCC by a group of MPs under the abstract constitutional review procedure. In such a setting, the SKCC declared itself to have the competence to assess the compatibility of a domestic legal norm with international treaties ‘to which consent was given by the NCSR and which were ratified and promulgated in a manner laid down by law’, as is provided in Article 125 of the Constitution. Therefore, the SKCC found its subject-matter jurisdiction by treating primary EU law as general international law: in other words, if international treaties (as defined by Article 125) belong to the ‘framework of reference’ for review, then primary EU law also belongs to that framework of reference. However, in this context, it remains unclear why the SKCC only formulated its jurisdiction in relation to primary EU law and not in relation to directly applicable treaties in general. After having declared its competence to rule on the matter, the SKCC carried on distinguishing three hypothetical situations in which national legislation could be challenged in light of EU law. In the first situation, the claimant would challenge, in parallel, the conformity of a national norm with both the Constitution and with EU law. In relation to this scenario, the SKCC stated: When assessing the alleged incompatibility of a domestic legal norm with EU law, a question of usefulness arises, if in that same proceedings the Constitutional Court concludes that the challenged legal norm … is incompatible with the Constitution … If the Constitutional Court, in proceedings pursuant to Art 125(1)(a) of the Constitution, declares that a domestic legal norm is incompatible with the Constitution, it is no longer necessary to assess its incompatibility with EU law (although the claimants wish so), because the potential incompatibility would lead to the same result and the same legal effects as have already been achieved by the decision according to which the challenged legal norm is incompatible with the Constitution. The Constitutional Court explains this ‘self-restraining’ approach to the exercise of its competence by reference to the fact that after declaring incompatibility with the Constitution, the proceedings are in fact devoid of their subject matter.

To sum up, once the SKCC has declared the unconstitutionality of a legal norm, it will no longer assess the alleged incompatibility with EU law, since the potential incompatibility with 100 See also Gabriela Dobrovičová and Martina Jánošíková, ‘Ústava Slovenskej Republiky a primárne právo Európskej Unie’ [‘The Constitution of the Slovak Republic and EU Primary Law’], in L Orosz, M Breichová Lapčáková and T Majerčák (eds), 20 rokov Ústavy Slovenskej republiky – I. ústavné dni (Košice, Univerzita Pavla Jozefa Šafárika, 2012) 185, 186 f. 101 ECJ 6/64 (Costa v E.N.E.L.) (n 77). 102 ECJ 09.03.1978 106/77 (Amministrazione delle Finanze dello Stato v Simmenthal SpA) ECLI:EU:C:1978:49. 103 Ibid, para 24 as quoted in SKCC PL. ÚS 3/09 (n 52) part V.3.4. 104 Similarly, more recent SKCC 17.06.2015 Decision PL. ÚS 37/2015, 10–11, where the Court confirmed that in the case of a conflict between domestic law and primary EU law, general courts are obliged by EU law itself to disapply that rule and, therefore, they cannot turn to the SKCC with a request to assess its constitutionality, since the SKCC is only competent to assess the constitutionality of a rule which is to be applied in a given case.

Slovakia  719 EU law would only lead to the same result as the (already established) incompatibility with the Constitution. The situation would, however, be different if the claimant only invoked incompatibility of the domestic legal norm with EU law (or a treaty), without claiming unconstitutionality of the provision. In this second situation, the SKCC would feel obliged to investigate the alleged incompatibility with EU law. In the words of the SKCC, it would be not only entitled, but in fact obliged … to assess the alleged incompatibility and to decide on it, either on the basis of the existing case law of the Court of Justice and the principles articulated therein, or after having referred a question to the Court of Justice, pursuant to Art 267 TFEU.105

This is not too controversial in light of the above-mentioned conclusion that primary EU law belongs to the framework of reference for constitutional review, by virtue of belonging to the group of ‘international treaties to which consent was given by the NCSR and which were ratified and promulgated in a manner laid down by law’, as foreseen by Article 125 of the Constitution. Third and most though-provokingly the situation would appear if the claimant invoked a conflict both with the Constitution and with EU law, but the SKCC would not find an incompatibility of the challenged legal norm with the Constitution. In such a case, the SKCC would feel obliged to decide about the alleged conflict, if the conflict cannot be resolved by applying the principles of euro-conform interpretation. In such a case, if it is authoritatively found that the challenged legal norm is incompatible with an international treaty under Art. 7(2) of the Constitution, a question may arise whether such situation does not require a constitutional change (which, however, does not belong to the competences of the Constitutional Court).106

This third scenario is in fact only a variation of the second scenario and they can both be summarised as follows. Even if there is no conflict with the Constitution (irrespective of whether it was claimed by the parties or not), the SKCC may review the validity of a challenged measure also in light of primary EU law. The SKCC’s reasoning is not controversial: It follows from the constitutional text itself, since the framework of reference is not limited only to the Slovak Constitution, but it also encompasses certain treaties to which EU law certainly belongs. The added value of the analysed case, therefore, only lies in the ‘order’ that the SKCC has established: Constitution first, EU law second (if still necessary). While not controversial, the reasoning remains slightly confusing. The SKCC has only addressed the position of primary EU law, leaving aside the status of other treaties foreseen by Article 125, although the Constitution clearly speaks of a much broader category of treaties than just the EU Treaties falling under Article 7(2) of the Constitution. Furthermore, as for the position of secondary EU law, the SKCC stated on another occasion that ‘a directive is not part of the referential framework in the constitutional review procedure’.107 Last but not least, the nudge at the end of the third scenario, suggesting the need for a constitutional change, seems rather unnecessary. If EU law requires more than the Constitution does, why would that be a reason to change the Constitution?108 105 Ibid. In SKCC IV. ÚS 206/08 (n 31) the Court acknowledged that it is itself a ‘court’ in the sense of Art 267 TFEU (then Art 234 TEC), see above. 106 SKCC PL. ÚS 3/09 (n 52) part V.3.4. 107 SKCC 02.06.2010, Decision PL. ÚS 8/2010. 108 On the other hand, the SKCC probably suggested that the compatibility of a statute with the Constitution and its parallel incompatibility with an international treaty could potentially create a tension between the Constitution and that treaty. However, if this was the hypothetical scenario, it must be pointed out that the SKCC did not offer a clear solution to such a situation. It can only be deduced from the reasoning in the analysed case that the SKCC would probably appeal

720  Zuzana Vikarská and Kamil Baraník It can be concluded that although this judgment remains (so far) the most important case dealing with the position of EU law in the Slovak legal order, it is far from comprehensive. The solutions outlined by the SKCC are rather simplistic, leaving out many more possible scenarios which could occur. Moreover, the principles formulated by the SKCC are found in the obiter dictum, ie the non-binding part of the decision. Therefore, not only does this case not offer a clear answer to the status of EU law in the national legal order, but it also does not bear precedential value for future cases of collisions between EU law and national law.

3.  Unconstitutionality of the Act on Electronic Communications (PL. ÚS 10/2014)109 The (in)famous Data Retention Directive,110 which was annulled by the CJEU in April 2014,111 has not only stirred waters at the EU level but has also given rise to a very interesting judgment of the SKCC. In October 2012, a group of Slovak MPs challenged a number of provisions of the Act on Electronic Communications, the Code of Criminal Procedure, and the Act on the Police Forces, for their alleged incompatibility with the Constitution, the Convention, and the Charter of the Fundamental Rights of the European Union (CFR). Although the claimants requested the SKCC to submit a reference for a preliminary ruling to the CJEU, the SKCC has decided simply to wait for the result of the pending case in Digital Rights Ireland. Two weeks after the CJEU’s judgment, the SKCC issued a partial decision in the case, by which it suspended the legal effects of the challenged provisions until the substantive decision in the case.112 This substantive decision was delivered one year later, on 29 April 2015. The SKCC opened its analysis by discussing the referential framework of constitutional review, acknowledging that even after the accession of the SR to the EU, the referential framework of the constitutional review remains limited to the norms of the Slovak constitutional order. However, the Constitutional Court cannot disregard the effects of EU law on the creation, application and interpretation of domestic norms in the fields of regulation that have their origin in EU law.113

The SKCC then elaborated on the principles of the scope of application of EU law, referring to Article 51(1) of the EU Charter and the explanatory notes thereto, as well as to the judgments of the CJEU in cases such as Wachauf,114 ERT,115 Annibaldi,116 Karner,117 DEB,118 Fransson,119 Ymeraga,120 and Siragusa.121 On the basis of this case law, the SKCC concluded that even after the the Parliament to change the CSK in order to bring it in line with the international obligations, as the CSK requires. This, however, is merely our interpretation of the SKCC’s reasoning. 109 SKCC PL. ÚS 10/2014 (n 72). 110 Directive 2006/24/EC of the European Parliament and of the Council of 15 March 2006 on the retention of data generated or processed in connection with the provision of publicly available electronic communications services or of public communications networks and amending Directive 2002/58/EC, OJ L 105, 13.4.2006, 54. 111 See CJEU 08.04.2014 Joined Cases C-293/12 and C-594/12 (Digital Rights Ireland and Seitlinger and Others) ECLI:EU:C:2014:238. 112 See SKCC 23.04.2014, Resolution PL. ÚS 10/2014, published in the Collection of Laws under No 128/2014 Coll 113 See PL. ÚS 10/2014 (n 72) para 62. 114 ECJ 13.07.1989 5/88 (Hubert Wachauf v Bundesamt für Ernährung und Forstwirtschaft) ECLI:EU:C:1989:321. 115 ECJ 18.06.1991 C-260/89 (ERT v DEP) ECLI:EU:C:1991:254. 116 ECJ 18,12,1997 C-309/96 (Annibaldi v Sindaco del Comune di Guidonia and Presidente Regione Lazio) ECLI:EU:C:1997:631. 117 ECJ 25.03.2004 C-71/02 (Karner) ECLI:EU:C:2004:181. 118 CJEU 22.12.2010 C-279/09 (DEB) ECLI:EU:C:2010:811. 119 CJEU 07.05.2013 C-671/10 (Åkerberg Fransson) ECLI:EU:C:2013:105. 120 CJEU 08.05.2013 C-87/12 (Ymeraga and Ymeraga-Tafarshiku) [2013] ECLI:EU:C:2013:291. 121 CJEU 06.03.2014 C-206/13 (Siragusa) ECLI:EU:C:2014:126.

Slovakia  721 annulment of the Data Retention Directive the domestic implementation thereof still remains to be seen as a measure adopted within the scope of application of EU law, since it represents derogation from the E-Privacy Directive,122 namely its Article 15(1).123 The SKCC therefore concluded that in the constitutional review of the challenged decisions … the Constitutional Court must also take into consideration the relevant provisions of the Charter, especially Arts. 7, 8, and 52(1) thereof and the relevant case law of the Court of Justice of the EU.124

In light of the SKCC’s decision in case PL. ÚS 3/09 analysed above, this is rather surprising. If primary EU law was explicitly included in the framework of reference in the previous case, why does the SKCC later claim that the referential framework remains limited to the Slovak Constitution? This contradicts not only the previous case law but, in fact, also the text of Article 125 of the Constitution. It also contradicts the reasoning found in the following part of the very same judgment. After excluding EU law from the referential framework, the SKCC pulls it back in: The Charter has not been adopted in the form of an international treaty; however, by the coming into effect of the Lisbon Treaty, it became a legally binding part of EU primary law, having the same legal value as the Treaties … The status of the EU founding treaties (TEU and TFEU) in the Slovak legal order is regulated by Arts. 1(2) and 7(5) of the Constitution. Therefore, the EU Charter, having the same legal value as the EU founding treaties, must be awarded in the Slovak constitutional order the same status as is awarded to the international treaties listed in Art. 7(5).

Further on, the SKCC continued: for the purpose of review of legal norms pursuant to Art. 125(1) of the Constitution, the framework of reference includes not only the Constitution, but also international treaties to which consent was given by the NCSR and which were ratified and promulgated in a manner laid down by law,125

whereas fundamental rights and freedoms must be interpreted in light of human rights treaties and the relevant case-law pertaining to these treaties.126 At this stage, the SKCC repeated its ruling from case PL. ÚS 3/09, namely the order of review in which the Constitution comes first and only in case that the challenged norm stands the review in light of the Constitution, the SKCC will continue to review it in light of primary EU law. As has been stated above, such a mechanism is not controversial when seen in the context of the wording of Article 125 of the Constitution. However, what remains unanswered is the status of other norms of EU law in the national legal order, as well as the mutual relationship between the norms of EU law and the national norms of constitutional status. These questions still remain to be addressed in the future decisions of the SKCC.

V.  Constitutional Rules and Practice on Implementing EU Law It follows from the Preamble to Protocol (No 1) on the Role of National Parliaments in the European Union, annexed to the TEU and the TFEU, that ‘the way in which national Parliaments 122 Directive 2002/58/EC of the European Parliament and of the Council of 12 July 2002 concerning the processing of personal data and the protection of privacy in the electronic communications sector (Directive on privacy and electronic communications), [2002] OJ L 201/37. 123 See PL. ÚS 10/2014 (n 72) para 68. 124 Ibid. 125 Ibid, para 70. 126 Ibid, para 74.

722  Zuzana Vikarská and Kamil Baraník scrutinise their governments in relation to the activities of the Union is a matter of constitutional organisation and practice of each Member State.’ In Slovakia, the relationship between the legislative and the executive branch has gone through some changes due to the accession to the EU and it is now governed by Constitutional Act 397/2004 Coll on the cooperation of the NCSR and the Government of the SR in EU matters.

A.  The Role of the Parliament and the Government in EU Matters Three instruments define the role of the NCSR in EU matters. These are the Constitution, Constitutional Act No 397/2004 Coll on the cooperation of the NCSR and the Government of the SR in EU matters, and Act No 350/1996 Coll on the rules of procedure of the NCSR.127 At the moment of accession, the involvement of the NCSR and the cooperation between the NCSR and the Government in EU matters had not been resolved. This gap was filled shortly after accession, in June 2004, by the adoption of Constitutional Act No 397/2004 Coll. It is a very short Act, consisting of only four articles, aiming primarily to define the respective powers of the NCSR and the Government in EU matters. Pursuant to Article 1, the Government has an obligation to inform the NCSR about the current EU agenda, about any draft legislation debated at the EU level, and about any draft opinions (draft EU legislation) or positions that are to be presented in the name of the SR. Pursuant to Article 2, the NCSR may endorse any such draft opinions, and any other opinions on EU matters. If the NCSR does not issue an opinion within two weeks, it is presumed that it has endorsed that opinion. Furthermore, Article 2(4) states that a parliamentary decision shall be legally binding for any member of the Government. Article 2(5) allows for an exception from this principle in pressing cases, in which case a member of the government must act in the interest of the SR without delay, but is subsequently obliged to inform the NCSR of the action. Constitutional Act No 397/2004 Coll further requires parliamentary deliberation on EU matters to take place at least once a year, on the basis of a report proposed by the Government. This deliberation should result in a set of recommendations given by the NCSR to the Government for the subsequent year. Pursuant to Article 3 of the Constitutional Act, any further details on the cooperation of the NCSR and the Government shall be laid down by law. Moreover, the Government has adopted its own resolution, the so-called Revised Mechanism of Preparation of Positions in October 2007, which regulates further details. As for the practical organisation of the EU agenda in the NCSR, Article 45 of the Rules of Procedure of the NCSR allows the NCSR to create specialised committees. On the basis of this provision, the NCSR created the European Affairs Committee (EAC) in April 2004. Before the creation of the EAC, EU matters were administered by the Foreign Committee, the Constitutional and Legal Affairs Committee, and the Committee for European Integration. The Rules of Procedure of the NCSR were then amended to include a separate provision on the EAC. This was Article 58a, which regulates the composition of the Committee (proportionate representation of all the parliamentary political parties), its competences and various procedural matters. The EAC currently consists of 15 members and 15 substitute members; it functions continuously (unlike other committees) and is one of the busiest parliamentary committees. However, there is quite a difference between the formally strong position of the EAC and its actual weight. The main

127 For a detailed study of the respective roles of the Parliament and the Government in EU matters, see Vladimír Bartovic, ‘National Council of the Slovak Republic in the EU Agenda: Giant in Theory, Dwarf in Practice’, in D Král and V Bartovic (eds), The Czech and the Slovak Parliaments after the Lisbon Treaty (Prague, Europeum, 2010) 47.

Slovakia  723 problems include lack of interest in the EU agenda, low administrative and expert capacity to handle the EU agenda, late involvement of the Parliament in the creation of positions, as well as the fact that the EAC members are simultaneously members of other parliamentary committees, which leaves them with little time to focus on the EAC agenda.128 The work of the EAC is complemented by the Department for European Affairs of the NCSR, which serves as the Committee’s secretariat.

B.  Techniques of Implementation of EU Law Pursuant to the third clause of Article 7(2) of the Constitution, ‘the enactment of legally binding acts that require implementation shall be executed by law or a government ordinance pursuant to Art 120(2).’ There are thus two modalities of implementation of EU law: a legislative technique, whereby a law is passed by the NCSR, and an executive technique, whereby the Government passes an ordinance. As for the legislative technique, there are no special provisions relating to the process of implementation of EU legal norms. The NCSR simply uses its legislative competence in order to adopt a domestic law that fulfils the aims outlined by the EU legislation. The EU provisions are often so detailed and precise that the Parliament merely copy-pastes the Slovak language version of the EU instrument into the implementing legislation. Academic authors have therefore spoken of the so-called ‘de-parliamentarisation’, which results in the Parliament stepping back from the decision-making processes and taking up the role of a mere observer.129 As for the executive technique, these so-called ‘government ordinances’ have been subject of various academic discussions, due to their questionable legitimacy and due to some uncertainty concerning their legal force (is their legal force equivalent to ordinary laws, or rather to other government ordinances?), but they certainly do offer a very effective way of implementing EU law.130 Constitutional Act No 397/2004 Coll undoubtedly strengthened the position of the NCSR in EU matters. However, it needs to be pointed out that the Government played a major role in the approximation of the Slovak legal system to the EU before accession and still plays a major role in the transposition of certain EU legal acts post-accession. Pursuant to the third clause of Article 7(2) of the Constitution, and pursuant to Article 120(2) of the Constitution, the Government has ‘the power to issue regulations to implement laws’. This power is regulated further by Act No 19/2002, on the conditions of issuing governmental regulations, which defines in its Article 2(1) the specific areas in which a governmental regulation may substitute a legislative transposition of EU law measures. These areas are custom duties, banking legislation, corporate accountancy and taxation, intellectual property, protection of workers, financial services, consumer protection, technical regulations and norms, use of nuclear power, transportation, agriculture, environment and free movement of labour. In practice, this brings a major decrease to the parliamentary workload and shifts the burden of transposition from the legislative to the executive branch. It also proves that the transposition of EU norms is seen as a technical issue, rather than a politically sensitive one. However, pursuant to Article 2(2) of the Act, such governmental regulations cannot regulate fundamental rights

128 Ibid. The title of this expert report gives a clear indication of this. 129 See eg Aleš Gerloch et al, Teorie a praxe tvorby práva [Theory and Practice of Law Making] (Prague, ASPI, 2008) 294 ff; or Antošová, ‘Ústavný súd’ (n 97) 11. 130 See Gabriela Dobrovičová, Vybrané otázky recentných prameňov práva v Slovenskej republike [Selected Issues of Recent Sources of Law in the Slovak Republic] (Košice, Univerzita Pavla Jozefa Šafárika, 2004) 22 f.

724  Zuzana Vikarská and Kamil Baraník and freedoms, state budget and other matters where the Constitution requires a law and cannot create new state institutions.

VI.  Resulting Relationship of EU Law and National Law The Slovak legal order, including the SKCC’s case law, finds itself somewhere between a genuine desire to be very open to everything that comes from the European Union, and a cumbersome inability to embrace the principles of EU law coherently and to find its proper place in the national legal order. On the one hand, the SKCC has embraced primary EU law norms in the constitutional framework of reference. On the other hand, it keeps emphasising the key role of the constitutional provisions therein. This could be seen as respect for the principle of sovereignty. However, one could also be rather sceptical and conclude that the SKCC itself is not entirely sure about the interactions and mutual relationships of various sources of norms in a pluralist legal order. Since its accession to the European Union in 2004, Slovakia has shown willingness to cooperate with the EU and its legal order. However, both the constitutional text and the case law that follows remain simplistic, confusing, and occasionally also contradictory. Regrettably, the Slovak legal scholarship has been rather scarce and scant in addressing these intriguing questions. Therefore, the tale of the effects of EU law in the Slovak legal order is yet to be written and it remains to be seen whether the newly composed ‘fourth court’ will shatter the foundations set in the previous years or will rather continue the path already started and (hopefully) clarify it.

References Z Antošová, ‘Ústavný súd SR po vstupe do Európskej Únie’ [‘The Constitutional Court of the Slovak Republic After EU Accession’], (2013) 5 Historia et Theoria Iuris 8. J Azud, ‘K otázke vplyvu vstupu do EÚ na suverenitu SR v kontexte čl. 7 Ústavy Slovenskej republiky’ [‘On the Issue of the Impact of EU Accession on the Sovereignty of the Slovak Republic in the Context of Art. 7 of the Constitution of the Slovak Republic’] (2003) 86 Právny obzor 596. B Balog, ‘Medzinárodné zmluvy v ústavnom systéme Slovenskej republiky’ [‘International Treaties in the Constitutional System of the Slovak Republic’], in Dny práva 2009 (Brno, Masaryk Univerzita, 2009). K Baraník, ‘Interakcia medzinárodného, supranacionálneho a ústavného práva pri ochrane ľudských práv v SR’ [‘Interaction of International, Supranational and Constitutional Law in the Protection of Human Rights in the Slovak Republic’] (2017) 69 Justičná Revue 1118. K Baraník ‘Why Have Constitutional Courts Been so Important for Democracy in Central Europe (… And So Hated by Those in Power)?’ (2018) 11 Juridiskā zinātne 77. K Baraník and M Giba, ‘Medzinárodné zmluvy a Ústavný súd Slovenskej republiky v treťom funkčnom období’ [‘International Treaties and the Constitutional Court of the Slovak Republic in the Third Term of Office’], in L Orosz et al (eds), Tretie funkčné obdobie Ústavného súdu Slovenskej republiky – VII. ústavné dni (Košice, Univerzita Pavla Jozefa Šafárika, 2019) 241. K Baraník, ‘Ústava na hvězdném nebi. Vztah Ústavy Slovenské republiky a mezinárodnému právu’ [‘The Constitution in the Starry Sky. The Relationship of the Constitution of the Slovak Republic and International Law’] (Prague, Leges, 2020). V Bartovic, ‘National Council of the Slovak Republic in the EU Agenda: Giant in Theory, Dwarf in Practice’, in D Král and V Bartovic (eds), The Czech and the Slovak Parliaments after the Lisbon Treaty (Prague, Europeum, 2010) 47.

Slovakia  725 L Berdisová, Constitutional Law (Trnava, Trnavská univerzita v Trnave. Právnická fakulta, 2013), http:// iuridica.truni.sk/sites/default/files/dokumenty/zahranicne-vztahy/en/publications/pdf/02Consti tutional%20Law.pdf. M Bobek, P Bříza and J Komárek, Vnitrostátní aplikace práva Evropské unie [National Application of European Union Law] (Prague, C.H. Beck, 2011). M Bobek and Z Kühn, ‘Europe Yet to Come: The Application of EU Law in Slovakia’, in A Lazowski (ed), The Application of EU Law in the New Member States – Brave New World (The Hague/Cambridge, TMC Asser Press/Cambridge University Press, 2010) 373. A Bröstl, Ústavné právo Slovenskej republiky [Constitutional Law of the Slovak Republic] (Plzeň, Aleš Čeněk, 2013). G Dobrovičová, Vybrané otázky recentných prameňov práva v Slovenskej republike [Selected Issues of Recent Sources of Law in the Slovak Republic] (Košice, Univerzita Pavla Jozefa Šafárika, 2004). G Dobrovičová and M Jánošíková, ‘Ústava Slovenskej Republiky a primárne právo Európskej Unie’ [‘The Constitution of the Slovak Republic and EU Primary Law’], in L Orosz, M Breichová Lapčáková and T Majerčák (eds), 20 rokov Ústavy Slovenskej republiky – I. ústavné dni (Košice, Univerzita Pavla Jozefa Šafárika, 2012) 185. M Domin, ‘Krátke zamyslenie nad potenciálnym referendom o vystúpení Slovenskej republiky z Európskej únie’ [‘A Brief Reflection on the Potential Referendum on the Withdrawal of the Slovak Republic from the European Union’], (2017) 69 Justičná Revue 1042. J Drgonec, Ústava Slovenskej republiky – komentár [Constitution of the Slovak Republik – Commentary], 3rd edn (Bratislava, Heuréka, 2012). J Drgonec, ‘Ústavné a úniové právo: začneme hľadať rozumnú rovnováhu?’ [‘Constitutional and EU Law: are we Going to Find a Reasonable Balance?’] (2013) 65 Justičná revue 1463. A Gerloch et al, Teorie a praxe tvorby práva [Theory and Practice of Law Making] (Prague, ASPI, 2008). M Jánošíková, ‘Ústava Slovenskej republiky a členstvo Slovenskej republiky v Európskej únii’ [‘The Constitution of the Slovak Republic and the Slovak Membership in European Union’], Acta Universitatis Carolinae Iuridica (2014) 249. V Karas, ‘Právna povaha Európskych spoločenstiev a Európskej únie a ostatná novela Ústavy SR’ [‘The Legal Nature of the European Communities and the European Union and Other Amendments to the Constitution of the SR’], (2003) 55 Justičná revue 373. H Kelsen, The Pure Theory of Law (transl M Knight) (Clark NJ, The Lawbook Exchange, 2005). J Klučka, ‘Medzinárodné práva a právny poriadok Slovenskej republiky’ [‘International Law and the Legal Order of the Slovak Republic’] (1993) 76 Právny obzor 295. S Krasner, Sovereignty: Organized Hypocrisy (Princeton NJ, Princeton University Press, 1999). J Kysela ‘K dalším důsledkům přijetí tzv. euronovely Ústavy ČR’ [‘On the other Consequences of the E ­ uroamendment to the Constitution of the Czech Republic’], (2002) 10 Právní rozhledy 525. T Langášek, Ústavní soud Československé republiky a jeho osudy v letech 1920–1948 [The Constitutional Court of the Czechoslovak Republic and Its Destiny in the Period 1920–1948], www.usoud.cz/ustavni-soudcsr-1920–1948/. J Malenovský, ‘K nové doktríně Ústavního soudu ČR v otázce vztahů českého, komunitárního a mezinárodního práva’ [‘On the New Doctrine of the Constitutional Court of the Czech Republic in the Relationship between Czech, Community and International Law’] (2006) 14 Právní rozhledy 782. J Malenovský, ‘Půlstoletí od rozsudku Costa/ENEL: Je unijní právo (stále) zvláštním právním řádem?’ [‘Half Century from the Costa/ENEL decision: Is EU Law (Still) a Special Legal Order?’], (2015) 154 Právník 193. R Procházka, Mission Accomplished: On Founding Constitutional Adjudication in Central Europe (Budapest, Central European University Press, 2002). R Procházka. ‘Referendum a ratifikácia zmluvy o pristúpení k EÚ’ [‘Referendum and Ratification of the Accession Treaty’], (2003) 55 Justičná revue 119. R Procházka and J Čorba, Právo Európskej únie [European Union Law] (Žilina, Poradca podnikateľa, 2006).

726  Zuzana Vikarská and Kamil Baraník V Schorm, ‘Osud vládního návrhu novely Ústavy v oblasti mezinárodních vztahů, Evropské unie a Severoatlantické aliance’ [‘The Fate of the Government‘s Draft Amendment to the Constitution in the Area of International Relations, the European Union and the North Atlantic Alliance’], in B Dančak, V Šimiček (eds), Aktuálnost změn Ústavy ČR (Brno, Masarykova univerzita, 1999) 67. E Stein, ‘Out of the Ashes of a Federation, Two New Constitutions’, (1997) 45 American Journal of Comparative Law 45. Z Vikarská and M Bobek, ‘Slovakia: Between Euro-Optimism and Euro-Concerns’, in A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (The Hague, T.M.C. Asser Press, 2019) 835.

26 Finland From Constitutional Outsider to Insider: The Changing Relationship between the Finnish Constitution and EU Integration TUOMAS OJANEN

I.  Main Characteristics of the National Constitutional System The current Constitution of Finland (Act No 731/1999)1 entered into force on 1 March 2000, replacing the earlier Constitution Act of 1919 and three other enactments enjoying constitutional status.2 In European comparison, the Constitution of 2000 stands out as a modern and unified constitutional document with a clear structure and is written concisely and lucidly. Indeed, the Constitution of Finland can be regarded as an example of Scandinavian design, where ‘form follows function’ for the purpose of making the Constitution understandable and workable. In total, the Constitution consists of 136 sections arranged in 13 chapters as follows: fundamental provisions (Chapter 1); fundamental rights (Chapter 2); the composition and activities of Parliament (Chapters 3 and 4); the Government and the President of the Republic (Chapter 5); legislation (Chapter 6); state finances (Chapter 7); international relations (Chapter 8); administration of justice (Chapter 9); supervision of legality (Chapter 10); administration and self-government (Chapter 11); national defence (Chapter 12); and final provisions (Chapter 13). The very first chapter, entitled ‘Fundamental Provisions’, defines the foundations of the ­constitutional-political system of Finland as a republican parliamentary democracy based on the rule of law, the principle of parliamentarism, the separation of powers, and the protection of fundamental and human rights. Up until 2012, the Constitution of Finland suffered from ‘European deficit’, as despite its great constitutional significance, European Union (EU) membership was insufficiently reflected in the text of the Constitution. After a recent amendment of the Constitution in 2012, however, the very first provision of the Constitution displays constitutional engagement with EU membership by providing that Finland ‘is a Member State of the European Union’. In addition, constitutional provisions acknowledging the possibility of the transfer of 1 Unofficial translation of the Constitution of Finland, including amendments up to 1112/2011, in English is available at www.finlex.fi/fi/laki/kaannokset/1999/en19990731.pdf. Translations will be used from this source. 2 Accordingly, there were four constitutional enactments enjoying constitutional status: the Constitution Act of Finland, the Parliament Act and two Acts on ministerial liability. All Acts were passed during the first years of Finnish independence (the Acts of 94/1919; 7/1928; 274/1922; and 273/1922).

728  Tuomas Ojanen powers to the EU or international organisations were enacted. The domestic decision-making system pertaining to EU affairs is regulated in more detail in Chapter 8 of the Constitution. One of the most distinctive features of the Constitution, including Finnish legal culture, is what can nowadays be called rights-based constitutionalism.3 The very first provision of the Constitution expresses commitment to rights-based constitutionalism by providing that the Constitution ‘shall guarantee the inviolability of human dignity and the freedom and rights of the individual and promote justice in society’. The same provision adds that Finland participates in international cooperation ‘for the protection of peace and human rights and for the development of society’. Fundamental rights are addressed in Chapter 2 which amounts to a broad catalogue of fundamental rights, with a range of economic, social and cultural rights, in addition to the more traditional civil and political rights. There are also specific provisions on responsibility for the environment and environmental rights, as well as for the right of access to information and the right to good administration. Fundamental rights are almost invariably granted to everyone within Finland’s jurisdiction, save for certain dimensions of the freedom of movement and electoral rights that are reserved to Finnish citizens only. The current significance of rights protection closely intertwines with another contemporary characteristic of the Finnish constitutional system: the existence of a pluralist system of constitutional review combining abstract ex ante review4 by the Constitutional Law Committee of Parliament and ex post review by courts.5 In this model, the ex ante constitutional review by the Constitutional Law Committee is supposed to remain primary, whereas judicial review under section 1066 is designed to address matters missed by the abstract ex ante review, inasmuch as unforeseen constitutional problems would arise in applying the law by the courts in particular cases. Hence, for all practical and theoretical purposes, the practice of the Constitutional Law Committee will be of great significance in this chapter. The constitutional position of the Constitutional Law Committee of Parliament bears many resemblances to centralised judicial review models with constitutional courts at their apex. However, the main difference is that the Constitutional Law Committee is a political organ composed of members of Parliament, albeit with a distinct constitutional mandate for ex ante review of the constitutionality of ‘legislative proposals and other matters brought for its consideration, as well as on their relation to international human rights treaties’ under section 74 of the Constitution. Despite its political organisation, the practice of Committee is characterised by legal argumentation and a search for interpretations that can be linked to the text of the Constitution, 3 See especially Juha Lavapuro, Tuomas Ojanen and Martin Scheinin, ‘Rights-Based Constitutionalism in Finland and the Development of Pluralist Constitutional Review’ (2011) 9 International Journal of Constitutional Law 505. See also Tuomas Ojanen, ‘The Europeanization of Finnish Law’ in P Luif (ed) Österreich, Schweden, Finnland. Zehn Jahre Mitgliedschaft in der Europäischen Union (Vienna, Böhlau, 2007) 145, 156 ff. 4 The supervision by the Constitutional Law Committee is abstract, not concrete, in the sense that the relation between the norm and the circumstances of a particular case is lacking, unlike in the case of concrete ex post (judicial) review where a court reviews the constitutionality of legislation in the light of all relevant circumstances of a concrete case to be decided. 5 See Lavapuro, Ojanen and Scheinin, ‘Rights-Based Constitutionalism’ (n 3). See also Tuomas Ojanen, ‘From Constitutional Periphery toward the Center – Transformations of Judicial Review in Finland’ (2009) 27 Nordisk Tidsskrift for Menneskerettigheter/Nordic Journal of Human Rights 194. 6 Section 106, entitled ‘The Primacy of the Constitution’, provides as follows: ‘If in a matter being tried by a court, the application of an Act of Parliament would be in manifest conflict with the Constitution, the court of law shall give primacy to the provision in the Constitution’. The criterion of a ‘manifest conflict’ is deliberately designed to subordinate judicial review to ex ante review by the Constitutional Law Committee. The travaux préparatoires of section 106 explicitly state that, as a rule, a court should not regard the conflict as manifest, if the Constitutional Law Committee, in its ex ante review, has already reviewed the constitutional issue at hand and held that the relevant Act of Parliament should be regarded as being in harmony with the Constitution. See Government Bill for the new Constitution No 1/1998, 164. See also the Report by the Constitutional Law Committee on Government proposal 1/1998. See also Report 10/1998 of the Constitutional Law Committee, 31.

Finland  729 its preparatory works as well to the Committee’s own previous interpretive practice and to the interpretive practice of international human rights treaty bodies. Before issuing its Opinions or Reports, the Committee regularly hears experts in constitutional law and international human rights law, notably university professors, whose opinions often have significant impact on the Committee’s statements on the constitutionality of legislative proposals and other matters. It deserves emphasis that ex ante review by the Committee of the constitutionality of matters pending before Parliament extends to cover proposals for directives, regulations or other EU measures. This is quite unique in European comparison, as the constitutional review of EU measures usually assumes the nature of ex post review by constitutional courts or other courts in other EU Member States. The possibility of ex ante constitutional review of proposals for EU measures also explains why the Constitutional Law Committee was able to express much earlier such constitutional concerns about eg the proposal for the ESM Treaty than the German Federal Constitutional Court advanced later ex post about the ESM Treaty.7 Finally, a distinct characteristic of the Finnish constitutional-political system has always been made by constitutional arrangements pertaining to the distribution of powers between Parliament, the Government and the President of the Republic. Traditionally, and even on the threshold of EU membership in the early 1990s, Finland still fell into the category of presidential democracies where the elected Head of State – the President of the Republic – enjoyed strong powers distinct from parliamentary decision-making and the requirement of parliamentary confidence. The most significant bastion of presidential power was in the area of foreign policy, but the President also enjoyed strong powers in such domestic affairs as legislation, the formation of the Government and the appointment of state officials. However, the trend has increasingly been to go away from the presidential focus of authority and move towards the Parliament-Government axis since the early 1980s. In practice, the emphasis on the parliamentary aspect of the Finnish constitutional-political system has entailed the strengthening of the Parliament vis-à-vis the Government, on the one hand, and the strengthening of the Government vis-à-vis the President of the Republic, on the other hand. The current Constitution of Finland has carried parliamentary pathos almost to completion by entailing a general restriction of the powers of the President and by affiliating the exercise of the remaining presidential powers to cooperation with the Government. Hence, the functioning of Finland’s constitutional and political system can no longer be defined with reference to the constitutional authority and political power of the President. Instead, it is the parliamentary mode of policymaking that matters predominantly, if not exclusively. The Prime Minister is nowadays the most significant political actor in Finnish everyday politics, and the leadership of the Prime Minister is nowadays explicitly recognised in section 66 of the Constitution.8

II.  Constitutional Culture The Constitution, as a legal and a political instrument, has traditionally been highly esteemed in Finland. The respect for the Constitution originates in the legal-positivist resistance by the 7 See eg Opinions 27/2011 and 1/2012 by the Constitutional Law Committee. The judgment by the German Federal Constitutional Court on the constitutionality of the ESM Treaty was given almost year later, see Judgment BVerfG 12.09.2012, 2 BvR 1390/12 (ESM (Aussetzung der Ratifikation I)) BVerfGE 132, 195 ECLI:DE:BVerfG:2012:rs20120912a. 2bvr139012. 8 For a brief overview of the history of the distribution of powers between state organs, see Ojanen, ‘Europeanization’ (n 3) 161 ff. See also Tapio Raunio, ‘Semi-presidentialism and European integration: lessons from Finland for constitutional design’ (2012) 19 Journal of European Public Policy 567.

730  Tuomas Ojanen Finnish legal and political elite to the campaigns of so-called ‘Russification’ between 1899 and 1905. For over a century, from 1809 to 1917, Finland was an autonomous Grand Duchy within the Russian Empire – Finland had its own legal system, including constitutional enactments inherited from the era of Swedish rule (pre-1809). During the years of ‘Russification’, however, the Finns fought against arbitrary Russian interference with Finland’s domestic legal and political affairs, advancing a constitutional challenge, essentially founded on a simple yet firm claim that all authorities, including those of the Russian Empire, had to strictly observe Finland’s constitutional enactments and Finnish law in general in the exercise of all their powers. As this constitutional challenge proved very successful, a strong tradition of legalism, including respect for the rule of law, started characterising Finnish legal culture from those years onwards.9 The Grand Duchy era also generated another constitutional idiosyncrasy, the institution of exceptive enactments.10 While, in the name of the legal positivist spirit originating in the rule of law, the Finns urged the Russians to abide by constitutional enactments originating in the period of Swedish rule, there were, simultaneously, pressing economic and social reasons to enact modern legislation which was at odds with the antiquated constitutional enactments. As the Finns wanted their constitutional challenge to ‘Russification’ to remain credible, the institution of exceptive enactments offered a way out. In essence, this institution makes it possible to adopt legislation that conflicts with the Constitution without amending the text thereof, subject to the proviso, however, that such legislation is approved in accordance with the procedure for constitutional enactments. As will be discussed in more detail later, the institution of exceptive enactment has also been used for the purpose of bringing into force obligations originating from EU membership that have been deemed to be in conflict with the Constitution. As with other Nordic countries, rights and judiciaries traditionally assumed marginal legal roles in the Finnish scene of constitutionalism until the late 1980s.11 The Finnish constitutional system followed both formally and practically the classic legislative supremacy principles with ideas about democracy as majority rule and about the law as a supreme expression of the people’s will at their apex. However, the years since the late 1980s have brought a marked change. As already noted, Finnish constitutionalism witnessed a shift from the legislative sovereignty paradigm to one in which legislative acts are increasingly subordinated to a rights-based system of pluralist review where both the democratically elected legislature through ex ante review by the Constitutional Law Committee of Parliament and the independent judiciary ex post review are entrusted with a duty to protect fundamental and human rights.12 Finally, it is important to note on the Finnish constitutional and political culture that Finland has historically been fairly homogeneous and state-centred in its understanding of community values. Aside from the civil war of 1918, there has been no significant ethnic, cultural, political or religious controversies that would divide society. Moreover, ‘consensual pathos’ has characterised Finnish political and constitutional culture in recent decades, thereby also contributing to 9 See in more detail Ojanen, ‘Europeanization’ (n 3) 146 ff. See about Finnish understanding on the Constitution and the legacy of the autonomy under the Russian Empire, also Anu Mutanen, ‘Towards a Pluralistic Constitutional Understanding of State Sovereignty in the European Union? The Concept, Regulation and Constitutional Practice of Sovereignty in Finland and Certain Other EU Member States’, Dissertation, University of Helsinki, 2015, 272–76 with references. 10 A brief historical review of this institution in English is provided in Martin Scheinin, ‘Constitutional Law and Human Rights’ in J Pöyhönen (ed), An Introduction to Finnish Law (Helsinki, Kauppakaari, 2002) 31, 55 f. 11 For the traditional ‘Nordic constitutional approach’ to rights and courts, see Ran Hirschl, ‘The Nordic counternarrative: Democracy, human development and judicial review’ (2011) 9 International Journal of Constitutional Law 449; and Eivind Smith, ‘Judicial Review of Legislation’, in H Krunke and B Thorarensen (eds), The Nordic Constitutions: A Comparative and Contextual Study (Oxford, Hart Publishing, 2018) 107. 12 For the role of rights and courts in the Finnish scene of constitutionalism, see Ojanen, ‘Periphery to Center’ (n 5) and Lavapuro, Ojanen and Scheinin, ‘Rights-Based Constitutionalism’ (n 3).

Finland  731 ‘consensual constitutional reforms’. For instance, Parliament adopted the Constitution of Finland of 2000 practically unanimously. However, the most recent constitutional amendment of 2012 that will be discussed in more detail below in section III.C proved an exception, as no less than 40 MPs voted against the amendment and another 40 were absent. These numbers are high, as the unicameral Parliament only consists of 200 MPs. Such wide resistance to the amendment primarily stemmed from those parts of the amendment that addressed the ‘European deficit’ of the Constitution. In recent years, issues revolving around European integration, in general, and the eurozone crisis, in particular, moved to the centre of the Finnish political arena in a manner that has been increasingly causing friction between political parties and different groups of society. In particular, recent years have witnessed a breakthrough of right-wing populism which can be characterised by such attributes as ‘anti-European’, ‘anti-immigration’ and ‘nationalistic’. These kinds of ‘anti-EU integration’ political trends are not constitutionally insignificant, either. This is especially so because the Constitution of Finland nowadays reflects a relatively strong commitment to EU membership, as will be discussed in more detail later.13

III.  Constitutional Foundations of EU-Membership A.  Constitutional Arrangements Regarding the Accession Treaty Finland joined the EU on 1 January 1995, along with Austria and Sweden. Finland has also been a eurozone member since 1 January 1999. In the mid-1990s, the Finnish Constitution was still introverted and nationalist.14 The sovereignty of Finland was understood in a very formal and rigid manner so that the transfer of powers to international organisations was almost ‘automatically’ found to be in conflict with the Constitution. In essence, the possibility of adhering to a very rigid and strict interpretation of sovereignty derived from the institution of ‘exceptive enactments’, as at the time it was allowed to approve the incorporation enactments of treaties conflicting with the Constitution through a decision of Parliament by a qualified majority of two-thirds without formally amending the Constitution. The institution of exceptive enactments was also applied to the bringing into force of the Treaty of Accession of 1994.15 It was deemed to be in conflict with the Constitution in several ways, the major reason simply being the incompatibility of the transfer of powers to the EU with the sovereignty of Finland. Accordingly, the Treaty of Accession was incorporated into Finnish law through an exceptive enactment (Act No 1540 of 1994), which was approved by a two-thirds majority in Parliament. In addition, Parliament consented to the ratification of the Accession Treaty by a simple majority decision. The domestic ratification and incorporation of the Accession Treaty was accompanied by a consultative referendum on 16 October 1994. The referendum was not a constitutional condition for accession, but aimed at enhancing the domestic democratic 13 See Janne Salminen, ‘Manifestations of the European Union Membership in the Constitution of Finland in the European Context’ (2010) 12 Europarättslig tidskrift 509. 14 Tuomas Ojanen, ’Constitutional amendment in Finland’, in X Contiades (ed), Engineering Constitutional Change. A Comparative Perspective on Europe, Canada and the USA (Abingdon, Routledge, 2013) 93, 110. 15 The instrument concerning the accession of Finland to the European Union is the Treaty between Member States of the European Union and the Kingdom of Norway, the Republic of Austria, the Republic of Finland, the Kingdom of Sweden, [1994] OJ C 241/14, as adjusted by Council Decision 95/1/EC, Euratom, ECSC, [1995] OJ L 1/1.

732  Tuomas Ojanen legitimacy of EU membership. In the referendum, a majority of 56.9 per cent of those who voted answered ‘yes’ to the following question: ‘Should Finland become a member of the European Union in accordance with the treaty which has been negotiated?’ The turnout was 74 per cent. However, a constitutional amendment was found necessary insofar as the domestic distribution of powers between the Government and the President, including the role of Parliament in EU affairs, was concerned.16 Actually, the accession of Finland to the European Economic Area (EEA) in 1994 had already made it necessary to reconsider the domestic distribution of powers between state organs. Although the powers of the President had already been subject to significant reductions in domestic affairs, the President still enjoyed strong powers in the sphere of foreign affairs in the early 1990s. Hence, one of the most important issues to be decided in Finland prior to embarking on the process of European integration was whether European affairs – first EEA affairs, and later EU affairs – should be considered a domestic or a foreign policy matter. In the latter case, they would have fallen within the competence of the President by virtue of section 33 of the Constitution Act of 1919, with the outcome that the constitutional pendulum would have lurched back again towards a strong presidency, thereby watering down constitutional amendments since the early 1980s to nudge the Finnish constitutional system towards parliamentarism. In particular, this outcome would have been a severe blow to the effective participation of Parliament in domestic decision-making pertaining to European affairs. However, the competences were arranged according to the same ratio as those used in domestic legislative affairs, instead of the competence arrangements typical of foreign policy. Accordingly, the Constitution was amended so that the main responsibility for the national preparation of EEA affairs and later EU affairs was given to the government whose members are both individually and collectively accountable to Parliament. Moreover, specific constitutional provisions were enacted for the purpose of ensuring the effective participation of Parliament in the national preparation of EU affairs (section 96), on the one hand, and Parliament’s right to receive information on all measures on which decisions are to be made in the EU whenever these fall within Parliament’s competence, on the other hand (section 97).17

B.  The New Constitution of 2000 The entry into force of the Constitution of Finland on 1 March 2000 modified the Finnish constitutional foundations of EU membership by introducing the so-called ‘internationalisation principle’, according to which ‘Finland participates in international cooperation for the protection of peace and human rights and for the development of society’ (section 1, subsection 3). According to the travaux preparatoires of the Constitution, this new clause was intended to reflect on the positive attitude of the Constitution towards international cooperation, including European integration, as well as to inform that the sovereignty clause of the Constitution must, under the current circumstances, be understood with relation to international obligations binding on Finland and, particularly, EU membership. 16 See also Tuomas Ojanen, ‘The Impact of EU Membership in Finnish Constitutional Law’ (2004) 10 European Public Law 531, 536 ff and 551–54. See generally, Janne Salminen, ‘Yhä läheisempään liittoon? Tutkielmia valtiosäännön integraationormin sisällöstä ja vaikutuksista’ [‘Towards an Ever Closer Union? Studies in the Constitutional Consequences of Integration’], Dissertation, Turku University 2015, discussing the changes as transformations of the integration norms of the Finnish Constitution. See also Mutanen, Constitutional Understanding (n 9), discussing from the comparative perspective the changes of the concept of sovereignty in Finland and providing a detailed account of the amendment process, drafting history and relevant literature in English. 17 See also Ojanen, ‘Impact’ (n 16) 554–57, and Veli-Pekka Viljanen, ‘Eurooppalaistuminen valtioelinten välisissä suhteissa’ [‘Europeanisation in International Relations’] (2003) 100 Lakimies 1169.

Finland  733 Since 2000 onwards, the Constitutional Law Committee of Parliament has consistently referred to this new constitutional provision in reviewing both EU measures and international treaties for their compatibility with the sovereignty of Finland. Moreover, the Committee has invariably regarded the EU as being sui generis and, accordingly, something quite different from international and regional organisations.18 In practice, this idea of the sui generis nature of the EU has resulted in a greater tolerance of limitations on sovereignty stemming from EU membership than those originating in international obligations. In addition, EU membership started to increasingly shape the interpretation of other constitutional provisions beyond the sovereignty clause of the Constitution.19 However, the Constitution still suffered from the ‘European deficit’ as its text failed to display appropriately the constitutional significance of EU membership. As the constitutional foundations of EU membership were premised on the basis of the institution of exceptive enactments, membership still appeared as an outsider to the constitutional system of Finland.

C.  The Constitutional Amendment of 2012 The latest amendment of the Constitution entered into force in 2012 with provisions explicitly designed to address the ‘European deficit’ of the Constitution. To begin with, section 1, subsection 3, of the Constitution, located in Chapter 1 entitled ‘Fundamental provisions of the Constitution’, was amended to include an explicit commitment to EU membership by providing simply, yet forcefully, that ‘Finland is a Member State of the European Union’. Moreover, new provisions on the transfer of powers were inserted in section 94, entitled ‘Acceptance of international obligations and their denouncement’ and section 95, entitled ‘Bringing into force of international obligations’. In essence, these new provisions provide that a ‘significant’ transfer of state powers to the EU or international organisation or an international body requires a decision taken with at least two-thirds majority in Parliament. By contrast, the transfer of powers that cannot be deemed to be ‘of significance with regard to Finland’s sovereignty’ can be approved by a decision taken by a simple majority in Parliament. As a result, there is no longer any meaningful scope of application for the institution of exceptive enactment as regards the transfer of powers to the EU. Instead, the crucial constitutional question simply is whether a given transfer of powers can be regarded as being ‘of significance’ within the meaning of the Constitution. It also deserves to be emphasised that the text of the Constitution remains silent on powers that cannot be transferred to the EU or international organisations. The importance of this constitutional amendment cannot be over-emphasised insofar as the constitutional foundations of EU membership are concerned. Whereas the application of exceptive enactments originally entailed the exclusion of EU membership from the Finnish constitutional system, including by considering the EU as something contradicting the Constitution, the Constitution of Finland nowadays displays a constitutional commitment to EU membership among the foundations of the Finnish constitutional system. As a result, EU membership has gone from being an outsider to the Constitution of Finland to being an insider, with a firm place among the foundations of the Constitution. 18 See eg Opinions 13/2009, 36/2006, 9/2006, 38/2001 of the Constitutional Law Committee. 19 For the evolution of the sovereignty doctrine and the constitutional effects of EU membership in general, see especially Mutanen, Constitutional Understanding (n 9). See also Salminen, ‘Manifestations’ (n 13) and T Ojanen, ‘The European Constitution in the Far North, in a Country Called Suomi’, in Anneli Albi, Jacques Ziller (eds), The European Constitution and national constitutions: ratification and beyond (Kluwer, 2007) 89–100.

734  Tuomas Ojanen Aside from the foregoing changes, the constitutional amendment of 2012 enhanced the identity of the individuals as ‘EU citizens’ by amending the provision on electoral and participatory rights so that ‘every Finnish citizen and every other citizen of the European Union resident in Finland, having attained eighteen years of age, has the right to vote in the European Parliamentary elections, as provided by an Act’. This amendment reflects the thinking by the Constitutional Law Committee about the Union as a community of not only the Member States but also the citizens.20 In addition, the constitutional amendment of 2012 introduced a new form of direct participation at the state level: citizens’ initiative, which offers citizens a possibility to have their legislative initiative considered by Parliament. According to section 53, subsection 3 of the Constitution, at least 50,000 Finnish citizens entitled to vote have the right to submit an initiative for the enactment of an Act to the Parliament. The Act on citizens’ initiative (Act No 12 of 2012) includes more specific provisions on the procedure to be followed in the making of a citizens’ initiative. Parliament has an obligation to consider a citizens’ initiative, but it is up to Parliament whether it wishes to approve an initiative, with or without changes. If an initiative is rejected by Parliament, a new initiative concerning the same matter can be set in motion. Citizens’ initiative has rapidly become a popular tool of participatory democracy in Finland, and nearly 20 initiatives have already reached the status of citizens’ initiative, with the required amount of endorsements, and have been submitted for Parliament’s consideration. Hitherto, however, only one initiative – the Equal Marriage Law – has resulted in legislation, with only minor changes.21

D.  European Integration Outside the EU Legal Order As a matter of Finnish constitutional law, EU matters are construed in a wide sense so as to secure the effective participation of Finland’s Parliament in the national preparation of European affairs. Hence, the constitutional framework pertaining to the EU also extends to cover wider European integration which formally falls outside the institutional framework of the EU, including the EU legal order, but which nonetheless has sufficiently close linkage to the EU. For instance, the constitutional basis of the eurozone regime has in recent years included the use of international agreements such as the Treaty Establishing the European Stability Mechanism (ESM). However, although the ESM Treaty is formally a treaty that was drafted and adopted outside the institutional and constitutional framework of the EU, the Constitutional Law Committee of Parliament nonetheless regarded it as falling within the scope of application of sections 96 and 97 of the Constitution. The ESM Treaty and other European measures provoked by the eurozone crisis created entirely new constitutional dynamics and debates on European integration in Finland. It is to be emphasised that ex ante constitutional review by the Constitutional Law Committee of Parliament of legislative proposals and other matters pending before Parliament also extends to proposals for various European measures with sufficient ‘EU linkage’. Hence, Finland has often been in recent years the very first Member State to express constitutional doubts about various proposals for European measures to combat the eurozone crisis and to develop the EMU in general. The ESM Treaty provides a neat illustration of ex ante review by the Constitutional Law Committee of constitutional implications of European measures pertaining to the eurozone crisis. First of all, the central constitutional premise of the Constitutional Law Committee of 20 See eg Opinions 36/2007 of the Constitutional Law Committee. 21 For more information on citizens’ initiative in Finland, see eg Parliament’s website, www.eduskunta.fi/EN/lakiensaataminen/kansalaisaloite/Pages/default.aspx.

Finland  735 Parliament has been the fiscal sovereignty of the EU Member States, including Finland, for the sake of securing democratic legitimacy at both the European and the domestic level. On a more concrete level, the Committee’s ex ante constitutional review has revolved around national budgetary sovereignty and the appropriate procedure for handling international agreements that affect the institutional framework of the EU and its legislative procedures. After all, as drafting, adopting, ratifying and incorporating international agreements involve only the contracting parties, it effectively disregards those procedures and other requirements of EU legislative processes that engage the major EU institutions, including the European Parliament and national Parliaments, in legislative and decision-making processes within the EU. During negotiations of the draft ESM Treaty, the Constitutional Law Committee of Parliament already addressed, in its ex ante review, the effects of the draft Treaty on national budgetary sovereignty and the participation of Parliament in the national preparation of European affairs. The Constitutional Law Committee took the view that the draft ESM Treaty impinged upon the legislative and budgetary powers of Parliament in a manner that would warrant both Parliament’s acceptance of the ESM Treaty and an Act of Parliament to bring it into force domestically in accordance with sections 94 and 95 of the Constitution. Moreover, the Committee required the adoption of unanimous decision-making in the stability mechanisms as a precondition for the compatibility of the treaty with the sovereignty of Finland. In particular, the so-called emergency decision-making procedure and the scope of the decision-making power by a qualified majority in the ESM Board of Governors provoked a constitutional concern. The Committee considered that, in light of the draft Treaty, Finland’s financial liabilities could be increased by a qualified majority decision of the Board of Governors above the maximum limit defined in the Treaty. The risk of increasing Finland’s financial liabilities against its will was considered by the Constitutional Law Committee to be incompatible with Finland’s sovereignty in general and the budgetary competence of the Finnish Parliament in particular. Accordingly, the Committee demanded the Government to make changes to the draft ESM Treaty for the purpose of eliminating constitutional problems.22 Later, these constitutional demands were effectively taken into account during the further negotiations of the ESM Treaty at the European level so that the draft ESM Treaty was altered to include the unanimity requirement insofar as decision-making on financial liabilities was concerned.23 Furthermore, the Constitutional Law Committee reviewed the ESM Treaty as regards its implications on the amount of Finnish capital investment in the ESM and the risks related to it against the so-called Constitution-based obligations of the state.24 The Committee required all financial liabilities and investments in the various parallel mechanisms to be calculated in order to get a sufficiently clear understanding of Finland’s liabilities as a whole. In order to be able to review appropriately the compatibility of the ESM Treaty with the Constitution, the Committee also took a holistic constitutional notice of the total amount of public debt and the risks of the investment. Given the total amounts of the Finnish liabilities under the ESM Treaty – the paidin capital (1.4 billion EUR) and on-demand callable capital (11 billion EUR) –, the Committee concluded that the total amount was, as such, very considerable in comparison with eg the total amount of the annual state budget (ca 54 billion EUR). Yet, as the paid capital was under 1 per cent 22 Opinions 25/2011, 22/2011, 1/2011 and 13/2012 of the Constitutional Law Committee. 23 Originally, the (draft) Fiscal Compact was intended to impose an obligation on Member States to amend their national constitutions so as to introduce a debt break. Largely due to constitutional criticism by some Member States, including Finland, no legal obligation to amend domestic constitutions found its way to the final form of Fiscal Compact. See Opinion 24/2011 of the Constitutional Law Committee. 24 See Opinions 13/2012 and concerning the European Financial Security Facility and the Framework Agreement Opinion 5/2011 of the Constitutional Law Committee.

736  Tuomas Ojanen of the GDP, and the maximum about 5 per cent of the GDP, the Committee concluded that the ESM Treaty did not so significantly curtail national budget autonomy as to be in conflict with the Constitution.25 Finally, when reviewing constitutional implications of the ESM Treaty, the Constitutional Law Committee attached great significance to the ability of Finnish Parliament to genuinely control and influence the Finnish member of the ESM Board of Governors. As sections 96 and 97 of the Constitution function to secure real and extensive information and participation rights to Parliament in the national preparation of EU affairs, the constitutional requirement regarding the ESM Treaty has been that the Government must inform Parliament well prior to decisionmaking within the ESM on eg the financial implications of the granting of significant loans in order to safeguard the prerogatives of Parliament appropriately. Where necessary, the position of Parliament should direct the decision-making choices of the Finnish representative within the ESM Board of Governors.26

IV.  Constitutional Limits to EU Integration A.  From Absolute Sovereignty Doctrine Towards a More Modern and Nuanced Approach Initially, the traditional sovereignty doctrine featured as the only Finnish constitutional limit to EU integration. Prior to and even during the initial years of Finland’s EU membership, the constitutional construction of sovereignty was markedly formal and rigid, with a focus on the opening provision of the Constitution Act of 1919, laconically providing that ‘Finland is a sovereign republic’ (section 1, subsection 1 of the Constitution Act of 1919). As the Constitution Act of 1919 lacked a specific constitutional provision for the transfer of powers to international institutions, not to speak of the EU in particular, constitutional review by the Constitutional Law Committee could not take advantage of such constitutional considerations that directed and shaped constitutional review of EU affairs in many other EU Member States. The interpretation of sovereignty was markedly strict in the sense that an issue of unconstitutionality was found almost automatically if an international treaty appeared to entail even a slight transfer of powers to international organisations or the authorities of other states. The explanation was the institution of exceptive enactments, allowing the possibility of enacting an exceptive enactment through a qualified majority of two-thirds, without formally amending the Constitution. Hence, there were actually no genuine material constitutional limits for EU integration. Instead, the constitutional limits remained formal and procedural in character as the constitutional attention was largely, if not exclusively, given to the proper parliamentary procedure. As already discussed, the Treaty of Accession of 1994 was incorporated into Finnish law through an exceptive enactment (Act No 1540 of 1994), which was approved by a two-thirds majority in Parliament. In addition, Parliament consented to the ratification of the Treaty of Accession by a simple majority decision. The same parliamentary procedures were applied to the Treaty of Amsterdam that revised the founding treaties of the EC and the EU in 1999. Regarding the Amsterdam Treaty, the essential constitutional question was whether this Treaty entailed such

25 Opinion 26 Opinion

of 13/2012 of the Constitutional Law Committee. 13/2012 of the Constitutional Law Committee.

Finland  737 qualitatively new derogations from the Constitution that transcended those made by the exceptive enactment of the Accession Treaty. The Constitutional Law Committee answered affirmative and, accordingly, an exceptive enactment was found necessary for the purpose of bringing into force the Amsterdam Treaty. The entry into force of the Constitution of Finland brought about a new understanding of the sovereignty of Finland, although the textual formulation of the sovereignty clause in the Constitution of 2000 is exactly the same as in the old Act of Constitution of 1919. Two distinct, yet interrelated reasons coalesce to explain the shift in the construction of sovereignty. On the one hand, the Constitution of 2000 places an entirely new emphasis on international cooperation, by containing a general statement on the positive attitude towards it as follows: ‘Finland participates in international cooperation for the protection of peace and human rights and for the development of society’ (section 1, subsection 3). According to the travaux préparatoires of the Constitution, this provision is designed to offer a new dimension for the construction of sovereignty so that international obligations, which are deemed to be ‘conventional’ in modern international co-operation and which only affect sovereignty in a ‘minor way’, are no longer as at variance with the sovereignty of Finland. Since 2000, the Constitutional Law Committee has consistently referred to this new constitutional provision in reviewing both EU measures and treaties for their compatibility with the sovereignty of Finland. On the other hand, the traditional and markedly formal and rigid understanding of sovereignty was replaced by the idea of the specificity of the EU among international organisations. In essence, the Union is perceived as being sui generis, as something much more than other international organisations yet something less than a (federal) state. In essence, the distinction between the EU and international organisations resulted in a greater constitutional tolerance of limitations on sovereignty stemming from EU membership than those originating in international obligations. However, this new construction of sovereignty does not imply that sovereignty has become totally irrelevant as a limit to EU integration. The overall approach of the Constitutional Law Committee is to accommodate the EU membership within the principle of sovereignty. Attention also deserves to be focused on the Committee’s fundamental conception of the EU. The Committee envisions the EU as an entity derived from, and legitimised by, the sovereign will of the Member States. In terms of traditional constitutional terminology, the Committee still regards KompetenzKompetenz as remaining in the hands of Finland and the other EU Member States. Moreover, the Committee is in the habit of thinking that Finland’s EU membership is about exercising Finland’s sovereignty in cooperation with other EU Member States for the benefit of European integration. These views on the nature of EU integration were particularly highlighted regarding the Treaty establishing a Constitution for Europe. In essence, Finland’s position was that the EU must continue to develop both as a close community of independent Member States and a community of citizens. The competence of the Union must also remain limited in the sense that the EU has competence only in the areas explicitly enumerated in the Treaty, while all other competences should remain with the Member States. Since the Member States should retain the power to set the limits of the Union’s competence, Kompetenz-Kompetenz also remains at Member State level.27 Aside from changing the constitutional setting relating to the interpretation of the sovereignty clause, the entry into force of the Constitution of Finland of 2000 introduced a distinct limit to European integration and international cooperation in general. According to section 94, subsection 3, ‘[a]n international obligation shall not endanger the democratic foundations of the

27 See

eg Opinion 36/2006 of the Constitutional Law Committee. See also Ojanen, ‘A Country Called Suomi’ (n 19).

738  Tuomas Ojanen Constitution’. Up until now, this constitutional limit has been addressed by the Constitutional Law Committee in its Opinions on the Treaty establishing a Constitution for Europe and the Lisbon Treaty, albeit in a very laconic and bland manner as the Committee simply contented itself with noting that these treaties appeared unproblematic in light of section 94, subsection 3, of the Constitution.28 The onset of the economic crisis in the autumn of 2008 and, especially, the eurozone crisis has brought about some new constitutional limits to EU integration insofar as closer EMU integration is concerned. On the one hand, the Constitutional Law Committee scrutinised various instruments relating to the eurozone crisis for their impact on the budgetary powers of Parliament. In particular, the Committee was concerned that the absolute amount of Finland’s liabilities under the ESM Treaty or other instruments pertaining to the eurozone crisis would not endanger, in light of the annual national budget, the ability of Finland to meet its obligations under the Constitution. Although the Committee has so far invariably concluded that Finland’s liabilities have neither conflicted with the budgetary powers of Parliament nor endangered the ability of Finland to observe its constitutional obligations, these considerations have nonetheless brought about new aspects to the ex ante review of EU measures for their compatibility with the Constitution by the Constitutional Law Committee.29 On the other hand, the Constitutional Law Committee has consistently emphasised in the context of various measures pertaining to the eurozone crisis the need to observe the strong constitutional prerogatives of Parliament as regards its rights of information and participation in domestic decision-making pertaining to EU affairs.30 As already noted, the 2012 constitutional amendment introduced entirely new provisions on the transfer of powers to the EU or international organisations that require a two-thirds majority in Parliament for ‘significant’ transfer of powers. E contrario, ‘insignificant’ transfer of powers can be decided by simple majority in Parliament. However, it warrants emphasis that the Constitution is silent on such powers that cannot be transferred, save for the condition under section 94, subsection 3 of the Constitution that ‘an international obligation shall not endanger the democratic foundations of the Constitution’. Moreover, as constitutional provisions on the transfer of powers are still very novel and the interpretive practice of the Constitutional Law Committee is yet minuscule, the full significance of these new constitutional provisions in the context of EU membership is still to emerge. Finally, it is worthy of mention that the Constitutional Law Committee has been relatively cautious in contesting EU legislation or their implementing enactments on ultra vires grounds. While the Constitutional Law Committee has emphasised as a starting point that proposals for EU legislation should stay within the limits of the competences of the EU as defined in the EU Treaties,31 there is so far no Opinion by the Committee or by a Finnish court for that matter 28 The Constitution of 2000 also introduced a limit to the use of the exceptive enactment by requiring that the derogation made by an exceptive enactment of the Constitution has to remain ‘limited’ (section 73 of the Constitution). Both the incorporation enactment of the Constitutional Treaty and the incorporation enactment of the Lisbon Treaty were reviewed in light of this requirement by the Constitutional Law Committee with the outcome that these treaties were found to entail a limited derogation of the Constitution within the meaning of section 73 of the Constitution by the Committee. After the entry into force of the recent constitutional amendment of 2012, introducing provisions on the transfer of powers, the institution of exceptive enactments, including the requirement of a limited derogation, is no longer applicable. See Opinions 36/2006 and 13/2009 of the Constitutional Law Committee. 29 See Opinion 13/2012 of the Constitutional Law Committee. For an overview of the Committee’s practice regarding various measures pertaining to the euro crisis, see Päivi Leino and Janne Salminen, ‘The Euro Crisis and Its Constitutional Consequences for Finland: Is There Room for National Politics in EU Decision-Making?’ (2013) 9 European Constitutional Law Review 451. 30 See eg Opinion 13/2012 the Constitutional Law Committee. 31 See eg Opinions of 36/2006 and 13/2009 by the Constitutional Law Committee.

Finland  739 proclaiming an EU measure as ultra vires. Besides, both the Constitutional Law Committee and Finnish courts have accepted that the validity of EU legislation is a matter of EU law in which the Court of Justice of the European Union (CJEU) has the final authority. However, the Constitutional Law Committee has sometimes entertained doubts as regards the appropriate legal basis of proposals for EU measures and, accordingly, has required the Government to pay attention to this issue in the preparation of the legislative proposals at the EU level.32 In conclusion, while a very formal and rigid sovereignty doctrine dominated Finnish constitutional practice and doctrine at the beginning of its EU membership, this doctrine has gradually been replaced by a much more modern approach. Today, the Constitution does not only reflect the sovereignty of Finland as founded on the principle of international co-operation in which EU membership plays a special role. By explicitly mentioning the membership of Finland in the EU in the very first provision of Chapter 1, entitled ‘Fundamental provisions of the Constitution’, the Constitution also displays commitment to EU membership by positioning this membership among central arrangements of the Finnish constitutional system. Moreover, there is no longer any room for the applicability of the traditional Finnish constitutional peculiarity, the institution of exceptive enactments, which initially presented EU membership as ‘constitutional outsider’ derogating from the Constitution. Instead, constitutional attention is now geared towards new provisions on the transfer of powers so that the key question is the level of significance of such transfer.

B.  The Domestic System of Rights Protection and Constitutional Identity as Limits to Domestic Reception of EU Measures On the verge of EU membership, a strong distinction between the domestic system for the protection of fundamental and human rights and EU legal order was made in Finland. These two were seen as completely unconnected with distinct fields of application. Soon, however, it was realised that EU membership inevitably affects the domestic system for the protection of fundamental and human rights. Some provisions of the 1995 Constitutional Rights Reform that entered into force on 1 August 1995, ie seven months after Finland’s accession to the EU, can already be understood as amounting to a kind of constitutional self-defence against those problematic tendencies that EU membership might entail in the long run in the field of fundamental and human rights. A reference can be made to a strict clause on the right of access to information (section 12, subsection 2), a relatively far-reaching clause in section 19 on the right to social security in which the guaranteed rights are granted to everyone (and not just EU citizens or workers and their families). Some social rights, either by means of the relevant constitutional provision itself or by means of Acts of Parliament, have also been guaranteed as subjective rights directly enforceable through the courts. Section 20, which guarantees responsibility for the environment and environmental rights, can also be understood as imposing restrictions on integration, and the same applies to section 21 on the right to good governance. Moreover, the Constitutional Law Committee of Parliament adopted the position that it is not permitted for domestic implementation of EU measures to lower the domestic level of protection of fundamental and human rights.33 This position was not simply paying lip-service to rights.

32 See 33 See

eg Opinion 15/2011 of the Constitutional Law Committee. especially Opinion 25/2001 of the Constitutional Law Committee.

740  Tuomas Ojanen The implementation of both the Council Framework Decision of 13 June 2002 on Combating Terrorism34 and the Council Framework Decision on the European Arrest Warrant35 illustrate how the observance of fundamental and human rights has come to take precedence over the ‘maximal’ implementation of EU measures in certain concrete circumstances.36 However, it should be emphasised that this kind of ‘rights-based constitutional resistance’ to the domestic implementation of EU legislation remained very rare. First of all, rather than being harmful to rights protection, EU membership has contributed the protection of fundamental and human rights in such areas as eg the principle of non-discrimination, environmental rights and participatory rights. Moreover, it has rather infrequently been possible to implement EU instruments without having to compromise the observance of fundamental and human rights. The practice of the Constitutional Law Committee also suggests a tendency to temper any open conflict between domestic fundamental rights and EU law in a variety of ways, often facilitated by the imprecise and open-ended wording of EU measures, as well as their status as minimum standard measures.37 Finally, a notable feature of the practice of the Constitutional Law Committee is made by its systematic references to the EU Charter of Fundamental Rights when reviewing the compatibility of proposals for EU measures or domestic implementing enactments with the Constitution or human rights treaties binding upon Finland. The very first reference by the Committee to the Charter can be found in its very first Opinion of the parliamentary session of 2001,38 and from that Opinion onwards the Committee has consistently referred to the applicable rights under the EU Charter of Fundamental Fights alongside relevant constitutional rights and international human rights treaties.39 Recently, the Constitutional Law Committee invoked, among others, the notion of constitutional identity in the context of a domestic bill on supplementary domestic legislation pertaining to the EU General Data Protection Regulation.40 It took the view that oversight of

34 2002/475/JHA: Council Framework Decision of 13 June 2002 on combating terrorism, [2002] OJ L 164/3. 35 2002/584/JHA: Council Framework Decision of 13 June 2002 on the European arrest warrant and the surrender procedures between Member States, [2002] OJ L 190/1. 36 See especially Ojanen, ‘Periphery to Center’ (n 5). 37 For an overview, see T Ojanen, ‘The Europeanization of Finnish Law’ in P Luif (ed) Österreich, Schweden, Finnland. Zehn Jahre Mitgliedschaft in der Europäischen Union (Vienna, Böhlau, 2007) 145. 38 Opinion 1/2001 of the Constitutional Law Committee. 39 See eg Opinion 18/2012 by the Constitutional Law Committee on the bill amending the Act on the Processing of Personal Data by the Police and the Act on the Processing of Personal Data by the Border Guard. The Committee explicitly referred to Art 8 of the EUCFR, Art 8 of the ECHR, including the judgment by the ECtHR in the case of Rotaru v Romania (ECtHR 04.05.2000 Application No 28341/95 ECLI:CE:ECHR:2000:0504JUD002834195), as well as the Council of Europe’s Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data, as premises of constitutional review. Similarly, Art 8 of the EUCFR, Art 8 of the ECHR, and the Council of Europe’s Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data were mentioned in Opinion 14/2009 by the Committee which also discussed the judgment of the ECtHR in the case of S and Marper v United Kingdom at length (ECtHR (GC) 04.12.2008, Applications Nos 30562/04 and 30566/04 ECLI:CE:ECHR:2008:1204JUD003056204). See also Opinion 11/2008 in which the Committee took constitutional notice of Opinion of the European Data Protection Supervisor on the Proposal for a Council Framework (PNR) data for law enforcement purposes, as well as Art 8 of the EUCFR, Art 8 of the ECHR and the Council of Europe’s Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data when assessing the position adopted by the Finnish Government on the proposal for Council’s framework decision on the use of PNR data for law enforcement purposes. Directive 2006/24/EC of the European Parliament and of the Council of 15 March 2006 on the retention of data generated or processed in connection with the provision of publicly available electronic communications services or of public communications networks and amending Directive 2002/58/EC, [2006] OJ L 105/54. 40 Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) [2016] OJ L 119/1.

Finland  741 the application of the Regulation by a supervisory authority (see Chapter VI of the Regulation) cannot cover the functioning of the Parliamentary Ombudsman and the Chancellor of Justice of the Government. This was because these institutions have constitutional mandates to ensure that all public authorities and officials, including the new domestic supervisory authority, observe the law and fulfil their duties. In its reasoning, the Constitutional Committee made a reference to Article 4(2) of the Treaty on the European Union (TEU), according to which the Union must respect national identities, inherent in their fundamental structures, political and constitutional, including regional and local self-government. In addition, the Committee referred to CJEU’s judgments in the Digibet and Albers41 and Sayn-Wittgenstein42 cases where the CJEU, among others, took judicial notice of Article 4(2) TEU. After a quite lengthy and complex reasoning, the Constitutional Law Committee concluded that the issue at stake was not a conflict between EU law and the Constitution but, instead, it was the extension of the oversight by the supervisory authority to the Ombudsman of Parliament and the Chancellor of Justice which would entail a serious infringement of the fundamental institutional arrangements under the Constitution of Finland. According to the Committee, the EU General Data Protection Regulation does not require such an outcome. It remains to be seen whether this is so. Nonetheless, this Opinion by the Constitutional Law Committee is most recent illustration how the Constitution of Finland can direct and define the domestic implementation of EU law.43

V.  Constitutional Rules and/or Practice in Implementing EU Law A.  Domestic Decision-making System in the National Preparation of European Union Affairs The domestic distribution of powers regarding EU affairs revolves around the GovernmentParliament axis. The general competence in EU matters belongs to the Government, whose members are individually and collectively accountable to Parliament. It deserves emphasis that nothing has been excluded from the Government’s competence and, accordingly, the Government is competent also in matters falling within EU’s Common Foreign and Security Policy. The competence of the Government is regulated in section 93, subsection 2, of the Constitution as follows: The Government is responsible for the national preparation of the decisions to be made in the European Union, and decides on the concomitant Finnish measures, unless the decision requires the approval of the Parliament. The Parliament participates in the national preparation of decisions to be made in the European Union, as provided in this Constitution.

Within the Government, the daily responsibility for the preparation, monitoring and determination of Finland’s positions in EU affairs rests with the relevant ministries. In addition, a specific system of coordination exists for the purpose of securing coherence of the domestic preparation of EU affairs within the government, including the ability of Finland to present a sufficiently clear and concise position in line with Finland’s general EU policy on issues under consideration in the EU. The coordination system involves competent ministries, the Cabinet Committee on European Union Affairs, the Committee for EU affairs and its EU sub-committees.

41 CJEU 42 CJEU

12.06.2014 C-156/13 (Digibet and Albers) ECLI:EU:C:2014:1756, para 32. 22.12.2010 C-208/09 (Sayn-Wittgenstein) ECLI:EU:C:2010:806. 14/2018 by the Constitutional Law Committee of Parliament.

43 Opinion

742  Tuomas Ojanen The Government Secretariat for EU affairs serves as the secretariat for the Cabinet Committee on European Union Affairs and the Committee for EU affairs. The ratio behind the structure and functioning of the coordination system is that the greater the political, economic or legal significance of the EU affair in question is, the higher the level handling that affair. This domestic decision-making system is supplemented by the Finnish Permanent Representation at Brussels. It plays a pivotal role in the relationship between the institutions of the EU and the domestic preparation of EU affairs in Finland. It also maintains essential contacts with the Permanent Representations of the other Member States, as well as keeps in touch with officials at all levels in the institutions of the EU, in particular the Commission. The Finnish Permanent Representative is also the member of the Committee of Permanent Representatives (COREPER). Distinct constitutional provisions are designed to secure the participation of Parliament in the consideration of acts and measures that are decided in the Union, but would, according to the Constitution, otherwise fall within Parliament’s competence (section 96), as well as Parliament’s right to receive information (section 97).44 In Parliament, the Grand Committee assumes main responsibility of the handling of EU matters whereas special committees may submit opinions to the Grand Committee. As it is also possible to request the Opinion by the Constitutional Law Committee of Parliament on the compatibility of proposals for EU legislation and other EU matters for their compatibility with the Constitution and international human rights treaties, the outcome is ex ante constitutional review of proposals for EU measures. However, matters relating to the EU’s Common Foreign and Security Policy are handled by the Foreign Affairs Committee. The Government provides information on EU issues to Parliament through Europe communications, Union communications, reports, statements and announcements. Overall, the system for the national preparation of EU affairs, including the participation of Parliament, has proven to be overall relatively efficient in a manner that has also been satisfactory for Parliament. The formally structured constitutional basis of the domestic system for the preparation of EU affairs has brought about stability and transparency to the handling of EU affairs in Finland and has equally managed to provide enough flexibility. Above all, the system has succeeded in serving its major purposes, ie to ensure the parliamentary focus of authority, including the effective participation of Parliament, in the national preparation of EU matters, as well as the possibility of Finland to present a sufficiently coordinated position, in line with its overall EU policy, on issues under consideration in the EU at various stages of preparation. The fact that the system has remained practically intact since the beginning of Finland’s membership also demonstrates that the domestic decision-making system has functioned sufficiently well. However, there has been one significant conundrum of the domestic distribution of powers in EU affairs, and that has related to the role of the President. Although the Government enjoyed general competence in EU affairs, including the EU’s common foreign and security policy affairs under section 93, subsection 2 of the Constitution, section 93, subsection 1 simultaneously provided that foreign policy of Finland ‘is directed by the President in cooperation with the Government’. This complex constitutional regulation has muddied the waters insofar as power constellations between the nexus of EU and foreign policy affairs have been concerned. Up until 2012, there was a great deal of debate over the role of the President in such EU affairs that include foreign and security policy issues. Although the competence of the government covers Common 44 See also, Ojanen, ‘Impact’ (n 16) 554–57, and Viljanen, ‘Eurooppalaistuminen’ (n 17). See also on sections 96 and 97 of the Constitution, Improving EU Scrutiny. Report of the Committee to Assess EU Scrutiny Procedures, Publications of the Parliamentary Office 4/2005.

Finland  743 Foreign and Security Policy issues of the Union, the Presidents insisted their right to be involved in decision-making pertaining to foreign and security policy matters. At level of daily politics, the debate over the role of the President boiled down to a question about the Finnish representation in the meetings of the European Council.45 Although the Constitutional Law Committee of Parliament has consistently emphasised since 1994 that the Prime Minister should represent Finland in the European Council,46 both President Martti Ahtisaari (1994–2000) and President Tarja Halonen (2000–12) were of the view that the President had the right to be present, together with the Prime Minister, in those meetings of the European Council in which common foreign and security policy affairs were on the agenda. As a result, the two Presidents participated in the clear majority of European Council meetings in the late 1990s and the 2000s. However, the entry into force of the Lisbon Treaty brought an end to the practice of ‘dual representation’ of Finland in European Council meetings. It was expressly noted by the Constitutional Law Committee that changes made by the Lisbon Treaty to the role and powers of the European Council rendered it necessary for the Prime Minister alone to represent Finland in the meetings of the European Council.47 After the entry into force of the Lisbon Treaty, the Presidents have no longer participated in the meetings of the European Council. Moreover, the amendment of the Constitution of 2012 codified the new practice in section 66, subsection 2 of the Constitution, according to which the ‘Prime Minister represents Finland on the European Council. Unless the Government exceptionally decides otherwise, the Prime Minister also represents Finland in other activities of the European Union requiring the participation of the highest level of State.’

B.  Constitutional Rules and Practices in the Domestic Implementation of EU Measures The Constitution of Finland unequivocally requires that EU legislation which is of a ‘legislative nature’ must be implemented through an Act of Parliament. EU legislation can be regarded as being of a ‘legislative nature’ within the meaning of the Constitution on such occasions on which the EU legal measure in question regulates the rights and obligations of private parties, or is in conflict with domestic legislation. The most frequently used methods of implementing EU measures are incorporation, either through an Act of Parliament in blanco or through a similar government decree; transformation, involving the enactment of domestic legislation in order to achieve normative harmony with the EU measure in question; and reference, involving special clauses in domestic legislation stipulating that the EU measure in question is to be taken into account in the application of the domestic legislation in question. By and large, the ‘implementation record’ of Finland in European comparison is relatively good, although Finland occasionally fails to abide by the deadline for the domestic implementation of Directives or other EU measures or otherwise fails to implement an EU measure in the most appropriate and effective way. As already noted, the domestic standard of rights protection has on a few occasions compromised the domestic implementation of EU legislation – the European Arrest Warrant Framework 45 For an overview, see Raunio, ‘Semi-presidentialism’ (n 8). 46 The Committee already took this view on the verge of Finland’s accession to the EU in 1994. See Report 10/1994 of the Constitutional Law Committee. For subsequent views of the Committee, see eg Opinions 10/1998; 36/2006; and 13/2008f of the Committee. 47 See the Opinion of 2/2010 by the Constitutional Law Committee.

744  Tuomas Ojanen Decision and the Framework Decision on Combating Terrorism. In addition, there has been constitutional concern that the domestic transposition of EU law may dilute the ‘quality of law’, particularly insofar as implementing enactments constituting an interference with fundamental rights are concerned. In such situations, the permissible limitations test under the Finnish Constitution requires, among other things, that legislative provisions on limitations must be sufficiently clear and precise. Against this background, the Constitutional Law Committee has emphasised the importance of taking into account the quality of law standard in the domestic implementation of EU legislation constituting an interference with fundamental rights. In particular, the Committee has adopted a critical stance regarding criminalisation ‘in blanco’ where an implementing enactment mechanically states that a certain piece of EU criminal law is in force domestically.48 Instead, the Committee has favoured the method of transformation as the appropriate method of implementing EU criminal law, as well as EU legislation entailing limitations on fundamental and human rights.

VI.  Resulting Relationship between EU Law and National Law The incorporation Act of the Accession Treaty (Act No 1540 of 1994) and written Finnish law in general are silent on the issue of the domestic effects and status of EU law within the Finnish legal system. Section 1 of the Incorporation Act of the Accession Treaty only provides that ‘The Accession Treaty of Finland … as well as those treaties that are referred to in Article 1, p ­ aragraph 1 of the said Treaty, is prescribed to be in force as they have been agreed upon’. However, the totality of the Accession Treaty package, as it has been agreed upon, reveals that Finland has subscribed to such fundamental principles of EU law as its primacy, direct and indirect effect, as well as to all other requirements of the EU legal order, and as seen in light of the case law of the EU courts. Finnish courts have not disputed this view. At least so far, there is no evidence of any overt resistance by the Finnish courts to embrace such fundamental qualities of EU law as its direct effect, indirect effect and primacy.49 However, it is somewhat striking that the cases in which Finnish courts have taken advantage of the primacy of EU law and, accordingly, have set aside conflicting provisions of Finnish law have remained relatively few in number, although Finnish courts decide annually thousands of cases falling within the scope of application of EU law. The first reason explaining this may be that Finnish legislation largely, if not exclusively, is in harmony with EU law. Moreover, the interpretative effect of EU law is by far the most frequent method of giving judicial effect to EU law by Finnish courts. This approach assumes special importance in cases in which where there is a prima facie tension between EU law and a provision of national law. In such circumstances, Finnish courts try to interpret, as far as possible, national law in conformity with EU law so as to avoid an open conflict. The Finnish constitutional system may go a long way towards explaining why Finnish courts clearly favour the indirect effect of EU law, instead of applying provisions of EU law directly to the facts of the case. Although the role of the courts has become stronger in recent years and the tendency has been increasingly towards rights-based judicial review of 48 See eg Opinions 9/2007, 53/2006, 40/2002, 31/2002, 6/2002, 45/2001 and 5/1998 of the Constitutional Law Committee. 49 For the reception of EU law by Finnish courts, see T Ojanen and J Salminen, ‘Finland: European Integration and International Human Rights Treaties as Sources of Domestic Constitutional Change and Dynamism’, in A Albi and S Bardutzky (eds) National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (The Hague, TMC Asser Press, 2019).

Finland  745 legislation, the democratically elected legislature still plays a central role in the Finnish constitutional system. As with other Nordic countries, Finland has no experience of totalitarian regimes and, accordingly, the outcome has been the flourishing of such fundamental principles of Nordic constitutionalism as parliamentary supremacy, popular sovereignty, democracy as a majority rule, and aversion to strong forms of judicial review. What also warrants attention is the lack of domestic case law pertaining to the constitutional aspects of EU membership. Indeed, a hard case that would genuinely invite a Finnish court to ponder its acceptance of the primacy of EU law is yet to emerge, although it is over 20 years since Finland became an EU Member State along with Sweden and Austria. Given the current emphasis on the protection of fundamental and human rights in the Finnish legal order and legal culture, as well as the above-discussed occasional constitutional resistance by the Constitutional Law Committee of Parliament, questions concerning the primacy of EU law over fundamental and human rights might present just such a hard case sooner or later.50 As with the European Convention on Human Rights (ECHR), EU membership has contributed to the trend of attaching greater significance to case law as a source of law, thanks to the importance of the case law of the EU courts within the EU legal order. Today, Finnish courts routinely refer to individual judgments by the CJEU, signalling their acceptance of the authority of that court insofar as the interpretation of EU law is concerned. Moreover, EU law has emerged as one of the major reasons behind the shift in legal reasoning from formal, rule-focused reasoning towards more heavily relying on reasoning from principle. Similarly, EU law has for its part affected the methods of interpretation, attaching more significance to what can be described as the teleological or purposive interpretation approach. Up until the end of 2017, Finnish courts had made one 115 references for a preliminary ruling by the CJEU, the Supreme Administrative Court of Finland being at the forefront as it alone had made 56 references for a preliminary ruling by the CJEU.51 Given that Finnish courts annually deal with at least hundreds, if not thousands, of ‘EU law cases’, the number of references is quite low and may even evince some sort of ‘passive judicial resistance’ or ‘shunning’ by Finnish courts of EU law. What is also of significance is that all references have so far concerned the interpretation of EU law only. Hence, a reference for a preliminary ruling inquiring into the validity of EU legislation is yet to emerge from the maze of case law of Finnish courts pertaining to EU law. This is somewhat puzzling in light of the gradually intensified tendencies towards rights-based constitutionalism since the late 1980s.

VII.  Concluding Remarks: From Constitutional Outsider to Insider Finland became a Member State of the EU along with Austria and Sweden in 1995. Aside from significant political, economic, legal, and social dynamics, the Constitution of Finland and the constitutional system in general have undergone a profound transformation due to EU membership. 50 On 25 February 2010, reference for a preliminary ruling from the Supreme Court of Finland was lodged before the CJEU in the case of Public prosecutor v Malik Gataev, Khadizhat Gataeva. The reference included a number of questions aimed at finding out whether the domestic implementing enactment of the European Arrest Warrant was in harmony with the Council Framework Decision on the European Arrest Warrant (2010/C 100/48, [2010] OJ C 100/32). However, the reference was subsequently withdrawn due to the cancellation of the arrest warrant (CJEU 03.07.2010 Order C-105/10 (Gataev and Gataeva) ECLI:EU:C:2010:176). 51 Court of Justice of the European Union, Annual Report 2017, 125.

746  Tuomas Ojanen During the first years of EU membership, the Constitution of Finland was still nationalist and introverted. The Constitution lacked a constitutional provision permitting the transfer of powers to international organisations, let alone the EU. The Constitution also otherwise assumed a minimalist approach to EU membership: nowhere was membership directly acknowledged, and the EU became visible only in constitutional provisions regulating the distribution of powers between Parliament, the Government, and the President. The Accession Treaty of Finland to the EU was given legal effect domestically by the Finnish constitutional idiosyncrasy of exceptive enactment. This enables the incorporation of treaties conflicting with the Constitution by a qualified majority of two-thirds without formally amending the Constitution. As the Constitution was silent on the derogations by EU membership, and the exceptive enactment simply stated that the provisions of the Accession Treaty ‘are in force’ domestically, EU membership initially remained an invisible constitutional outsider. Gradually, however, the constitutional impact of EU membership became stronger and in stages extended to cover practically all levels of Finnish constitutional law, including constitutional culture and such fundamentals as the sovereignty of Finland, democracy, rule of law, the domestic system for the protection of fundamental and human rights, and separation of powers. As a consequence, the text of the Constitution has also begun to increasingly reflect Finland’s EU membership. The entry into force of the new Constitution of Finland on 1 March 2000 provided a more positive attitude towards international cooperation, including European integration, although it also introduced a material limit by providing that an international obligation ‘shall not endanger the democratic foundations of the Constitution’ (section 94, subsection 3). In 2012, a constitutional amendment further addressed the ‘European deficit’ of the Finnish Constitution by providing that ‘Finland is a Member State of the European Union’ (section 1, subsection 3). In addition, new constitutional provisions on the transfer of powers to the EU or international organisations were enacted rendering unnecessary the need for adopting exceptive enactments for such a transfer. As a result of these constitutional amendments, EU membership has become a ‘constitutional insider’ with a firm place among the foundations of the Finnish constitutional system. The Finnish Constitution nowadays reflects a strong commitment to EU membership. Finnish courts and other authorities do not hesitate to embrace their obligations under EU law. However, the domestic political context of EU membership has become more ambivalent in recent years. As with Denmark and Sweden, Finland has witnessed the rise of populism and nationalism, including anti-immigration, anti-EU and even neo-Nazi movements. It cannot be excluded that these kinds of social-political trends will increasingly result in tensions or even conflicts between law and politics even in the Nordic countries. However, there is a long way to go before these contemporary phenomena mutate into such authoritarianism and illiberal outcomes that the very foundations of the Nordic constitutions based on democracy, rule of law and the protection of human rights are endangered.

References R Hirschl, ‘The Nordic counternarrative: Democracy, human development and judicial review’ (2011) 9 International Journal of Constitutional Law 449. J Lavapuro, T Ojanen and M Scheinin, ‘Rights-Based Constitutionalism in Finland and the Development of Pluralist Constitutional Review’ (2011) 9 International Journal of Constitutional Law 505. P Leino and J Salminen, ‘The Euro Crisis and Its Constitutional Consequences for Finland: Is There Room for National Politics in EU Decision-Making?’ (2013) 9 European Constitutional Law Review 451.

Finland  747 A Mutanen, ‘Towards a Pluralistic Constitutional Understanding of State Sovereignty in the European Union? The Concept, Regulation and Constitutional Practice of Sovereignty in Finland and Certain Other EU Member States’, Dissertation, University of Helsinki 2015. T Ojanen, ‘The Impact of EU Membership in Finnish Constitutional Law’ (2004) 10 European Public Law 531. T Ojanen, ‘The Europeanization of Finnish Law’ in P Luif (ed) Österreich, Schweden, Finnland. Zehn Jahre Mitgliedschaft in der Europäischen Union (Vienna, Böhlau, 2007) 145. T Ojanen, ‘The European Constitution in the Far North, in a Country Called Suomi’, in Anneli Albi, Jacques Ziller (eds), The European Constitution and national constitutions: ratification and beyond (Kluwer, 2007) 89–100. T Ojanen, ‘From Constitutional Periphery toward the Center – Transformations of Judicial Review in Finland’ (2009) 27 Nordisk Tidsskrift for Menneskerettigheter/Nordic Journal of Human Rights 194. T Ojanen, ’Constitutional amendment in Finland’, in X Contiades (ed), Engineering Constitutional Change. A Comparative Perspective on Europe, Canada and the USA (Abingdon, Routledge, 2013) 93. T Raunio, ‘Semi-presidentialism and European integration: lessons from Finland for constitutional design’ (2012) 19 Journal of European Public Policy 567. J Salminen, ’Manifestations of the European Union Membership in the Constitution of Finland in the European Context’ (2010) 12 Europarättslig tidskrift 509. J Salminen, ‘Yhä läheisempään liittoon? Tutkielmia valtiosäännön integraationormin sisällöstä ja vaikutuksista’ [‘Towards an Ever Closer Union? Studies in the Constitutional Consequences of Integration’], Dissertation, Turku University, 2015. M Scheinin, ‘Constitutional Law and Human Rights’ in J Pöyhönen (ed), An Introduction to Finnish Law (Helsinki, Kauppakaari, 2002) 31. E Smith, ‘Judicial Review of Legislation’, in H Krunke and B Thorarensen (eds), The Nordic Constitutions: A Comparative and Contextual Study (Oxford, Hart Publishing, 2018) 107. V-P Viljanen, ‘Eurooppalaistuminen valtioelinten välisissä suhteissa’ [‘Europeanisation in International Relations’] (2003) 100 Lakimies 1169.

748

27 Sweden JOAKIM NERGELIUS

I.  Main Characteristics of the National Constitutional System Sweden has no less than four constitutional Acts, namely the Instrument of Government (hereinafter referred to as IG), and also the Freedom of the Press Act (FPA), the Freedom of Speech Act (FSA) and the Act of Succession of 1810. The Swedish legal system is based, like most other legal systems in the world, on the idea of an existing hierarchy of norms. Still, that is not very clearly expressed in the Swedish Constitution, where it may only indirectly be deduced from the role on judicial review in Chapter 11, Article 14 of the IG, according to which courts or other public bodies may set aside provisions that are in conflict with rules of fundamental law ‘or other superior statute’. Without an existing hierarchy of norms, this provision, until recently rarely used, would of course be meaningless. Nevertheless, this concept is yet to be clearly developed in Swedish constitutional law; even the idea of the Constitution as higher law, or lex superior, was unclear at the point of enactment of the IG (which is by far the most important constitutional Act) in 1974. At that time, a perspective inspired greatly by popular sovereignty prevailed, according to which the Constitution contains rules that describe the political situation and issues, such as what should happen and which procedure is to be followed when the Prime Minister is appointed or laws are enacted. In other words, the Constitution was to be primarily descriptive in its nature. The opposite and internationally more common view, based on the idea of constitutionalism, according to which the Constitution is the highest law and against which all other laws are to be measured and reviewed, has only quite recently, in the last 15 or 20 years and in the light of the influence of European law, been seriously discussed. Undoubtedly, that new perspective is now gaining ground. It is observable in the official governmental instructions (direktiv 2004:96) to the parliamentary committee preparing a constitutional reform for 2010 (Grundlagsutredningen).1 Historically speaking, it may be noted that judicial review of legislation did not become a part of the constitutional text until 1979, when Chapter 11, Article 14 IG was enacted. The existence of judicial review as such has however been discussed in the legal doctrine since at least the 1880s.

1 In the doctrine, this tension between old and new ideas and perspectives, between popular sovereignty and constitutionalism in other words, is analysed in particular by Joakim Nergelius, Svensk statsrätt, 4th edn (Lund, Studentlitteratur, 2018) in ch 1 and in the structure of the book, which is built around this distinction. The main report from Grundlagsutredningen is SOU 2008:125, followed by the Governmental Bill 2009/10:80.

750  Joakim Nergelius It was generally acknowledged by the doctrine in the 1930s and finally accepted as a fact by the Supreme Court in a case in 1964.2 It was not until 1979, in the second reform wave of the new Constitution of 1974, that judicial review, which actually presupposes a hierarchy of norms, was finally written into the new constitutional text. This was limited by the reservations that followed from the requirement of so-called ‘manifest error’ and also by some influential political statements in some of the travaux préparatoires. A notable example was by the Constitutional Committee, arguing that restraint by the courts in their exercise of this new possibility was necessary to avoid the gradual undermining of Swedish popular sovereignty and democracy.3 In its text, unchanged between 1979 and 2010, Chapter 11, Article 14 stated that if a court or other public body found that a provision conflicted with a rule of fundamental law or other superior statute, or that a procedure laid down in law had been disregarded in any important respect when the provision was made, the provision could not be applied. However, if it had been approved by the Parliament or the Government, it should be waived only if the error was manifest (the so-called uppenbarhetskrav, or requirement of manifest error). Thus, judicial review pertained and continues to belong formally to both the courts and other public authorities,4 a fact that has in reality probably diminished the importance of judicial review. The main trait or characteristic of the Swedish Constitution, identified here as IG, is in line with the above-noted emphasis on popular sovereignty. According to its opening sentence, in Chapter 1, Article 1, Section 1, ‘all public power in Sweden proceeds from the people’. This is not only a declaratory statement against the concept of separation of powers, but a viewpoint and attitude that influences other parts of the Swedish constitutional system. One aspect of this principled approach, though perhaps not the most important one, is the relative ease with which the four constitutional acts may be changed. According to Chapter 8, Article 14 IG, the constitutional acts or fundamental laws are enacted, changed or amended by means of two parliamentary decisions of identical wording (see also Chapter 8, Articles 15 and 18). The second decision, however, may not be taken until elections for the Parliament have been held and the newly elected parliament has been convened. At least nine months shall elapse between the day when the matter was first submitted to the Swedish Parliament (Riksdag) and the date of the election, unless the Constitutional Committee of the Parliament (Konstitutionsutskottet, KU) grants an exception from that provision by means of a decision supported by five-sixths of its members.5 But it should be noted that a qualified majority of the 349 MPs is not required on any of those occasions. Instead, a majority of the MPs present and voting is sufficient both times, which is surprising and even astonishing when viewed from an international perspective. A certain protection for parliamentary minorities is provided by Chapter 8, Article 16 IG, which stipulates that a referendum shall be held on a proposal concerning a fundamental law which is held in abeyance over an election, on a motion to this effect by at least one-tenth of the MPs, provided that at least one-third of them will concur in approving the motion.6 2 Nytt Juridiskt Arkiv (hereinafter NJA) 1964, 471, concerning opening hours in shops (although that law was actually considered to be constitutional). From 1981 on, NJA can be referred to online at https://lagen.nu/dataset/dv (in Swedish). For a thorough historical account, see Nergelius, Statsrätt (n 1) 213 ff. 3 See in particular KU 1978/79:39, 13. 4 See now Ch 12, Art 10 IG, which is formally identical with Ch 11, Art 14, though only the latter article has any real significance. 5 In Section 2 of the same article it is also stated that the Riksdag may not adopt a decision held in abeyance over an election a proposal for the enactment of fundamental law which conflicts with any other proposal concerning fundamental law currently being held in abeyance, unless at the same time it rejects the proposal it first adopted. 6 Such a motion must be put forward within 15 days from the date on which the Riksdag adopted the proposal which is to be held in abeyance.

Sweden  751 The referendum will then be held simultaneously with the parliamentary election. The proposal is rejected if a majority of those taking part in the referendum vote against it and if the number of those voting against will exceed half the number of those who registered a valid vote in the general election. It could be mentioned here that only two referenda on EU matters have been held in Sweden. The first was held in November 1994, when 52.3 per cent voted for membership and 46.7 per cent were against. The second referendum was on EMU membership, the legality of which was extremely doubtful, held in September 2003 when 56.7 per cent voted against. Apart from that, only four national referenda have been organised.7 Still, from an international perspective, it seems clear that the Swedish constitutional Acts are unusually easy to change or amend. This point is emphasised by the fact that the highly technical nature of the FPA and the FSA, which both contain huge sections of criminal, procedural and civil law, has led to a series of regular constitutional changes (in total more than 200) since 1974. When no constitutional amendments were proposed before the election in 2006, this was the first and only time since 1974 that this had happened! It should also be noted that the possibility for a referendum on a proposed constitutional amendment mentioned above, which should in its result be legally and constitutional binding, has so far never been used. From a political point of view, however, those rules are best understood from the aspect that they contribute to consensus concerning constitutional changes among the main political parties, particularly the social democrats and the right-wing parties. In fact, politically disputed or contested constitutional amendments have so far been extremely unusual, a fact that has, in a way, until now diminished the practical importance and consequences of the fact that the constitutional acts are so comparatively easy to change or amend. According to Chapter 1, Article 4, Section 1 of the IG, the Parliament or Riksdag is the foremost representative of the people. It is thus logical, given the emphasis on popular sovereignty that is so characteristic for the Swedish Constitution, that legislation is a dominant source of law and that it is easy to pass. The main rule on how legislation is passed is to be found in Chapter 4, Article 7 IG, which states that when a vote is taken in the Parliament, the opinion in which more than half of those voting concur will constitute its decision. Enactment of legislation is of course included here, without being expressly mentioned. There are specific rules on requirements for a qualified majority, in certain exceptional situations, to be found both in the IG and the special Riksdag Act (Riksdagsordningen), but these are highly irregular. However, at the same time, regardless of its traditional emphasis on popular sovereignty, it is clear that the Swedish Constitution is currently going through a transitional phase. It is moving away from a traditional, total focus on popular sovereignty to a situation characterised by a separation or division of powers between many different actors (or by a kind of ‘constitutional pluralism’). As follows from Chapter 10, Article 6 and Chapter 8, Articles 14–15, the transfer of powers to the EU is quite clearly regulated by the Constitution, while the supremacy and direct effect of EU law that follows from EU membership is not as such regulated or even mentioned in the Constitution (or in any other law for that matter). Instead, this has been left to the courts to regulate and here it is evident that both of these two crucial concepts have now been acknowledged

7 These have concerned the use and sale of alcoholic beverage (1922), driving on a certain side of the road (1955, when 89% of the voters wished to continue driving on the left side of the road, only to face a change in 1967 anyway), 1957 (on the pension system), and 1980 (on the use of nuclear power stations).

752  Joakim Nergelius by the Swedish courts, and thus in Swedish law, as explained below. This is in fact one of the main reasons behind the recent Swedish constitutional changes.

II.  Constitutional Culture A.  The Position of Constitutional Rights in the Constitution Chapter 2 of the IG is rather long (25 articles) and contains provisions on most of the human rights enumerated in various catalogues of this kind, eg the well-known conventions of the United Nations and the Council of Europe. What is striking, however, is the emphasis put on the techniques and possibilities for limiting the scope of the rights by ordinary legislation. The European Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR) has in fact been a part of Swedish law since 1 January 1995. No law or other provision contrary to Sweden’s obligations under the Convention may be promulgated (IG Chapter 2, Article 19). The exact meaning of this is unclear, as opinions differ as to what the status of such a law should be if it were to be promulgated nevertheless. Even the preparatory documents are silent on this issue. Given the traditionally sceptical attitude of most of the political parties in Sweden towards judicial review, it was impossible to write into the Constitution that the ECHR would prevail over a conflicting Swedish law, since this would have enabled Swedish courts to set aside Swedish laws in a potentially high number of cases. Still, it seems to be difficult to avoid the conclusion that Swedish courts do indeed have to do precisely that when serious conflicts between the two legal systems arise. The applicable and binding rules on human rights in Sweden are to be found in Chapter 2 IG, in the FPA of 1949 and the FSA from 1991, as far as freedom of speech in books and the media is concerned, and finally, since 1995, in the ECHR. Chapter 2 IG guarantees the freedom of expression, freedom of information, freedom of assembly and manifestation, freedom of association and freedom of worship to all citizens. As far as the freedom of expression is concerned, reference is made to the FPA and the FSA, which regulate freedom of expression in the media. Some forms of free speech, such as art exhibitions and theatre performances, are however exclusively protected by Chapter 2 IG. The same applies to political meetings, subject to the condition that they are not broadcast; in the latter case, the FSA is applicable. Chapter 2 protects the ‘negative’ dimension of the above-mentioned freedoms. All citizens are thus protected against coercion by the public authorities to disseminate or not to disseminate their political, religious or cultural views, to participate in or abstain from a meeting or manifestation or to join or quit a political, religious or cultural association. As concerns procedural rights, it has to be noted that retroactive legislation is prohibited both in penal and tax law (IG 2:10)8 and that courts may not be established for specific cases. More complicated is the provision that every citizen shall be entitled to a hearing before a court if he is deprived of his liberty after having committed a crime or because he is suspected of such an act (or for other reasons). Here, the definition of deprivation of liberty has caused some difficulties in relation to administrative sanctions like the obligation formerly imposed on foreigners to reside in and not leave certain municipalities (which has now been abolished).

8 That this may be important, though it has sometimes been questioned, is shown by the case NJA 2000, 132, where a tax law was seen as having retroactive effects and thus was considered unconstitutional by the Supreme Court.

Sweden  753 Some further rights are protected by the constitutional text but in weaker terms, such as the right to exercise a trade and practise a profession, or the rights of the Sami population. This is also true for the protection of property, which was reinforced in 1995 but is still rather easy to restrict (although compensation must always be promptly paid).9 We may also note the weak or unclear legal position that the ECHR has in Swedish law, where, according to IG Chapter 2, Article 19, no act of law or other provision may be adopted which contravenes Sweden’s obligations under the Convention. The ECHR as such was incorporated through an ordinary law (1994:1219). As mentioned above, this does not say anything about what would actually happen should a Swedish law nevertheless be contrary to this provision.10 In other words, the supremacy of the ECHR in relation to Swedish national law is not explicitly recognised. Chapter 2, Articles 12–13 as well as Articles 20–24 IG are in one way or another concerned with limitations of the rights guaranteed in Chapter 2. Articles 20–24, which will be addressed here in greater detail, are somewhat complex. The relationship between these articles is that Article 20 mentions articles that may be limited by law, while Articles 21–22 lay down general conditions for restrictions of the rights that may in fact be restricted. Articles 23–24 then regulate more closely the conditions for the restriction of the different forms of free speech and freedom of information, assembly and manifestation as well as association referred to in Article 1. Article 20 begins by mentioning that these rights, together with those guaranteed in Articles 6 and 8 as well as the right of public attendance at a court trial may in fact be restricted by an Act of Parliament. Such restrictions may, however, be imposed only to achieve purposes that are acceptable in a democratic society, and must never be disproportionate with regard to such purposes. Furthermore, they may not be extended so far as to constitute a threat to the free formation of opinion, and must never be imposed solely on the grounds of political, religious, cultural or other opinions. A parliamentary minority which comprises at least one-sixth of Members of Parliament (hereinafter MPs) voting on the issue has the right to postpone the final adoption of a law restricting the exercise of basic rights (Article 22). Whether a bill is in fact of this restrictive kind or not will be decided by the KU. As for the other freedoms guaranteed in Chapter 2 (Articles 14–18), they may be restricted by law in accordance with the conditions in the provision defining the right. The constitutionality of such legislation, eg in relation to property, may of course be reviewed by the courts. The rights related to free speech may be restricted for the purposes defined in Articles 23 and 24, eg public safety, the integrity of the individual and national security. Some of these purposes, such as national supply of vital goods and services, may seem a bit odd as a justification for the restriction of freedom of speech, as may the circulation of traffic as a justification for restricting freedom of assembly. It can also be observed that the freedom of expression may always be restricted ‘where particularly important grounds so warrant’, a limitation formulated in very broad terms. Generally speaking, the scope of the individual rights guaranteed in Chapter 2 IG is quite comprehensive, but the emphasis on the conditions and the forms in which they may be restricted reflects the general tendency towards parliamentary rule discussed above. Apart from the fivesixths majority required for the rejection of a request for the postponement of the adoption of a

9 As shown most recently in the case NJA 2014, 332. 10 This issue is not dealt with in a satisfactory way by the travaux préparatoires either; see prop (bill) 1993/94:117 and KU 1993/94:24. For criticism against this fact, see Ulf Bernitz, European Law in Sweden. Its Implementation and Role in Market and Consumer Law (Stockholm, Faculty of Law Stockholm University, 2002).

754  Joakim Nergelius law restricting certain basic rights (in particular rights related to freedom of speech and personal liberty), qualified majorities are not needed in order to restrict or limit a fundamental right.

B.  Judicial Review In the Swedish Constitution, as noted above, the idea or even existence of a hierarchy of norms may only indirectly be deduced from the provision on judicial review in Chapter 11, Article 14 IG. According to this, courts or other public bodies may set aside provisions that are in conflict with rules of fundamental law ‘or other superior statutes’. Once again, the main trait or characteristic feature of the Swedish Constitution, identified here as IG, is its emphasis on popular sovereignty. This is exemplified by its above-mentioned opening sentence in Chapter 1, Article 1 IG, which states ‘[a]ll public power in Sweden proceeds from the people’. This is a declaratory statement of a general constitutional attitude and is directed against the concept of separation of powers, since courts are never elected by the people. However, since 1 January 2011, Chapter 11, Article 14 has had the following wording: If a court finds that a provision conflicts with a rule of fundamental law or other superior statute, the provision shall not be applied. The same applies if a procedure laid down in law has been disregarded in any important respect when the provision was made.

In the case of review of an act of law under paragraph one, particular attention must be paid to the fact that the Riksdag is the foremost representative of the people and that fundamental law takes precedence over other law.11 In particular, the second section of this article requires further analysis. It may of course be seen as an attempt to reconcile the old popular sovereignty with the somewhat newer elements of separation of powers and judicial authority that do now influence the constitutional development to a high degree. However, it has also been suggested that it above all underlines the supremacy of the Constitution, since it is the Parliament that enacts or changes the Constitution (see Chapter 8, Articles 14–17).12 One thing that is clear from the wording of Chapter 11, Article 14, however, is that the review exercised by a court may be either formal or material in its nature. In contrast, what is not quite clear from the Article is whether the courts have an obligation to exercise judicial review whenever they find such a situation in a case before them or whether they may do so only when the argument is being raised by the parties in a case.13 Surprisingly enough, this issue remains unresolved. What is clear is that there seems to exist no autonomous right to an abstract judicial review, independent of a specific case or dispute. Although established in Sweden in 1987,14 it was not until the so-called Unibet case that the European Court of Justice confirmed that EU law does not impose any specific conditions on the Member States in terms of allowing specific forms of judicial review, provided effective remedies do exist before independent courts of law.15 From jurisprudence, although not from the text itself, it also becomes clear, logically enough and in line with what was just said, that judicial review is normally not of – or focussed on – a law or 11 An identical rule on judicial review of the public authorities now exists in Ch 12, Art 10 IG. 12 See Nils Karlson, ‘Lagrådets roll – idag och imorgon’ [‘The role of the Law Council – Today and Tomorrow’], (2009) Svensk Juristtidning 269. 13 See Nergelius, Statsrätt (n 1) 255 for further discussion. 14 NJA 1987, 198. 15 ECJ 13.03.2007, Case C-432/05 (Unibet), ECLI:EU:C:2007:163.

Sweden  755 some other rule in itself but rather of the practical use or application of the rule by a public authority in a specific case.16 Swedish courts have acknowledged the supremacy of EU law in some important cases, notably since 2005 (while generally using the Swedish word företräde, which may be translated as either supremacy or priority). However, since legal support for the supremacy principle cannot be found in Chapter 10, Article 6 IG, it must instead be derived from general principles of EU law, which is also what the Supreme Administrative Court did in an important landmark case in 1997, as explained below.17

C. Legislation There are rules in the special Riksdagsordningen regulating the internal life of the Riksdag on the various parliamentary committees (utskott), not least on their role in the preparation of legislation. The IG is rather silent on this point. It simply states in Chapter 4, Article 3 that the Riksdag elects committees from among its members in accordance with the rules laid down in the Riksdagsordningen and that those shall include a Committee on the Constitution (KU) and a Committee on Finance (Finansutskottet, FiU). Thus, those two committees are constitutionally regulated and therefore enjoy a somewhat stronger position than the other ones, of which the Committee on Taxation (Skatteutskottet, SkU) is the only one mentioned in an ordinary provision of the Riksdagsordningen. The other committees (currently 14) can be abolished or changed by a single decision of Parliament, due to differences between the ordinary and additional provisions of the Riksdagsordningen stipulated by Chapter 8, Article 17 IG (according to which the ordinary provisions of the Riksdagsordningen are changed in the same way as the constitutional acts or by one single decision supported by a majority of three-quarters of the voting MPs, while additional provisions may be changed in the same way as ordinary laws, ie by a single decision supported by a majority of those voting). In Chapter 4, Article 5 IG it is also mentioned that any matter raised by the Government or by an MP shall be prepared by a committee before it is to be decided, unless the IG states otherwise. Thus, the position of the committees is in fact very strong, particularly in Swedish political culture, where minority governments are frequent18 and parliamentary compromises are often sought within the committees.19 Every committee consists of an odd number of members not lower than 15 and currently 17. Proportionality between the political parties is always aimed for. The committees must deliver reports on all matters which have been referred to them and which have not been withdrawn.20

16 For some very clear examples of this, see the two so-called Kurd cases, NJA 1989, 131 and 1990, 636. Also NJA 1986, 489 may be observed here. 17 The so-called Lassagård case, Regeringsrättens årsbok (RÅ) 1997, ref 65. This case is commented in detail by Joakim Nergelius, ‘The Impact of EC Law in Swedish National Law – A Cultural Revolution’, in I Cameron and A Simoni (eds), Dealing with Integration Perspectives from Seminars on European Law 1995–96, Vol 2 (Uppsala, Iustus, 1998) 165. See also Torbjörn Andersson, ‘Effective Protection of Community Rights in Sweden – Judicial Review of Administrative Decisions applying the Common Agricultural Policy’, in I Cameron and A Simoni (eds), Dealing with Integration Perspectives from Seminars on European Law 1995–96 (Uppsala, Iustus, 1996) 141. 18 Since the IG was enacted, majority governments have only occurred 1976–78, 1979–81 and 2006–10. 19 Matters are divided between the different committees according to their ‘subject area’, as follows from rules in Chapter 7 Riksdagsordningen. A committee may transfer a matter to another committee if deemed necessary, provided that the latter one consents. 20 The right of the committees to take initiatives of their own is in fact very scarcely used, with an exception for the general right to take such initiatives for the FiU.

756  Joakim Nergelius That matter brings us to how decisions are finally made in individual matters. This is in fact based on the assumption that matters initiated within the parliament shall be treated by different committees before returning to the chamber for a decision of some kind. The general rule is that a committee report shall be notified to the Chamber and then tabled twice there before decision (see Chapter 11, Articles 2 and 19 of the Riksdagsordningen, which includes a few exceptions from the main rule). The MPs may reflect upon the report before moving for adoption or rejection of its various proposals (Chapter 11, Article 4). The matter itself is then taken up for settlement once the Chamber has declared the debate closed, in response to a proposal from the Speaker. It is then normally settled by acclamation or, if an MP so requests, by the holding of a special vote. The settlement by acclamation is considerably less complicated. Where a special vote is held, the Speaker has a considerable influence on the procedure and on which proposals are put against each other, not least in the case of a tied vote. In that case, a lottery may ultimately decide the outcome, although Swedish politicians have been trying to avoid lottery situations ever since the days of the equilibrium between the two main blocks, with 175 MPs representing both sides, from 1973–76. Now, since the Riksdag has 349 members, that ought not to be possible any more, but could still occur if an MP is absent or in similar situations.21 Needless to say, the committees in general and the KU in particular have also been involved in numerous discussions, before and after the enactment of the IG, on how to interpret and apply the former and new rules on judicial review.22 Generally speaking, it must be said that this committee and the MPs in general did not anticipate the rather quick development of the jurisprudence after 2005 or 2010 but do now seem to accept it.

III.  Constitutional Foundation of EU Membership and Closely Related Instruments A.  Historical Background In order to explain the reasons why Sweden joined the European Union in 1995, some historical background is necessary. To start with, the effects of the economic crisis in the early 1990s were very important in this respect. Suddenly, Sweden was no longer seen as a model for the rest of the world, but rather a symbol of a societal model that no longer worked and was in need of profound change. The initial response to the crisis by the Swedish population, in the aftermath of the events of 1989, was to be strongly in favour of joining the EU. At that point, EU membership was perceived as a first necessary step towards renewed economic prosperity. Gradually, however, from 1992 until early 1994, this opinion changed and instead the Union came to be seen as a threat to the Swedish welfare state. Finally, Sweden voted in favour of joining the EU in November 1994 (by 52.3 per cent), after Austria and Finland had both voted in favour (with considerably larger majorities).23 Economic 21 Concerning details in the voting rules, see Nergelius, Statsrätt (n 1) 63 ff. 22 See eg KU 1978/79:39 and for a comment in the doctrine, Bertil Bengtsson, ‘Om konstitutionsutskottets lagprövning’ [‘On the Constitutional Review of the Constitutional Committee’], in B Bengtsson, 14 uppsatser 1983–1991: ersättningsrätt, kontraktsrätt, fastighetsrätt, offentlig rätt och processrätt (Stockholm, Allmänna förlag, 1992) 345. 23 For a comparison with these two countries, and their respective reasons for joining the EU, see Joakim Nergelius, ‘Reasons Why a Small Country Would Want to Become a Member of the European Union (Taking the Nordic Countries as Examples)’, in Institut Suisse de Droit Comparé Lausanne, European Integration: History and Perspectives (Zurich, Schulthess, 2002) 65.

Sweden  757 arguments affected the outcome significantly, but if the Swedish EU debate is compared with, for instance, the situation in Denmark, it may be observed that the idea of the EU as a political instrument was not at all neglected or rejected in Sweden. On the contrary, in 1994 the EU was described by the Social Democrat Government that had then just returned to power as mainly a political means to balance the power of global capitalism.

B.  The Constitutional Provisions The new Chapter 1, Article 10 which since 2011 states that Sweden is now a member of the EU, may be worth noting once more. In 1995, when Sweden joined the EU, there was no constitutional provision reflecting this fact. As follows from Chapter 10, Article 6 and Chapter 8, Articles 14 and 15 and as mentioned above, the transfer of powers to the EU is quite clearly regulated by the Constitution. However, the supremacy24 and direct effect of EU law that follows from EU membership is not regulated or even mentioned in the Constitution (or in any other law for that matter). Instead, this has been left to the courts to regulate and it is clear that both of these crucial concepts are now acknowledged by the Swedish courts, and thus in Swedish law, with quite dramatic consequences, as explained below. The constitutional regulation covering EU membership as well as other kinds of international agreements is found in Chapter 10, Article 6 of the IG. A few words could be said about the historical background of this regulation, which today reads as follows: Within the framework of European Union cooperation, the Riksdag may transfer decision-making authority which does not affect the basic principles by which Sweden is governed. Such transfer presupposes that protection for rights and freedoms in the field of co-operation to which the transfer relates corresponds to that afforded under this Instrument of Government and the European Convention for the Protection of Human Rights and Fundamental Freedoms.

The Riksdag may approve a transfer of authority, provided that at least three-quarters of those voting and more than half the members of the Riksdag vote in favour of the decision. The Riksdag’s decision may also be taken in accordance with the procedure prescribed for the enactment of fundamental law. Such a transfer cannot be decided until the Riksdag has approved the agreement under Article 3.25 Roughly speaking, Article 3 was introduced when Sweden joined the European Union in 1995, while some other important articles in Chapter 10 that are dealt with below date back to the enactment of the new Constitution in 1974 or even earlier.26 As of 1964, the possibility to transfer decision-making powers to an international organisation aimed at peaceful cooperation was established, although this could only take place to a limited extent. Whilst these old provisions could not be used to facilitate EU membership, the constitutional reform itself was enacted as early as 50 years ago with a future relationship with the European Community in mind (given

24 As explained above, the Swedish term företräde may be translated either into supremacy or priority, but I prefer the former one, which is in my view more accurate. 25 Quoted from the official publication: Sveriges Riksdag, The Constitution of Sweden. The Fundamental Laws and the Riksdag Act. With an introduction by Magnus Isberg (Stockholm, Sveriges Riksdag, 2016). 26 A useful comment on the historical background of this Article, and its history before 1995, is provided by Karl-Göran Algotsson, Sveriges författning efter EU-anslutningen [Sweden’s Constitution after EU Accession] (Stockholm, SNS förlag, 2000) 273. See also Erik Holmberg and Nils Stjernquist (eds), Vår författning [Our Constitution], 13th edn, (Stockholm, Norstedts Juridik, 2003) 198; and Bernitz, EU Law in Sweden (n 10) 28.

758  Joakim Nergelius that Sweden had applied for association with the Community already in 1961). These rules were then transferred to the new Constitution in 1974 together with the additional provision that competences could also be transferred to an international court.27 However, it may be said that Chapter 10, Articles 7–10 permit transfers of legislative powers, budgetary powers including the use of the assets of the state and treaty-making power, all to a limited extent, to international organisations other than the EU. The possibility to transfer rights of decision-making to the EU itself is, and has been ever since 1995, much wider and more general. From a material point of view, such a transfer is limited only by the standard of the protection of human rights within the EU. From what has been said above, it follows that when Sweden was going to join the EU, the question was whether the three formerly closing sections of Chapter 10, Article 5 (now Articles 7–10) would be sufficient or not. In other words, was a constitutional reform required in order to enable Sweden to join the European Union? The first committee that attempted to deal with this question, the so-called Constitutional Committee for Joining the (then) EC,28 answered in the affirmative but proposed a new rule in Chapter 1, Article 10, according to which the supremacy of any kind of EC law in relation to any Swedish law, including the Constitution, should be clearly stated or declared. However, this proposal faced significant criticism in the spring of 1993, mainly for political reasons. Above all, it was seen as a sign of Swedish resignation or capitulation before Europe and was identified with an unlimited transfer of power to the EU institutions in Brussels.29 Consequently, the Government quickly proposed three other alternatives, one of which was similar to the one finally chosen.30 It should be noted that the solution chosen was very much influenced by the jurisprudence of the German Constitutional Court, the Bundesverfassungsgericht.31 The constitutional background to this is that according to Chapter 8, Article 14 IG, constitutional amendments must be presented in Parliament nine months before a general election. Thereafter, they must be approved by a simple majority of the members of Parliament in two subsequent and identical decisions before and after the election (this rule applies to all four Swedish constitutional acts, ie not only to the IG but also the FPA, the FSA and the Act of Succession). Such an election was going to take place in September 1994 and thus, in February 1994, it was also decided that a referendum on EU membership should be held in November of that same year. Consequently, the proposal for constitutional reform had to be presented to the Parliament in December 1993 at the latest, in order to facilitate EU membership, given that the content of the then existing Sections 2–4 of Chapter 10 Article 5 IG32 was not sufficient for this. Furthermore, from the point of view of the Government, this proposal had to have a different content or at least a different form than the first, unpopular proposal, which had been seen as a kind of gift to the already very strong ‘No side’ in the forthcoming referendum. 27 Some minor changes were also introduced in 1976 and 1985; see Algotsson, Sveriges författning (n 26) 274. 28 SOU 1993:14. 29 For a further analysis of this debate, see Algotsson, Sveriges författning (n 26) 273; as well as Ulf Bernitz, Sverige och Europarätten [Sweden and EU law] (Stockholm, Nordstedts Juridik AB, 2002) 109 ff; and Mats Melin and Göran Schäder, EU:s konstitution: maktfördelningen mellan den europeiska unionen, medlemsstaterna och medborgarna [The Constitution of the EU. The Distribution of Power Between the EU, the Member States, and the Citizens], 4th edn (Stockholm, Norstedts Juridik, 1999) 168 ff. 30 These were all presented in the report Ds 1993:36, ‘Våra grundlagar och EG – förslag till alternativ’ [‘Our Constitutions and the EC – Proposals for Alternatives’]. 31 In particular the so-called Solange cases, (BVerfG 29.05.1974, 2 BvL 52/71 [Solange I] BVerfGE 37, 271 ECLI:D E:BVerfG:1974:ls19740529.2bvl005271; and BVerfG 22.10.1986, 2 BvR 197/83 [Solange II] BVerfGE 73, 339 ECLI: DE:BVerfG:1986:rs19861022.2bvr019783), as well as the Brunner or Maastricht case, BVerfG 12.10.1993, 2 BvR 2134 (Maastricht) BVerfGE 89, 155 ECLI:DE:BVerfG:1993:rs19931012.2bvr213492. 32 Ie the former Sections 1–3.

Sweden  759 Compared with the earlier text, the main changes in the new proposal were that the former requirement that transfer of competences could only occur to a limited extent was abolished and, most importantly, that the previous mention of transfers, in general, to an ‘international organisation for peaceful cooperation’ was replaced by a specific mentioning of transfer of decision-making powers to the three European Communities (which in 2002 was changed to the EU). However, this transfer could only take place under the condition (or, as it read, ‘as long as’, ‘så länge’ (cf Solange as in the German cases)) that the protection of fundamental rights within the EU was as far-reaching as that envisaged by both Chapter 2 IG (dealing with basic human rights) and the ECHR.33 Here, the influence of the case law of the German Constitutional Court seems clear. It seems evident that the governmental committee analysing the constitutional conditions for future EU membership, faced with severe time pressures since it was necessary, due to Swedish constitutional requirements, to present a proposal for constitutional amendment no later than in the middle of December 1993, suddenly stumbled upon the Maastricht judgment of the Bundesverfassungsgericht delivered on 12 October 1993.34 This appeared to present a solution to the Swedish dilemma, since it provided conditions for further transfer of decision-making powers and for future development within the EU. Subsequently, the committee also discovered the two Solange judgments from 1974 and 1986, the language of which later found its way into the Swedish Constitution.35 In other words, the position of the German Constitutional Court was suddenly also the position chosen by the Swedish constitutional legislator. The formulation based on the Solange doctrine (‘så länge’, ‘as long as’) was subsequently changed in 2002, giving way to the rather vague notion of ‘the principles of the form of government’.36 Nevertheless, the basic content of this former first section of Chapter 10 Article 5 IG, now Chapter 10, Article 6, remains the same. So what does this Article mean in reality, then? First and foremost, Chapter 10, Article 6 IG is a rule to be used when new treaties are agreed. It has been used in relation to the ratification of the treaties of Amsterdam, Nice and Lisbon. Thus, it made it possible for Sweden to join the EU once the referendum in November 1994 had resulted in a positive vote. Additionally, it made it possible for Sweden to ratify new treaty changes in a rather uncomplicated manner, since the only material requirement for this was and still is that the protection of basic human rights within the EU continues to be reasonably high (which, as we know, has not been a problem so far37). The reasons for choosing this particular solution were twofold. Firstly, it would mean that in the future Sweden would not be bound to accept the legal situation that would exist at the time of joining the EU and, secondly, legal acts of the Union could not be questioned in Swedish domestic law by using national rules on the protection of basic rights. According to this line of reasoning, which bears clear traces of arguments used not only by the German Constitutional 33 This then means that these two catalogues of basic human rights should be read together, in order to analyse whether the protection within EU law corresponds to their added value of protection of basic human rights. However, the two catalogues are mutually overlapping, albeit with some differences. In reality, if the question would ever become a real problem, the assessment of the applicable human rights standard could turn out to be quite tricky. 34 BVerfG, Maastricht (n 31). 35 BVerfG, Solange I and Solange II (n 31), respectively. 36 A proposal based on governmental bill SOU 2001:19. 37 Instead, a number of new rules on fundamental rights have been included in the Treaty on European Union, starting with the Treaty of Amsterdam in 1999. Furthermore, the CFR became more important in the jurisprudence of CJEU. However, recent case law from the CJEU such as CJEU 26.02.2013 C-399/11 (Melloni) ECLI:EU:C:2013:107, or CJEU 18.12.2014 Avis 2/13 (Accession of the Union to the ECHR) ECLI:EU:C:2014:2454 may indicate a reduced human rights protection; see Joakim Nergelius, The accession of the EU to the European Convention on Human Rights. A critical analysis of the Opinion of the European Court of Justice (Stockholm, Swedish Institute for European Policy Studies, 2015).

760  Joakim Nergelius Court in its Maastricht judgment of 1993 but also by the Danish Supreme Court (Höjesteret) in a similar judgment in 1998,38 legal acts from the EU that fall outside the area where a transfer of sovereignty has taken place (ie outside the areas where the Union is competent to act, cf Articles 2–4 TFEU) are ultra vires and thus not applicable within the domestic legal system or by the national courts, who may simply set them aside. This may, in fact, be too simplified in reality; the main problem with such reasoning is, of course, that according to the settled jurisprudence of the Court of Justice of the European Union (hereinafter CJEU), it has only the competence to decide whether a particular EU act is in fact invalid (because it is ultra vires or due to other reasons).39 Here, national courts and authorities in several Member States seem not to agree with the CJEU. However, should such a situation occur before a Swedish court, where the court might actually be uncertain as to the validity of an EU rule, it would quite simply have to ask for a preliminary ruling from the CJEU according to Article 267 TFEU.40 The reference to the level of protection of basic rights within EU law as a criterion for enabling a transfer of sovereignty is interesting, since this is something which has never before been so clearly identified as an important aspect of the Swedish Constitution. Traditionally, the national protection of basic rights is also rather weak in Sweden, where there is no constitutional court and not even a strong tradition of judicial review by the ordinary courts, as discussed above. Once again, the influence of the German constitutional jurisprudence, ‘discovered’ by the constitutional legislators at a crucial moment in 1993 should be stressed. Furthermore, since the protection of basic rights within the EU has grown ever stronger since 1994, as highlighted above, it seems clear that Sweden may continue to transfer decision-making powers to the EU in the case of new treaty amendments or for other reasons. The new declaratory rule in Chapter 1, Article 10, which was enacted in 2010, states that Sweden is a member of the European Union and also that Sweden participates in the UN, the Council of Europe and other forms of international cooperation. An important fact to be observed here once again is the relative ease with which the Swedish Constitution(s) may be changed. According to Chapter 8, Article 14, the constitutional acts or fundamental laws are enacted, changed or amended by means of two parliamentary decisions of identical wording, though the second decision may not be taken until the elections for the Riksdag have been held and the newly-elected Parliament has been convened.41 It must be noted, however, that a qualified majority of the 349 MPs is not required on either of these occasions; instead, a majority of the MPs present and voting is sufficient on each occasion. However, constitutional amendments are normally made with consensus amongst the main political parties. A certain degree of protection for parliamentary minorities is provided by Chapter 8, Article 16, which stipulates that a referendum shall be held on a proposal concerning a constitutional amendment which has been voted by the Riksdag for the first time, on a motion to this effect by at least one-tenth of the MPs, provided that at least one-third of the latter concur in approving the motion.42 The referendum will then be held simultaneously with the election of the new Parliament, although this has so far never happened. The proposal is deemed to have failed if it 38 Danish Supreme Court 06.04.1998, Case I 361/1997 (Maastrichtsagen. Realiteten), Ugeskrift for Retsvæsen 1998, 800 H. 39 See for example the well-known case ECJ 22.10.1987 314/85 (Foto-Frost v Hauptzollamt Lübeck-Ost) ECLI:EU:C:1987:452. Cf on the initial position of Swedish courts Joakim Nergelius, Amsterdamfördraget och EU:s institutionella maktbalans [The Amsterdam Treaty and the EU’s Institutional Power Balance] (Stockholm, Norstedts Juridik, 1998) 156 f, 170. 40 Cf Bernitz, EU Law in Sweden (n 10) 116. In fact, Swedish courts have often been criticised for asking too few questions to the CJEU for preliminary rulings (Art 267 TFEU). The EU Commission in 2004 officially criticised the Swedish courts, in particular the highest instance on this point (2003/2161, C (2004) 3899), which led to some minor changes. 41 Cf n 5. 42 Cf n 6.

Sweden  761 is rejected by more than half of all votes validly cast. If it is successful, it proceeds to the Riksdag for final consideration. These rules are best understood if one bears in mind their political purpose, which is to facilitate consensus on constitutional changes among the main political parties, notably the Social Democrats and the right-wing parties. In fact, politically disputed or contested constitutional amendments have so far been extremely unusual, which has somewhat mitigated the practical impact of the generous rules on the adoption of constitutional amendments.

C.  The Stance of Swedish Courts Towards the Primacy of EU Law Swedish courts have acknowledged the supremacy of EU law in some important cases. However, since legal support for the supremacy principle cannot be found in Chapter 10, Article 6 IG, it must instead be derived from general principles of EU law. An important landmark case in 1997, the so-called Lassagård case,43 is of fundamental importance as to the extension of the scope of judicial review in Sweden and its interaction with EU law. It clearly shows that EU law prevails over domestic law. Secondly, it shows the extent to which EU law has developed the scope of judicial review in Sweden. Application of the general principle of effective legal protection by the Supreme Administrative Court has led to the invalidation of the national regulation limiting such review, and to the adoption of a new provision ensuring the general competence of the administrative courts to review administrative decisions. Both that case and some other, subsequent important cases do in fact merit closer attention here. In Lassagård, the plaintiff applied for an agricultural subsidy to a regional administrative authority (Länsstyrelsen) in May 1995. The application, based on an EC Regulation, was rejected because the time limit was not respected. Lassagård exercised its right to appeal to the superior administrative agency. The agency, however, upheld the previous decision in January 1996. There was no general recourse to judicial review of decisions made by administrative authorities and the review was conditioned by the existence of an explicit statutory provision. Finally, after numerous appeals to different courts, the case reached the Supreme Administrative Court.44 The main question at issue was to consider whether the lack of judicial review was contrary to Community law. Interestingly, the issue of Community law was not elucidated by the lower administrative courts that preferred to rely on Article 6 ECHR. The Supreme Administrative Court stressed that no right to judicial review was explicitly provided by the EU (then EC) law, ie by the applicable EC regulation. Thus, it was necessary to consider the general principles of Community law enshrined in the jurisprudence of the ECJ. The general principle that enshrines the right to effective judicial protection results from the Johnston case,45 in which it was stated that the national constitutions and the ECHR (Articles 6 and 13) constituted sources of inspiration regarding the elaboration of the right to 43 Cf n 17. 44 The company appealed to the District Administrative court in Jönköping in May 1996 (the regional agency was located in Jönköping), and the Court considered that the right to a subsidy constituted a civil right in the sense of Art 6 ECHR. By consequence, Lassagård was entitled to judicial review. However, the competence was not fixed. The Administrative Court of Appeal of Jönköping in August 1996 considered that the absence of judicial review was contrary to Art 6 ECHR and declared the administrative court to be competent. As to the substance, the Court of Appeal handed the case to the District Administrative Court of Halmstad. (Lassagård was located in Halmstad.) The Court declared itself competent but rejected the application since it was lodged too late. Then, Lassagård appealed to the Administrative Court of Gothenburg, that declared itself incompetent. 45 ECJ 15.05.1986 222/84 (Johnston v Chief Constable of the Royal Ulster Constabulary) ECLI:EU:C:1986:206.

762  Joakim Nergelius effective judicial protection. The national court did however not refer to Johnston but to Borelli.46 This case constituted an application of the stare decisis principle, and emphasised that all types of decisions from national authorities regarding individuals’ rights stemming from Community law must be subject to judicial review and that Community law prevails over national law. A Community law right was, thus, encroached.47 Consequently, the Court considered that a court should have jurisdiction to try the case, thereby putting aside the domestic rule, under which there existed no possibility of judicial review. Following the judgment in the Lassagård case, the Administrative Procedure Act (Förvaltningslagen) was amended in 1998 when an Article 22(a) was incorporated so as to give a general competence to the administrative courts to consider appeals against decisions taken by administrative authorities. If any single decision should be seen as symptomatic or symbolic for the development described here, along with the far-reaching changes we have seen in the last few years, it was probably the bold decision of the Swedish Labour Court (Arbetsdomstolen) in April 2005. The case concerned the Swedish Constructors Union who had decided to initiate a blockade against a Latvian company that was doing construction work in a town outside Stockholm, using lowpaid Latvian construction workers, and who could not agree with the Union on the terms of the collective agreement that the latter wanted to sign.48 To start from the beginning, in the spring of 2004, Laval, a Latvian company, sent some 30 workers to Vaxholm in the archipelago east of Stockholm to work on the construction of a school. The Swedish Construction Workers’ Union urged that Laval should adhere to the central or collective labour agreement (kollektivavtal) in the building sector. This had been concluded between itself and the respective employers’ organisation and regulated the levels of payment for different jobs, since Sweden has no laws on minimum wage. When the parties were not able to agree to this, the union initiated sanctions and a blockade of the school under construction, which was later supported by the Electricians’ Union. As a consequence, Laval was prevented from working in Sweden, lost its contract for this particular site and later went bankrupt. However, before its bankruptcy materialised, it initiated proceedings before the Labour Court, urging that the actions brought by the union(s) should be declared illegal and that it should be awarded financial compensation. The Labour Court then asked for a preliminary ruling concerning the compatibility of the actions or sanctions against Laval with EU law (in particular, but not only, in the field of labour law).49 Avoiding the complicated details of Laval,50 we may note that the ECJ found that the freedom to provide services outweighed the right to initiate sanctions aimed at imposing the rules in a collective labour agreement on workers posted from another Member State, although the latter freedom or right could be seen as part of the human rights included as General Principles of EU law. As the ECJ expressed it, Article 3(7) of Directive 96/71 concerning the posting of workers in the framework of the provision of services cannot be interpreted as allowing the host Member State to make the provision of services in its territory conditional on the observance of terms and conditions of employment which go beyond the mandatory rules for minimum protection. 46 ECJ 03.12.1992 C-97/91 (Oleificio Borelli v Commission) ECLI:EU:C:1992:491. 47 Also, the court found the Swedish ordinance incompatible with Art 6.1 ECHR. 48 The Labour Court, which started its work in 1929, is generally seen as a very important part of the allegedly successful ‘Swedish model’. It consists of three members each from the Employers organisation and the National Union, as well as one professional judge. This somewhat corporative character of the court has been subject to criticism, but in ECtHR 26.10.2004, Application No 41579/98 (AB Kurt Kellermann v Sweden) ECLI:CE:ECHR:2004:1026JUD004157998, by a narrow margin the ECtHR found that the composition of the court met the requirements of Art 6 ECHR. 49 Arbetsdomstolen 29.04.2005, Case A 268/04. 50 ECJ 18.12.2007 C-341/05 (Laval un Partneri) ECLI:EU:C:2007:809.

Sweden  763 The Swedish Labour Court then, two years later – a delay that must be some kind of record but was partly attributable to the fact that the court may have wished for legislative changes that would have made its task easier, but that never arrived – settled the case by awarding Laval compensation of roughly EUR 55,000 on the basis that the actions taken by the unions violated EU law.51 As well as the ECJ judgment, this latter judgment was highly controversial, though of course generally accepted. It may be noted that the then ruling right-wing Government in Sweden (2006–14) did not greet it with any enthusiasm. Undoubtedly, this is partly because the fear of provoking the powerful trade union remains strong. In general, the courts have not been wholly unwilling to face up to this new challenge, though some clear errors have been made by different Swedish courts in this respect (not least by the two Supreme Courts), as mentioned above. Still, given the reactions and reflexes of courts, such as the Labour Court, it appears just to state that while the full impact of European law and, in particular, EU law had not dawned on the Swedish courts during the first 10 years of EU membership, this is certainly no longer the case.52 A clear sign or indication of this fact came when the EU Commission in the fall of 2004 addressed the Swedish government with a request for an explanation of the fact that Swedish courts had so rarely, since 1995, asked for preliminary rulings from the ECJ.53 In response, Sweden proceeded to enact a new law on 1 July 2006. This states that if a right to appeal (prövningstillstånd) is not granted by any of the two highest courts in cases where matters of EU law are invoked, those two courts have to state the reasons for such an omission or decision.54 Undoubtedly, this ‘move under pressure’ from the Swedish legislator – which proved that previous critics of the exaggerated cautiousness in this respect from the highest courts were not entirely wrong – has contributed to a sharpened and heightened awareness concerning the importance and relevance of European law among Swedish lawyers. Still, it was only in 2005 that the Supreme Court showed that it was finally willing to grasp and comply with the full effect not only of the supremacy of EU law, but also of the ECHR. In two judgments during that year (as well as in one request for a preliminary ruling), the Supreme Court arrived at conclusions that, in light of its previous case law, can only be described as sensational. This is all the more so, given the Court’s use of argumentations that were highly unusual and unexpected from a historical perspective. The two judgments that were delivered by the Supreme Court both concerned the ECHR, the status of which in Swedish domestic law has in fact been disputed since 1 January 1995, when Sweden joined the EU. At the same time, a decision was made by the Swedish Parliament to 51 Arbetsdomstolen 02.12.2009, AD 2009 nr 89. 52 For a comprehensive though critical study, see Ulf Bernitz, Förhandsavgöranden av EU-domstolen. Svenska domstolars hållning och praxis [Preliminary Rulings of the Court of Justice of the EU: The Development of Swedish Courts’ Attitudes and Practices 2010–2015) (Stockholm, Swedish Institute for European Policy Studies, 2010). 53 Cf n 40. Two cases led the Commission to initiate this action: In part the Supreme Administrative Court’s Barsebäck judgment (RÅ 1999, ref 76), concerning the shutdown of a privately owned reactor in the Barsebäck nuclear power plant. It was criticised that the Court decided such a complex issue, involving both EU law and constitutional law, without asking for a preliminary ruling from the ECJ. Another vivid debate followed the Lyckeskog case (ECJ 04.06.2002, C-99/00, ECLI:EU:C:2002:329), where the Court of Appeal in Gothenburg asked whether it was to be considered as the highest instance in a petty criminal case. During the proceedings in that case, it seems to have occurred to the EU Commission that when the two highest courts in Sweden decide not to grant a prövningstillstånd, they are not obliged to justify their decisions, which means that they could in fact ‘ignore’ Art 234 TEC s 3 (today Art 267 TFEU) without there being any real possibility to control their actions. The Commission, unsurprisingly, found this possibility in combination with the generally low frequency of demands for preliminary rulings from Swedish courts unsatisfactory and thus decided to take action against it. 54 Lag (2006:502) med vissa bestämmelser om förhandsavgörande från EG-domstolen [Law (2006: 502) regarding certain provisions on preliminary rulings from the ECJ]. It was based on the report Förhandsavgörande från EG-domstolen [Preliminary rulings from the ECJ], Ju2004/9463/DOM.

764  Joakim Nergelius finally incorporate the ECHR into Swedish national law, which was necessary in order to make it legally binding in Sweden, given the country’s traditionally dualistic view of the relationship between international and national law. Given the ECHR is also a part of EU law and that the Convention was already being used by Swedish courts as a means of interpretation,55 this was a logical move. That said, it was not done by clearly stating that the Convention was now superior to Swedish law. Instead, as mentioned previously, Article 19 was invoked in Chapter 2 IG, according to which laws may not be enacted if they are contrary to the Convention. This article does however not reveal what would actually happen should a law that contrary to the ECHR in fact be passed by the Swedish Parliament, which is the result of a legally not so fortunate political compromise from 1993.56 One of the results of this solution is that a certain basic right is better protected if it may be invoked under EU law than if it is ‘only’ part of and protected by the ECHR. This is demonstrated by the Lassagård case from 1997, where the fact that a right to a legal hearing and a fair trial could be deduced not only from ECHR but also from EC law was decisive for the outcome, as seen above.57 In other words, the ECHR, despite being referred to in Chapter 2, Article 19 IG, has not been given constitutional status but is incorporated through an ordinary law.58 Chapter 2, Article 19 IG simply states ‘No act of law or other provision may be adopted’ which contravenes Sweden’s undertakings under the ECHR. Thus, as mentioned above this may be seen as a rule directed to the legislator rather than the courts.59 What does this mean in practice? It could of course be seen to indicate that when a conflict, which is not manifest, arises between the ECHR and a Swedish law or ordinance, the law or ordinance shall be applied.60 Concurrently, it must be noted that in the travaux préparatoires, both from the Government and the KU, mention is made of the ‘special character’ and status of the ECHR. This ought to give it a certain weight of its own in future conflicts with domestic laws (or ordinances).61 Based on those remarks, Swedish courts should be able to find at least some room for manoeuvre if they want to secure the respect (or even supremacy) of the ECHR. Still, the Swedish courts were for a long time reluctant to use the ECHR, at least when this would mean setting aside a national law. This is perhaps not entirely surprising against the outlined background. In one case from 1998,62 the Supreme Court as mentioned above even refused to uphold the Convention (as well as the FSA) against a demand from Swedish and Norwegian authorities to execute a penal judgment from a Norwegian court (which was all the more embarrassing since Norway was six months later found by the ECtHR to have violated the

55 See eg, the Supreme Court cases NJA 1988, 572 and 1991, 512 (I). 56 The travaux préparatoires where some not so convincing explanations for the compromise sought are presented can be found in SOU 1993:40, prop. 1993/94:117, and the report from the KU 1993/94:24. For critical discussions in the doctrine, see Bernitz, EU Law in Sweden (n 10), ch 5, ‘The Incorporation of the European Human Rights Convention into Swedish Law – A half Measure’; and Nergelius, Statsrätt (n 1) 174 ff. 57 RÅ 1997, ref. 65. 58 Lag 1994:1500. 59 Cf Iain Cameron, An Introduction to the European Convention on Human Rights, 4th edn (Uppsala, Iustus Förlag, 2002) 155 ff; as well as Bernitz, EU Law in Sweden (n 10) 81 ff. 60 This is the view presented by Cameron, Introduction (n 59) 159. Of course, it shall also be noted that the similarities between ECHR and Ch 2 IG, which contains a rights catalogue, are striking. The main difference is that the protection of personal integrity and family life (Art 8 ECHR) is definitely weaker in the Swedish constitution(s), given not least the wide right to access to documents in Sweden. 61 Prop 1993/94:117, p 37 s as well as KU 1993/94:24, p 19 s. 62 NJA 1998, 817.

Sweden  765 ECHR concerning the same issue).63 At the same time, both the highest courts have been very clear in acknowledging the impact of the principle of proportionality flowing from ECHR.64 In the last few years, questions related to administrative tax sanctions (or penalties) and the possibility for Swedish courts to impose damages when the Convention has been violated have been the focus of the discussion. In both areas, it seems clear that while ECtHR has found that certain violations of ECHR had taken place in Sweden, Swedish courts were slow or even reluctant to accept the consequences of those judgments and come to the same conclusions as the ECtHR in almost identical cases.65 As far as economic compensation for violations is concerned, Swedish courts tended to view this as an exclusive competence of the ECtHR, though this was neither quite in line with the case law of that court or with ECHR itself (see not least its Article 13). The importance of the so-called Lundgren case66 should thus be seen in this light. In that case, which must be seen as a breakthrough for the direct applicability of the ECHR in Swedish law, a man had been informed that he was suspected of economic crimes in 1991. Since then, he had difficulties supporting himself; his company went bankrupt and after that it was hard for him to find any other job. A criminal charge was raised against him in 1993, but the trial did not take place until 1997 and the acquitting sentence from the Court of First Instance did not receive full legal force and effect (laga kraft) before the fall of 1998. This procedural time extension was considered as unacceptable and contrary to Article 6 ECHR by the Supreme Court. Accordingly, it granted Lundgren, the formerly accused man who was now the claimant in the tort case, financial compensation on the basis that Swedish public authorities had not complied with the ECHR. The judgment followed previous cases where lower courts had refused to grant financial compensation due to the fact that the ECHR had been violated by public authorities. Still, the Supreme Court in 2003, in the so-called Holm case,67 had indicated that such financial compensation could after all occur, had the Convention really been violated, which was thus found to be the case here, two years later. Lundgren was awarded compensation from the state for his financial losses. The awarded sum of SEK 700,000 was based on his loss of income during the relevant number of years and calculated according to the rules in the Swedish Tort Law (Skadeståndslagen), although in reality it stemmed from the fact that the Swedish authorities had violated Article 6 of ECHR. In addition, he was also granted compensation for that violation in itself (so-called ideell skada or immaterial damage) to the tune of SEK 100,000. In this respect, the right to compensation, interestingly enough, was considered a part of the obligation that Sweden owes within its national law, according to Article 13 of ECHR, to provide efficient remedies in order to prevent or compensate

63 ECtHR 09.07.1998, Application No 21980/93 (Bladet Tromsø A/S and Pål Stensås v Norway) ECLI:CE:ECHR:1999:0 520JUD002198093. The case concerned Norwegian fishermen who had publicly criticised methods used in hunting seal. For this, they were condemned to penalties by Norwegian courts, which in the case before ECtHR was considered to be a violation of their freedom of speech (Art 10 ECHR). The Swedish case concerned a fisherman, Odd Lindberg, who had made critical comments in a TV programme which was later broadcast in Sweden (and thus also seen in Norway, where almost everyone can watch Swedish TV). He was therefore also sentenced to a fine by a Norwegian court. The Norwegian authorities then chased him, after he had taken refuge in Sweden, eager to have the penalty imposed. They were then helped by the Swedish authorities and the Supreme Court could not find that the ECHR prevented them from executing the fine in Sweden (despite the fact that under the FSA, the editor of the TV programme and not Lindberg himself was legally responsible for any comments made in the programme). From at least two points of view (and maybe also considering so-called ordre public in private international law), therefore, this judgment must be considered as a clear error. 64 See eg, NJA 2001, 439. 65 The procedure for imposing administrative tax sanctions (skattetillägg) has for example been found by the ECtHR to be contrary to Art 6 ECHR in the Cases Janosevic and Västberga Taxi/Vulic v Sweden, 34619 and 36985/97, judgment 23 Jul. 2002, Reports 2002 p. 66 NJA 2005, 462. 67 NJA 2003, 217.

766  Joakim Nergelius violations of the Convention within due time. This jurisprudence has been reinforced in subsequent case law in 201268 and 2014 with respect to the catalogue of rights in Chapter 2 IG.69 The circumstances in the other remarkable case from 2005, the internationally well-known and observed Reverend Åke Green case, which has received considerable attention worldwide, were perhaps even more peculiar. The case concerned a reverend member of a small religious community, and not of the official former state church (Svenska kyrkan). During a sermon in July 2003, he made an unprecedented, unexpected and very brutal or even violent attack against homosexuals as a group. Quoting and referring to the bible he warned, among other things, against an increased spread of Aids through legalisation of homosexual relations, called the legislators who had made such official relations legally possible ‘lunatics’ and denounced homosexuals as a ‘cancer tumour’ in society. He also made it clear that he had been in touch with media before the speech and had actually been active in trying to spread his views in the general debate. Consequently, newspapers and a local TV station were invited to his sermon. The Swedish law on agitation or incitement against certain minorities or groups of people (hets mot folkgrupp) had actually been amended only a few years before to include incitement against persons based on their sexual orientation. Thus, it was fairly clear that Reverend Green had in fact violated this Swedish law and for this he was sentenced for this to one month in prison by the city court in Kalmar. Nevertheless, the question remained whether this Swedish law was compatible with the Swedish constitution and the ECHR, both protecting freedom of speech as well as freedom of religion. The Court of Appeal did not think so and found that freedom of religion, given that the remarks were made during a sermon, outweighed the Swedish legislation (given also that according to the Swedish Constitution, IG Chapter 2, Articles 1, 23 and 24, the freedom of religion is as such unrestricted and not possible to limit by law). The state prosecutor (Riksåklagaren) who has an unlimited right to bring cases to the Supreme Court, appealed against this acquitting sentence and claimed that Green should be sentenced to a prison punishment. The reasoning of the Supreme Court in this case was particularly interesting – and simultaneously slightly puzzling and confusing – from a European law perspective. The Supreme Court analysed the new Swedish legislation in some detail and first of all, quite surprisingly or at least unexpectedly, came to the conclusion that it was not contrary to the protection of freedom of speech and religion in the Swedish constitution. This was done by invoking the requirement of manifest error in IG Chapter 11, Article 14, which may have been right or wrong in this case but was here definitely not done in a very convincing way. Above all, it shall be noted that this restriction on judicial review by the courts applies also in relation to the ECHR, according to IG Chapter 2, Article 19 and its travaux préparatoires.70 Thus, once the Supreme Court had come to that conclusion, it should in fact have been impossible also to claim that the Swedish law was incompatible with ECHR and, consequently, that Green should be acquitted. Nevertheless, the Supreme Court managed to arrive at exactly that result, using a quite complicated line of reasoning according to the case law of the Strasbourg court. This clearly shows that the only kind of speech in religious situations that is not protected by the freedoms of religion and speech in Articles 9–10 of the Convention is so-called ‘hate speech’ or, in other words, clear incitements to violent acts or persecution of specific groups of individuals. Once again, this conclusion may be right or wrong as such71 but in this specific case it led to some quite strange and puzzling results. 68 NJA 2012, 211 (I-II). 69 NJA 2014, 323, 332. 70 But not in relation to EC law, for obvious reasons, as shown eg, by the Lassagård case (RÅ 1997 ref 65). 71 It was in fact based on a heavy amount of quoted case law, including many old and new cases falling under both those two articles, which must as such be described as solid.

Sweden  767 First of all, the Supreme Court showed a clear reluctance to exercise judicial review as such and decide whether the law in question was after all incompatible with ECHR. Instead, it said that if Green were to be sentenced and he then brought the case to the Strasbourg court – which he had in fact publicly declared that he was going to do – it would be likely that the ECtHR would find that Sweden had violated the Convention. But as everyone can see, this judgment is based more heavily on speculation about the possible outcome of a case yet to be brought to Strasbourg, and decided in a certain manner in four or five years from now, than the existing legal rules that the Supreme Court has to apply. This attitude may of course be seen as very pro-European indeed, but at the same time the Supreme Court may here be criticised for not exercising its true competence to decide a case and, in fact, even for abdicating from its position as the highest court of the country (and conveniently, in a very controversial case, more or less handing over that role to the Strasbourg court). What seems to have been the crucial issue, from the point of view of the Swedish Supreme Court, is to be able to avoid applying the Swedish law without really stating that it is contrary to the ECHR. This explains the emphasis and detailed analysis of the jurisprudence of the Strasbourg court in the judgment. Referring to that jurisprudence, the Supreme Court found that it would be impossible to sentence Reverend Green to a punishment, though the new law is still not considered as incompatible with the Convention. Thus, the judgment may be seen as ‘pro-European’ while still expressing a traditional, cautious attitude to the exercise of judicial review.72 Time will tell whether it will in the future be considered as a wise and bold judgment or not. That said, there is no doubt that the Supreme Court adopts a very – perhaps overly – respectful approach in relation to the ECtHR.73 On top of those two judgments, we may then also mention a third interesting case from 2005, where the Supreme Court, contrary to its previous positions, and without too many errors in its handling of EU law, asked for a preliminary ruling in the already mentioned Unibet case.74 That request was particularly surprising since it is an established fact within Swedish law that no right to an abstract judicial review exists, as mentioned above.75 In other words, the courts will only exercise judicial review – under or within the restrictions imposed by Chapter 11, Article 14 IG – when there is a concrete case or dispute pending where the law or legal act in question may be applied and not ever in abstracto, just because a certain party may believe for one reason or another that a certain legal act is unconstitutional.76 Against that background, it was highly unexpected for the Supreme Court in November 2005 to ask the ECJ if this lack of scope in the exercise of judicial review or rather, this limitation of the possible ways under which judicial review may be exercised by Swedish courts, was compatible or not with EU law.77 Considering the circumstances of the case, the request was even more surprising, in particular given that the ECJ has in fact never demanded any specific details or specific design of any particular model of judicial review within the legal systems of the Member States, as long as a right to an effective remedy can be said to exist (also as far as review of the relationship between 72 In this latter respect, it is similar to yet another – strictly national – case on judicial review from 2005, NJA 2005, 33. 73 Apart from that, it also seems clear that it puts opinions made during the exercise of religious activities in a somewhat ‘preferred position’ in relation to other, similar verbal attacks, a fact that became even clearer when the Supreme Court in 2006 came to a totally different conclusion in a similar case where some neo-nazis who had attacked homosexuals in leaflets, using in fact less provocative words than Reverend Green, were found guilty; see NJA 2006, 467. 74 ECJ C-432/05 (n 15). 75 See in particular NJA 1987, 198, where this was clearly declared by the Supreme Court. See also RÅ 1994 refs 277 and 654. 76 In order to simplify here, I do not deal with certain special features of Ch 11, Art 14 IG, like the fact that not only courts but also other public authorities may exercise judicial review and that such review may also concern the compatibility of a regulation or decree with a superior rule. 77 Högsta Domstolen 24.11.2005, Cases Ö-4474-04 and Ö-752-05.

768  Joakim Nergelius EU law and national law is concerned). In this particular case, the gambling company Unibet had been charged by a Swedish prosecutor for violating Swedish gambling and lottery rules. Thus, a criminal case was pending in which Unibet was free to invoke an alleged unconstitutionality or incompatibility with EU law of the Swedish legislation. At the same time, Unibet had sued the Swedish state and asked for financial compensation due to the fact that the company could not freely exercise its trade. And if that was not enough, Unibet had also initiated an administrative procedure in which it was complaining or appealing against not being granted a certain gambling licence. Also in both those two latter proceedings, Unibet was free to invoke any legal arguments it wished, urging the courts to exercise judicial review included.78 Consequently, to put it mildly, the likelihood of the ECJ finding that Swedish law on this particular point was contrary to EU law seemed limited. The judgment confirmed this by emphasising that the ECJ leaves it to the Member States to design particularities in the Court system provided a right to an effective remedy exists. The cautious attitude of the Supreme Court when asking for the opinion of the ECJ in this case definitely stands in strong contrast to the attitude(s) previously shown by that Court in these kind of cases. It is a further example of a change in the positions taken by Swedish courts in relation to European law, the reasons for which we will now analyse. There is in my view no doubt that the case law from the Swedish Supreme Court presented and analysed above does indeed represent a true shift in the positions taken by the highest Swedish judicial body in relation to European law, in particular when compared to the attitude shown before by that particular court (and partly also by the Supreme Administrative Court). The two ECHR-related judgments are thus important and significant. At the same time, the question may be raised if they, along with the preliminary ruling by the Supreme Court in the Unibet case, represent a sincere desire of the Supreme Court to deal with European law in a serious way or instead should be seen as a step back under pressure, at gunpoint so to speak.79 Thus, though highly interesting, those two important judgments and the surprising request for a preliminary ruling concerning a very peculiar constitutional matter, are not entirely convincing in their apparent pro-European convictions. Further substantive proof of a true change of mind from the Swedish judges is wanted and indeed desired, as well as in cases that do not involve European law so clearly, for critical parts of the doctrine to believe this reversal of opinion is to be taken seriously. However, in 2013 the Supreme Court had to acknowledge the supremacy of European law in relation to Swedish legislation, in the so-called ne bis in idem cases.80 Having refused to set aside Swedish legislation imposing administrative tax sanctions and criminal sanctions for the same action81 and refrained from asking for a CJEU preliminary ruling, the Supreme Court in a plenary judgment, as consequence of the Åkerberg Fransson case82 (where a preliminary ruling had been requested by a local court in the north of Sweden), reversed its position and admitted that the system violated the ECHR as well as the Charter of Fundamental Rights of the EU (CFR).

78 The case was a spin-off from one of those procedures in which Unibet claimed that it had the right to have judicial review exercised independently of the specific dispute taking place before the court. When the Court of First Instance rejected that claim, Unibet appealed both to the Court of Appeal and later to the Supreme Court with the latter taking an unexpected interest in the arguments raised by the company. 79 More so since the EU Commission in 2004 had protested against the limited use of the preliminary reference procedure by Swedish courts (cf nn 40 and 53). 80 NJA 2013 502, 746. 81 NJA 2010, 168. 82 CJEU 26.02.2013 C-617/10 (Åkerberg Fransson) ECLI:EU:C:2013:105.

Sweden  769 In Åkerberg Fransson, para 21, the CJEU stated that since the fundamental rights guaranteed by the Charter must therefore be complied with where national legislation falls within the scope of European Union law without those fundamental rights being applicable. The applicability of European Union law entails applicability of the fundamental rights guaranteed by the Charter.

And in para 29, the CJEU stated that 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.

Finally, in para 48, the CJEU stated with a clear message to the Swedish Supreme Court, that had previously made some severe mistakes in its handling of ne bis in idem matters,83 that European Union law precludes a judicial practice which makes the obligation for a national court to disapply any provision contrary to a fundamental right guaranteed by the Charter conditional upon that infringement being clear from the text of the Charter or the case-law relating to it, since it withholds from the national court the power to assess fully, with, as the case may be, the cooperation of the Court of Justice, whether that provision is compatible with the Charter.

At least after these new judgments in 2013, it is clear that the supremacy of EU law is now acknowledged by Swedish courts. Going further, it must be remarked that within Swedish law, EU law offers better protection for basic human rights than the ECHR. This conclusion is based on the fact that Swedish rules were set aside in the Lassagård case because they were contrary to Articles 6 and 13 ECHR as part of EU law, as well as to the general principles of EU law, but not due to the application of the ECHR itself or ‘in its own right’.84 In this regard, it is also interesting to note that the Supreme Administrative Court in Lassagård did not refer to the special rule on judicial review in Chapter 11, Article 14 IG. This constitutional provision reflects the traditional Swedish stance on limited judicial review (in 1997 more so than today, as explained above). Thus, the Lassagård case also confirmed that no constitutional limitations regarding the power of the court to set aside legislation as unconstitutional do apply with regard to EU law. As stated above, as the protection of basic rights within the EU has grown stronger since 1994, it seems clear that Sweden may continue to transfer decision-making powers to the EU. Another possible limit for such a transfer consists of the rather vague notion of the so-called ‘principles of the form of government’. This vague expression was introduced into the constitutional text, Chapter 10, Article 6, in 2002 and must be understood as referring to the important principles for the Swedish Constitution mentioned in Chapter 1 IG (such as monarchy, local and municipal autonomy and impartiality within the public administration). This perhaps stems from the somewhat unique Swedish tradition concerning access to documents and protection for ‘whistle-blowers’ and other media sources as provided for in the FPA and the FSA. Proposals for a referendum on the decisions to transfer sovereignty and decision-making powers to the EU were tabled in relation to the ratification of the EU Constitution in 2005 and the Lisbon Treaty



83 See 84 See

the cases NJA 2010, 168; and NJA 2011 444. Nergelius, ‘Impact’ (n 17).

770  Joakim Nergelius in 2009, but only from the left-wing parties who did not have the number of votes necessary to make this happen.

D.  European Integration Outside the EU Legal Order Sweden joined the Stability and Growth Pact (SGP) and the Treaty on Stability, Coordination and Governance (TSCG) in 2013, but has not been very much affected by the euro crisis. The duties arising from the TSCG, namely those set out in Articles 3(1), 4, 5 and 6, have been accommodated in the national legal order. The official position of the Swedish Government is that Sweden already meets those duties, as a result of economic measures as well as legal and constitutional changes carried out in the 1990s. Thus, there was no need for any further measures for Sweden to sign and ratify the TSCG. In this regard, one of the main arguments presented for Sweden’s ratification of the Treaty was the limited participation in the future decision-making: Article 12 Section 6 provides that the original eight states which joined the Treaty, although not being euro states, shall be present at the special summits whenever possible and at least once a year. In particular, during the euro crisis but also earlier, the pure political problem of remaining outside the euro and the limited influence that this gives to Sweden in the shaping of Europe’s economic policies have from time to another been underlined in the public debate. The emergence of the TSCG thus presented the Swedish Government with a choice between joining the ‘hardliners’ United Kingdom and the Czech Republic, who are unlikely to join the euro in the foreseeable future, or attempting to obtain a limited influence by gaining a position nearer the centre or ‘inside, though still outside’. The result in terms of political influence must, however, be seen as very meagre when Article 12 is analysed. This, of course, also reflects the fact that Sweden’s influence, in legal terms, on the shaping of the ‘ever more detailed economic governance regime for euro area Member States’ has been absolutely none at all.

IV.  Constitutional Limits to EU Integration As mentioned above, it is clear that Chapter 10, Articles 7–10 IG permit transfers of legislative powers, budgetary powers including the use of the assets of the state, and treaty-making power, all to a limited extent, to international organisations other than the EU, while the possibility to transfer rights of decision-making to the EU itself is, and has been since 1995, much wider and more general. From a material point of view, such a transfer is limited only by the standard of the protection of human rights within the EU, which must be equivalent to those of the ECHR and Chapter 2 of the IG. Generally speaking, the scope of the individual rights guaranteed in Chapter 2 IG is quite comprehensive, but the emphasis on the conditions and the forms in which they may be restricted reflects the general tendency towards parliamentary rule discussed above. Apart from the fivesixths majority required for the rejection of a request for the postponement of the adoption of a law restricting certain basic rights (in particular rights related to freedom of speech and personal liberty) stipulated in Chapter 2, Article 22, qualified majorities are not needed in order to restrict or limit a fundamental right. A few words should also be said about the formal requirements for a transfer of decisionmaking powers to the EU. Such a transfer by a single decision of the Riksdag requires a majority of 75 per cent of the voting members; the alternative is that the transfer takes place in the same

Sweden  771 way as an amendment of the Constitution, ie through two identical decisions, one before and one after a general election (Chapter 8, Articles 14–15 IG).85 It may be noted that while Swedish public opinion was for a long time regrettably hostile towards EU membership (as shown most clearly in the controversial referendum on the euro in September 2003), parties that are in favour of membership have had a solid majority within the Parliament, having held more than 80 per cent of its 349 seats at all times since 1994 (and even before). This also means that the transfers of decision-making powers that have been made so far, in 1994, 1998 (Amsterdam), 2001 (Nice), and 2008 (Lisbon), have all been taken through one single decision with the necessary majority.86 From a theoretical point of view, this type of regulation makes it possible for Sweden to adhere to new treaties or approve treaty changes without a referendum and without having to change the Constitution, which is a practical advantage. Also, the process for ratification as such is uncomplicated, not least compared with the requirements in many other Member States. As stated above, the level of protection of human rights within EU law sets the limit here, but since the protection of basic rights within the EU has grown stronger since 1994, it seems clear that Sweden may continue to transfer decision-making powers to the EU.

V.  Constitutional Rules and/or Practice on Implementing EU Law A.  Parliamentary Control The European Parliament (EP) is not mentioned in the Swedish Constitution (except in IG, Chapter 8, Article 2, Section 1 p 6, which makes a short reference to the organisation of EP elections). However, a by-law regulation of the Riksdag concerning its functions in respect of European affairs can be found in Chapter 13 Riksdagsordningen. In fact, it has been changed recently, as will be explained below. Furthermore, according to Chapter 10, Article 10 IG the Government shall keep the Parliament continuously informed and confer with bodies appointed by the Riksdag with regard to developments within the framework of EU cooperation.87 In practice, this consultation and information is done through regular information provided by ministers to a specific parliamentary committee, the so-called EU-nämnden. More detailed regulations on how this information shall take place are found in Chapter 13 Riksdagsordningen. The parliamentary EU Committee is not a regular committee like the other parliamentary committees, which are regulated in Chapter 7 Riksdagsordningen. Still, it is mentioned in Chapter 9, Articles 20–23 Riksdagsordningen, which, according to the rules on amendment of that Act in Chapter 8, Article 16 IG, means that it cannot be abolished by a simple legislative majority. The Committee’s composition reflects and is based on the outcome of the general elections. 85 As follows from Sections 3 and 4 of this Art, this could actually also lead to a referendum on the issue, should onethird of the MPs so wish. So far, however, this possibility has never been used. 86 A question to be asked in this respect is whether there may be any material differences between what may be transferred to the EU using one or the other of the two procedures. Personally, I believe that neither the wording of Ch 10, Art 6 IG nor the practical experience during more than twenty years of membership point towards any such difference, but the issue has been raised in the doctrine (eg by Holmberg/Stjernquist, Författning (n 26) 199 ff). 87 ‘The Government shall keep the Riksdag continuously informed and consult bodies appointed by the Riksdag concerning developments within the framework of European Union cooperation. More detailed rules concerning the obligation to inform and consult are laid down in the Riksdag Act.’

772  Joakim Nergelius Some important changes in the Riksdagsordningen entered into force on 1 December 2009 (ie on the same day as the Lisbon Treaty), concerning specific procedures related to supervision of the principles of subsidiarity and proportionality.88 Based on a report from the KU,89 a document-based or sector-specific model of scrutiny of the principle of subsidiarity was introduced. According to this, the nature of the matter that will be reviewed determines which of the parliamentary committees will be responsible for exercising the new test on whether new legal acts do respect subsidiarity as introduced by the Lisbon Treaty. The parliamentary committee may, within two weeks, ask the Government to give its opinion on whether the legislative proposal in question violates the principle of subsidiarity. If the majority of the committee should then find that the principle of subsidiarity has been violated, this will also probably become the official Swedish position. The committee will make a statement or report to the Riksdag, suggesting that it shall declare the position of the committee to be the official position of the Parliament and that the Parliament should subsequently send this reasoned and justified position to the presidents of the European Commission, the Council of Ministers and the EP. Furthermore, should the contested act nevertheless enter into force, the Parliament may, in line with the Protocol on the application of the principles of subsidiarity and proportionality, urge the Government to initiate proceedings before the Court of Justice.90 Nevertheless, the various committees dealing with other political issues are far more established and have greater decision-making power, notably in legislative and budgetary issues, while the EU Committee has until now been seen only as a forum for discussion and information. Normally, formal hearings of the ministers take place before the meetings of the Council of Ministers, according to the rules in the Riksdagsordningen. The requirement is that at least five members of the EU Committee must request a hearing. Although this is relatively informal and unregulated, the ministers are supposed to represent an official Swedish line that enjoys the support of a significant majority of the MPs. Concerning Articles 48 and 49 TEU, Parliament shall, according to a new rule in Chapter 9, Articles 20–23 of the Riksdagsordningen, approve or reject any such new initiative having first heard the responsible committee on its views. It may be noted that the obligation for the Government to inform the Riksdag has expanded and does now include all documents originating from the EU (Riksdagsordningen Chapter 9, Articles 20–23). Generally speaking, it seems to be a good idea for national Parliaments to organise special committees for scrutiny of EU affairs, of the kind that has existed in Denmark since 1973 and since 1995 also in Sweden. The attention of such committees may – and should – also and not least be directed to economic matters related to the euro (in particular, after the entering into force of the TSCG). The Swedish EU-nämnden thus prepared the implementation of the TSCG, working together with the Committee for Economic Affairs.

B. Referenda Only two referenda on EU matters have been held in Sweden. The first was held in November 1994, when 52.3 per cent voted for membership and 46.7 per cent against. The second was the



88 Further, 89 KU 90 See

but mainly editional changes in the Riksdagsordningen were introduced on 1 September, 2014. 2009/10:2. KU 2009/10:2 p. 8.

Sweden  773 legally extremely doubtful referendum on EMU membership in September 2003, with 56.7 per cent voting against. While the first referendum was held for an obvious reason, the background of the second referendum warrants further explanation. The reason for this peculiar Swedish deviation from its obligations as an EU member simply has to do with domestic politics. After the referendum on EU membership in 1994, only two official No-side parties, the Environmentalists and (former) Communists, together with the small Farmers Party, which had been in favour of EU membership as such but was opposed to the idea of a common currency, demanded a referendum on EMU. Thus, a huge parliamentary majority was against holding a referendum. However, upon returning to Swedish domestic politics in the fall of 1997 and after having acted as a UN peacebroker in Bosnia and Herzegovina for two years, the right-wing opposition leader and former Prime Minister, later also Foreign Minister (2006–14) Carl Bildt, in a surprising move during a parliamentary debate in the build-up to the 1998 parliamentary election, called for a referendum to be held on this topic. Since he had at that time already consulted the Liberals and the Christian Democrats, only the governing Social Democrats (who held 45 per cent of the seats in Parliament) at that time did not formally support the idea of a referendum. This meant that the idea suddenly became a political reality or even necessity. It may also be added that the Social Democrats never formally opposed the idea, which was logical given the internal division within the party over this issue. Thus, no political party or force, including the media, ever really opposed the idea of organising a referendum on the EMU, at least not on legal grounds. The very simple legal fact that EU membership and the result of the referendum in 1994 did not formally or legally allow for this second referendum was never really invoked in the Swedish debate, which is quite surprising.91 From a legal point of view, it is clear that since 1997, when the original decision was made not to participate in the third step of EMU, Sweden has violated its obligations as an EU member. Sweden thus became the first country to meet the so-called convergence criteria but not to participate in the EMU, without any legally binding exception that enables it to remain outside (as in the cases of Denmark and the UK). However, the introduction in 2009 of Article 139 TFEU, with the enactment of the Lisbon Treaty, has probably made Sweden’s situation easier from a strictly legal point of view, since Sweden may now according to this rule be treated as a Member State ‘with an exception’, like Denmark or the UK – even though no such exception has ever really existed. As it turned out, 57 per cent of the voters voted against the euro on 14 September 2003, only three days after the assassination of Foreign Minister Anna Lindh towards the end of the campaign. It may be asked if the result might have been different had the Government spent more time and energy arguing in favour of the common EU currency in 1999 or 2000. The legally rather unclear situation to which the referendum led is all the more regrettable since all the other constitutional preparations for full membership in the EMU were conducted or implemented already in 1998–99 and done so in accordance with the provisions of Chapter 8, Articles 14–15 IG.92 These consisted of amendments to Chapter 9 Articles 12 and 13 IG, regulating the financial power of the state. According to these rules, the Riksbank (central bank of the Realm), working under the Riksdag and formally independent from the Government, is

91 For the only real exception from this, see Bernitz, EU Law in Sweden (n 10) Ch 3; Bernitz, Sverige (n 29) ch 9 (which were written and published before the EMU referendum). See also Joakim Nergelius, En olaglig folkomröstning. Sverige och euron [An Illegal Referendum. Sweden and the Euro] (Stockholm, Santérus, 2008). 92 It may also be noted, in this respect, that Sweden actually conducted an austere economic policy in 1994–96 to meet the convergence criteria necessary for EMU membership. This makes the subsequent development even less logical and harder to understand.

774  Joakim Nergelius responsible for monetary policy. No public authority may determine how the Riksbank decides in matters of monetary policy. The 11 members of the governing council of the Riksbank are all appointed by the Riksdag, which also considers whether the members of the Governing Council or the Executive Board of the Bank shall be discharged of their responsibility. A member of the Executive Board may be removed from office only if he or she no longer fulfils the requirements laid down for performing his or her duties or has been found guilty of gross negligence. All of these rules, aimed at ensuring the independence of the Riksbank, meet the TFEU’s criteria to join the EMU. Consequently, all possible practical measures in Swedish public life during the last 20 years, be they economic, legal or constitutional, have been directed towards facilitating full Swedish membership in the EMU and adopting the euro as the national currency. The only thing that has been lacking is a clear will on the part of the main political parties.

VI.  Resulting Relationship between EU Law and National Law To summarise this report in a few words, it may be fair to say that though the Swedish constitution does not give full recognition to the supremacy of EU law, this has gradually been acknowledged by the jurisprudence of the Swedish courts and must be seen as finally accepted in 2013.93 In addition, other respects of the legal situation and the constitutional regulation are now in line with the requirements following from EU law, with one major exception, which is the lack of participation in the EMU. For this deviation, no legally binding exception exists (except for Article 139 TFEU).

References K-G Algotsson, Sveriges författning efter EU-anslutningen [Sweden’s Constitution after EU Accession] (Stockholm, SNS förlag, 2000). T Andersson, ‘Effective Protection of Community Rights in Sweden – Judicial Review of Administrative Decisions applying the Common Agricultural Policy’, in I Cameron and A Simoni (eds), Dealing with Integration Perspectives from Seminars on European Law 1995–96 (Uppsala, Iustus, 1996) 141. B Bengtsson, ‘Om konstitutionsutskottets lagprövning’ [‘On the Constitutional Review of the Constitutional Committee’], in B Bengtsson, 14 uppsatser 1983–1991: ersättningsrätt, kontraktsrätt, fastighetsrätt, offentlig rätt och processrätt (Stockholm, Allmänna förlag, 1992) 345. U Bernitz, Sverige och Europarätten [Sweden and EU law] (Stockholm, Nordstedts Juridik AB, 2002). U Bernitz, European Law in Sweden. Its Implementation and Role in Market and Consumer Law (Stockholm, Faculty of Law Stockholm University, 2002). U Bernitz, Förhandsavgöranden av EU-domstolen. Svenska domstolars hållning och praxis [Preliminary Rulings of the Court of Justice of the EU: The Development of Swedish Courts’ Attitudes and Practices 2010– 2015) (Stockholm, Swedish Institute for European Policy Studies, 2010). I Cameron, An Introduction to the European Convention on Human Rights, 4th edn (Uppsala, Iustus Förlag, 2002). E Holmberg and N Stjernquist (eds), Vår författning [Our Constitution], 13th edn, (Stockholm, Norstedts Juridik, 2003). N Karlson, ‘Lagrådets roll – idag och imorgon’ [‘The role of the Law Council – today and tomorrow’], (2009) Svensk Juristtidning 269.



93 Notably

through the cases NJA 2013, 502, 746, both following the CJEU’s Åkerberg Fransson judgment (n 82).

Sweden  775 M Melin and G Schäder, EU’s konstitution: maktfördelningen mellan den europeiska unionen, medlemsstaterna och medborgarna [The Constitution of the EU. The Distribution of Power Between the EU, the Member States, and the Citizens], 4th edn (Stockholm, Norstedts Juridik, 1999). J Nergelius, Amsterdamfördraget och EU’s institutionella maktbalans [The Amsterdam Treaty and the EU’s Institutional Power Balance] (Stockholm, Norstedts Juridik, 1998). J Nergelius, ‘The Impact of EC Law in Swedish National Law – A Cultural Revolution’, in I Cameron and A Simoni (eds), Dealing with Integration Perspectives from Seminars on European Law 1995–96, Vol 2 (Uppsala, Iustus, 1998) 165. J Nergelius, ‘Reasons Why a Small Country Would Want to Become a Member of the European Union (Taking the Nordic Countries as Examples)’, in Institut Suisse de Droit Comparé Lausanne, European Integration: History and Perspectives (Zurich, Schulthess, 2002) 65. J Nergelius, En olaglig folkomröstning. Sverige och euron [An Illegal Referendum. Sweden and the Euro] (Stockholm, Santérus, 2008). J Nergelius, Svensk statsrätt, 4th edn (Lund, Studentlitteratur, 2018). J Nergelius, The accession of the EU to the European Convention on Human Rights. A critical analysis of the Opinion of the European Court of Justice (Stockholm, Swedish Institute for European Policy Studies, 2015). Sveriges Riksdag, The Constitution of Sweden. The Fundamental Laws and the Riksdag Act. With an introduction by Magnus Isberg (Stockholm, Sveriges Riksdag, 2016), www.riksdagen.se/globalassets/07.dokument--lagar/the-constitution-of-sweden-160628.pdf.

776

28 United Kingdom PAUL CRAIG

There is inevitably something strange in writing a chapter concerning the impact of UK constitutional law on the limits of EU integration when the United Kingdom has already left the EU. The chapter was, however, drafted prior to the completion of Brexit and it is, in any event, worth reflecting on the important issues posed by this study. This is in part because of their intrinsic significance, and in part because UK membership of the EU has been the catalyst for major judicial decisions concerning the nature of the UK constitutional order, which will continue to have more general relevance even after Brexit. The structure of this chapter follows the general remit of the Member State studies. It begins with a consideration of the principal characteristics of the UK unwritten constitution, followed by reflection on the nature of UK constitutional culture. The focus then shifts to what were the constitutional foundations of UK membership of the EU, and the national constitutional limits on EU integration. The penultimate section of the chapter examines the constitutional precepts that governed implementation of EU law into UK law, and the final section contains more general thoughts about the nature of the interaction between EU law and UK constitutional law while the United Kingdom was still a Member State of the EU.

I.  UK Constitution: Principal Characteristics The United Kingdom has no written constitution, but it does have a number of principles that shape the UK constitutional order. There is also more recent judicial recognition of a category of constitutional statutes that are regarded as especially important in fashioning the UK constitutional landscape. Prominent examples of UK constitutional principles are parliamentary sovereignty, the rule of law and separation of powers, while the category of constitutional statutes includes Magna Carta, the Bill of Rights 1689, the Human Rights Act 1998, the European Communities Act 1972, the Scotland Act and the Government of Wales Act 2006. The courts occupy an important place in the UK constitutional schema, since they interpret the constitutional principles. It is the courts that decide whether to recognise a category of constitutional statutes. While the Supreme Court is most important in this respect, the nature of the UK judicial system is that other superior courts, such as the High Court and Court of Appeal, may also make important contributions to this interpretive exercise. The UK judicial system is, therefore, one in which points of constitutional significance may be raised before any court, the corollary being that no court has a monopoly over constitutional interpretation in the manner that is common in many continental legal systems. This is exemplified by the fact that, as will be

778  Paul Craig seen below, the initial legal decision concerning recognition of a category of constitutional statutes was given by a court other than the Supreme Court.

II.  UK Constitution: Underlying Culture There are two features of UK constitutional culture that are noteworthy, which were especially relevant to relations with the EU. First, the United Kingdom is, in relative terms, open to influence and ideas that emerge from other legal systems. We are in that sense a country willing to import legal concepts, as well as export them. Thus, the UK courts had no trouble accepting the direct effect of EU law, and they regarded it as natural that EU principles of judicial review would bind Member States when acting within the scope of EU law, with the consequence that principles such as proportionality and legitimate expectations would be applied in cases of this nature, even if they were not part of the domestic legal order. Second, the UK Constitution is also adaptable, and able to respond to novel challenges. This is exemplified by the judicial acceptance of the supremacy of EU law in the Factortame case.1 The parliamentary debates and academic discourse concerning UK membership of the EEC in the 1970s were replete with analysis of the difficulty, or impossibility, of the United Kingdom accepting the supremacy of EEC law, given the prominence within the UK legal order of the sovereignty of Parliament.2 The legal reality proved to be otherwise. Thus, when we joined the EEC in 1972 the initial approach of the UK courts was to avoid the supremacy issue if possible through rules of interpretation or construction, the assumption being that it was not necessary to deal with a difficult problem, until it really could not be avoided on the facts of a particular case. When the issue could not be avoided in the Factortame case, Lord Bridge in the House of Lords managed to secure an accommodation between traditional conceptions of parliamentary sovereignty and membership of the EU in the manner explicated below. A problem that had been thought to be impossible, or excessively difficult, was surmounted through the adaptability of the common law.

III.  EU Membership: UK Constitutional Foundations The absence of a written constitution in the United Kingdom means, ex hypothesi, that there is no written constitutional provision that defined the nature of the relationship between the national and EU legal orders, or the terms of that relationship. It was, therefore, for the courts to determine this relationship in the light of existing primary legislation. This was clarified by the Supreme Court’s decision in the HS2 case,3 which shed important light on the conceptual foundation on which EU law was accorded supremacy in the event of a clash with UK law. The ECJ made clear from the outset that EU law has supremacy over national law in the event of a clash, but it is nonetheless for each Member State to decide whether to accept this supremacy, and whether to ground its acceptance in the ECJ’s communautaire reasoning, or in domestic constitutional principle, or in some admixture of the two.4 1 House of Lords (hereinafter UKHL) 09.07.1990, Factortame Ltd v Secretary of State for Transport (No 2) [1991] 1 AC 603. 2 Danny Nicol, EC Membership and the Judicialization of British Politics (Oxford, OUP, 2001). 3 UK Supreme Court (hereinafter UKSC) 22.01.2014, R (on the application of HS2 Action Alliance Ltd) v Secretary of State for Transport [2014] UKSC 3. 4 Paul Craig and Gráinne de Búrca, EU Law, Text, Cases and Materials, 7th edn (Oxford, OUP, 2020) ch 10.

United Kingdom  779 The communautaire approach is based on the ECJ’s reasoning in the seminal Costa decision,5 where it provided justificatory rationales for the supremacy of EU law over national law that were contractarian, functional, egalitarian and analytical. The contractarian argument was that EEC law should be accorded primacy because it flowed from the agreement made by the Member States when they joined the EEC through which they limited their sovereign rights. This was reinforced by a more functional argument, to the effect that the aims of the Treaty could not be achieved unless primacy was accorded to EEC law. A level playing field was integral to a common market and would be undermined if a Member State could choose not to accept an EU law, since it would thereby jeopardise the uniformity that was central to realisation of the EEC’s objectives. The egalitarian argument was related to, but distinct, from the functional. Ratification of a multilateral treaty is akin to joining a club, from which the recipient gains benefits and accepts constraints on autonomous individual action. This is the logic of collective action.6 Viewed from this perspective, if a Member State claims that its law should be accorded priority in the event of a conflict with EU law it is seeking unequal treatment in its favour, by continuing to take the benefits of membership while refusing to accept the attendant obligations. This was reinforced by an analytical argument that the obligations undertaken by the Member States in the Treaty would be merely contingent rather than unconditional if they were to be subject to later legislative acts by Member States that could trump EEC law. The reality, however, is that the great majority of Member States have chosen to base acceptance of supremacy on a provision of national constitutional law. Lord Bridge’s reasoning concerning supremacy in Factortame7 contained elements of both approaches. His Lordship acknowledged the functional argument by recognising that the supremacy of Community law could be regarded as inherent in the EEC Treaty, given its objectives. He noted also the contractarian line of reasoning, when he stated that such supremacy was well established when the United Kingdom joined the Community, and therefore the United Kingdom knew the terms of the agreement that it entered into, with the consequence that ‘whatever limitation of its sovereignty Parliament accepted when it enacted the European Communities Act 1972 was entirely voluntary’.8 These arguments cohered with those used by the ECJ and were reinforced by national legal foundations, the terms of the European Communities Act 1972, under which ‘it has always been clear that it was the duty of a United Kingdom court, when delivering final judgment, to override any rule of national law found to be in conflict with any directly enforceable rule of Community law’.9 It was, nonetheless, the domestic acceptance of supremacy that was emphasised in Thoburn.10 Laws LJ held that the constitutional relationship between the United Kingdom and the EU was not to be decided by the ECJ’s jurisprudence, which could not itself entrench EU law within national law. The constitutional relationship between the EU and the United Kingdom, including the impact of EU membership on sovereignty, was to be decided by the common law in the light of any domestic statutes. The approach in Thoburn was reinforced by the reasoning in HS2. The case arose as a challenge brought by action groups against government plans for a high-speed rail network, known 5 ECJ 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66. 6 Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge MA, Harvard University Press, 1965); James Buchanan and Gordon Tullock, The Calculus of Consent: Logical Foundations of Constitutional Democracy (Ann Arbor MI, University of Michigan Press, 1965). 7 UKHL, Factortame (No 2) (n 1) 658. 8 Ibid. 9 Ibid. 10 High Court of England and Wales (hereinafter EWHC), Queen’s Bench Division 18.02.2002, Thoburn v Sunderland City Council [2003] QB 151.

780  Paul Craig as HS2. The HS2 project was to be approved by Parliament through a hybrid bill procedure, which is the same as the ordinary public bill procedure, with the addition of a stage after the second reading whereby those most directly affected by the Bill would be able to voice their views through a special select committee hearing. The principles of the Bill would, however, be set at the second reading following debate, and these included the running of a high-speed rail line between London, Birmingham and the West Midlands. The established convention was that a select committee in a hybrid bill could not hear petitions that challenged the principle of the Bill, unless instructed to do so by the House at second reading, and the Government did not intend to allow this in the present case. The Government confirmed, moreover, that the Bill to authorise HS2 would be subject to the party whip system, whereby party members are instructed to vote in accordance with the wishes of the Government. The claimants argued, inter alia, that the hybrid bill procedure did not comply with the requirements of the Environmental Impact Assessment Directive,11 because the party whipping of the vote and the limited opportunity to examine the environmental information during the Parliamentary process prevented effective public participation as required by Article 6(4). This requirement did not, however, apply to projects that were exempted by Article 1(4) of the Directive, which provided that this ‘Directive shall not apply to projects the details of which are adopted by a specific act of national legislation, since the objectives of this Directive, including that of supplying information, are achieved through the legislative process’. The rationale for this exemption was that where a decision was reached by a legislative process, this itself entailed public participation, since the legislature was composed of democratically elected representatives of the public, thereby ensuring indirect, but nevertheless representative, public participation. The CJEU set two requirements that a legislative process must possess in order to comply with Article 1(4):12 it had to be substantive, and not merely formal, and information had to be available to those taking part in order that they could understand the relevant issues. The claimants contended that a decision by Parliament would be compatible with the EIA Directive only if MPs were allowed a free vote, regardless of their party allegiance, or of their membership of the Government. They also argued, in reliance on Opinions of Advocate General Sharpston and Kokott, that compliance with Article 1(4) required national courts to review the adequacy of the legislature’s consideration of the information, in order to assess the quality of the debate and the extent to which members participated in it. Thus Advocate General Sharpston had said that in assessing whether a legislative process was adequate, the national court should take into account, inter alia, whether the appropriate procedure was respected and whether the preparation time and discussion time was sufficient ‘for it to be plausible to conclude that the people’s elected representatives were able properly to examine and debate the proposed project’.13 In a similar vein, Advocate General Kokott stated that a national court should assess whether the national legislature ‘performs its democratic function correctly and effectively’, and whether the legislature ‘was able properly to examine and debate the environmental effects of the project’.14 The Supreme Court rejected the claimants’ argument and found in favour of the Government. It held that the hybrid bill procedure met the twin conditions established by the CJEU. It was a 11 Directive 2011/92/EU of the European Parliament and of the Council of 13 December 2011 on the assessment of the effects of certain public and private projects on the environment, [2012] OJ L 26/1–21. 12 CJEU 25.11.2011 C-128–131, 134–135/09 (Boxus and others v Région wallonne) ECLI:EU:C:2011:667, 37; CJEU 16.022012 C-182/10 (Solvay v Région wallonne) ECLI:EU:C:2012:82, 30–39; CJEU 11.09.2012 C-43/10 (Nomarchiaki Aftodioikisi Aitoloakarnanias and Others v Ypourgos Perivallontos, Chorotaxias kai Dimosion ergon) ECLI:EU:C:2012:560, 78–91. 13 CJEU, Cases C-128–131, 134–135/09 Boxus (n 12) 84. 14 CJEU, Case C-43/10 Nomarchiaki (n 12) 136 f.

United Kingdom  781 substantive legislative process, Parliament’s role was not merely formal and there was the possibility that the Bill would be amended. Parliament would receive information when debating the Bill, indeed there was a superabundance of information, such that most MPs would be able to digest only an executive summary. The Supreme Court held, moreover, that there was nothing in the CJEU’s case law to suggest that the influence of parliamentary parties, or Government, over voting in legislatures was incompatible with Article 1(4). This was more especially so, given that projects such as HS2 were matters of national political significance, and thus it was unsurprising that it should be considered appropriate to make such decisions through the legislative process, rather than by ordinary development control. The Supreme Court also rejected the suggestion that national courts were required to review the adequacy of the legislature’s consideration of the information, or that they should strike down legislation if the consideration of the information was found inadequate. It held that this requirement, which was derived from the Opinion of Advocate General Kokott and that of Advocate General Sharpston, had not been endorsed by the CJEU. If this were to be a requirement of EU law flowing from Directive 2011/92, the Supreme Court concluded that it would be constitutionally problematic for the United Kingdom. This was because Article 9 of the Bill of Rights 1689 precluded the impeaching or questioning in any court of debates or proceedings in Parliament. This limit had been reiterated in subsequent case law, which reinforced the precept that the courts would not look behind the parliamentary roll and question the nature of parliamentary proceedings.15 This was the foundation for much of the subsequent constitutional discussion in the Supreme Court, which raised issues concerning the legal relationship between the United Kingdom and the EU, sovereignty and constitutional instruments. The Supreme Court considered the extent, if any, to which the principles in Article 9 of the Bill of Rights concerning the limits of judicial intervention in the parliamentary process might be said to have been implicitly qualified, or even abrogated, by the European Communities Act 1972. The argument was that if the CJEU were to decide that national courts were obliged to scrutinise the parliamentary process in the manner indicated by Advocate General Sharpston and Advocate General Kokott, then the UK courts would be bound to accept this result. This was because the United Kingdom had accepted the supremacy of EU law, with the consequence that the European Communities Act 1972 would be regarded as qualifying the force of Article 9 of the Bill of Rights Act 1689. Lord Reed in HS2 rejected this argument in the following terms, and all other members of the Supreme Court agreed with his judgment.16 Contrary to the submission made on behalf of the appellants, that question cannot be resolved simply by applying the doctrine developed by the Court of Justice of the supremacy of EU law, since the application of that doctrine in our law itself depends upon the 1972 Act. If there is a conflict between a constitutional principle, such as that embodied in article 9 of the Bill of Rights, and EU law, that conflict has to be resolved by our courts as an issue arising under the constitutional law of the United Kingdom. Nor can the issue be resolved, as was also suggested, by following the decision in R v Secretary of State for Transport, Ex p Factortame Ltd (No 2) [1991] 1 AC 603, since that case was not concerned with the compatibility with EU law of the process by which legislation is enacted in Parliament.

Lord Reed found it unnecessary to develop this point further because he concluded that EU law did not require the kind of national oversight of the parliamentary process suggested by the 15 UKHL 22.03.1842, Edinburgh and Dalkeith Railway Co v Wauchope (1842) 8 Cl & F 710; UKHL 23.06.1871, Lee v Bude and Torrington Junction Railway Co (1871) LR 6 CP 576; UKHL 30.01.1974, Pickin v British Railways Board [1974] AC 765; UKHL 13.07.2003, Wilson v First County Trust Ltd (No 2) [2004] 1 AC 816. 16 UKSC, HS2 (n 3) 79. See also 203–05.

782  Paul Craig claimants to follow from the Opinions of Advocate General Sharpston and Advocate General Kokott.17 Lord Reed’s dictum reinforces the view of Laws LJ in Thoburn, that supremacy of EU law did not flow from the jurisprudence of the CJEU, but from its acceptance in and through the European Communities Act 1972. The corollary was that any conflict between the 1972 Act and another constitutional principle would be resolved by UK courts as a matter of UK constitutional law, even where the reason for the clash flowed from EU law. Factortame was, moreover, held not to be determinative of the relationship between EU law and national law where the case concerned the compatibility with EU law of the process by which legislation was enacted in the United Kingdom. It can be accepted that the foundation for EU law in the United Kingdom resided in the European Communities Act 1972. It can be accepted also that it was ultimately a matter of UK constitutional law to determine the impact of EU membership on sovereignty in the light of existing domestic legislation. It followed that any conflict between the European Communities Act 1972 and other statutes such as the Bill of Rights Act 1689 was to be resolved as a matter of UK constitutional law by the UK courts. While this can be accepted, it did not therefore mean that in determining the impact of EU membership on UK sovereignty, the arguments used by the CJEU were, or should have been, irrelevant. The contractarian, egalitarian, functional and analytical arguments could properly be regarded as having force, when deciding as a matter of UK constitutional law on the effect of EU membership on sovereignty. The fact that such arguments could properly be taken into account did not compel the conclusion that the UK courts should reach the same conclusion as the CJEU, to the effect that any norm of EU law was supreme over any norm of national law. The great majority of national courts place some qualifications on the extent to which the supremacy of EU law will be accepted within their legal system. This does not, however, alter the preceding point, which is that the normative arguments used by the CJEU could properly have been regarded as relevant when determining the outcome of a conflict between EU and national law. They could have been regarded as having normative force in terms of their ability to persuade, as opposed to being accepted merely by virtue of their authority as stemming from the CJEU. While the Supreme Court, as seen above, emphasised that it was for UK constitutional law to determine the relationship between the United Kingdom and the EU, it was also willing to recognise the novelty and force of EU law in the national legal order. Thus in Miller,18 the Supreme Court held that while statute gave effect to treaties, by prescribing the content of domestic law in the areas covered by them, the European Communities Act 1972 went considerably further by authorising a dynamic process through which, without further UK legislation, EU law became a source of UK law, and took precedence over domestic UK law, including statutes. While the ECA 1972 was the formal basis for EU law in the United Kingdom, the substantive reality was that ‘where EU law applies in the United Kingdom, it is the EU institutions which are the relevant source of that law’,19 since they made the rules that the ECA then incorporated into UK law. Thus, so long as the ECA 1972 remained in force, the EU Treaties, EU legislation and the rulings of the CJEU were ‘direct sources of UK law’;20 they constituted ‘an independent and overriding source of domestic law’.21

17 Ibid

109 f. 24.01.2017, R (on the application of Miller) v Secretary of State for Exiting the European Union [2017] UKSC 5. 19 Ibid 61. 20 Ibid. 21 Ibid 65. 18 UKSC

United Kingdom  783

IV.  Constitutional Limits on EU Membership A.  The European Union Act 2011 Prior to Brexit, the principal statutory limit to EU integration was the European Union Act 2011, which would have been regarded as a constitutional statute in accordance with the interpretation of that concept set out above, and considered in more detail below. The statute was complex, and imposed a regime of parliamentary and referendum ‘locks’ on Treaty amendments and a range of other EU decisions. A brief outline of the principal provisions can be given here. Section 2 dealt with Treaty amendment pursuant to the ordinary revision procedure in Article 48(2)–(5) TEU. It specified that a statement must be laid before Parliament in accordance with section 5, the Treaty amendment must be approved by an Act of Parliament and the referendum condition or exemption condition must be met. The referendum condition required, as specified in section 2(2), that the Act of Parliament approving the Treaty amendment could not come into force until there had been a positive vote in a national referendum. The only qualification was where the exemption condition applied. This obviated the need to hold a referendum if the Treaty amendment did not fall within section 4, which defined the notion of Treaty change. Section 4 was, however, very broad, and it would have been rare for the obligation to hold a referendum to be negated by the exemption condition. The House of Lords proposed an amendment to the referendum condition, such that a positive vote in a referendum was not per se dispositive if the voter turnout was less than 40 per cent. This amendment was, however, rejected by the House of Commons.22 Section 3 dealt with Treaty reform pursuant to the simplified revision procedure in Article 48(6) TEU. It provided that where the European Council adopted a decision pursuant to Article 48(6) TEU, a Minister of the Crown could not confirm UK approval of the decision unless a statement has been laid before Parliament in accordance with section 5, the decision had been approved by statute, and the referendum condition, exemption condition or significance condition had been met. The only difference between sections 2 and 3 was, therefore, that the requirement to hold a referendum was obviated in section 3 not only where the exemption condition applied, but also where the significance condition in section 3(4) was applicable. This made a referendum unnecessary if the change only came within section 4(1)(i) or (j),23 and such change was not significant for the United Kingdom. Section 6(1) specified further instances where confirmation by statute and referendum was required. It stated that a minister of the Crown could not vote in favour of a decision to which the section applied, unless the draft decision had been approved by Act of Parliament and the referendum condition was met. Section 6(5) defined the terrain to which section 6(1) applied. It was concerned primarily with changes to Treaty voting rules and legislative procedure pursuant to passerelle provisions. Thus, for example, any EU draft decision made pursuant to the general passerelle clause in Article 48(7) TEU, whereby there can be a shift from unanimity to qualified majority voting, or from the special to the ordinary legislative procedure, had to be approved by statute and a referendum in relation to any provision listed in Schedule 1 of the Act,24 and so too 22 House of Lords Select Committee on the Constitution’s Thirteenth Report of Session 2010–11: European Union Bill, Government Response to the Report, www.parliament.uk/documents/lords-committees/constitution/Government Response/GvtResEUBillReport.pdf. 23 European Union Act 2011, s 4(1)(i) the conferring on an EU institution or body of power to impose a requirement or obligation on the United Kingdom, or the removal of any limitation on any such power of an EU institution or body; (j) the conferring on an EU institution or body of new or extended power to impose sanctions on the United Kingdom. 24 Ibid, s 6(5)(b).

784  Paul Craig had draft changes made pursuant to more particular passerelle provisions.25 Section 6(5) also mandated the need for a statute and referendum for other EU decisions, which are not passerelle provisions. The House of Lords proposed an amendment, such that a referendum would not be required for most of the subject matter covered by sections 6(1) and 6(5), the rationale being that it was not sufficiently important to warrant a referendum. The House of Lords Select Committee on the Constitution took the view that the requirement of a referendum in over 50 policy areas constituted ‘a radical step change’,26 because most such areas did not concern fundamental constitutional issues.27 This amendment was rejected by the House of Commons. The Government argued that the Act provided clarity as to when a referendum was required,28 and that a referendum was warranted for the issues in section 6(5), because they could be just as significant as changes made by sections 2 and 3.29 The European Union Act 2011, section 7, stipulated instances in which EU decisions required statutory approval, although not a referendum. Section 7(1) provided that a minister could not confirm approval of certain EU decisions, unless they were approved by Act of Parliament. These were:30 a decision under Article 25 TFEU that permitted adoption of provisions to strengthen or add to the rights of EU citizens in Article 20(2) TFEU; a decision under Article 223(1) TFEU that permitted the laying down of provisions necessary for the election of MEPs in accordance with that Article; a decision under Article 262 TFEU that permitted conferral of jurisdiction on the CJEU in disputes relating to the application of acts adopted on the basis of the EU Treaties which created European intellectual property rights; and a decision under the third paragraph of Article 311 TFEU to adopt a decision laying down provisions relating to the system of EU own resources. Section 7(3) further provided that a Minister of the Crown could not vote in favour of, or otherwise support, a long list of decisions unless the relevant draft decision was approved by Act of Parliament.31 The United Kingdom’s zeal for control over EU decisions was further apparent in section 8, which was concerned with Article 352 TFEU, the EU’s flexibility clause. A Minister of the Crown could not vote in favour of, or otherwise support, an Article 352 decision unless one of the following mechanisms was complied with in relation to the draft decision. The first option was for the draft decision to be approved by Act of Parliament.32 The second option was that the Minister of the Crown moved a motion in each House of Parliament that the House approved the Government’s intention to support a specified Article 352 draft decision, because the measure to which it related was required as a matter of urgency, and each House agreed to the motion without amendment.33 The third option was for the Minister of the Crown to lay before Parliament a statement specifying a draft decision and stating that in the Minister’s opinion the decision relates only to one or more exempt purposes.34 The European Union Act 2011, section 9, contained requirements that had to be satisfied in relation to the Area of Freedom, Security and Justice, AFSJ. Protocol (No. 21) of the Lisbon 25 Ibid, s 6(5)(a), (f)–(j). 26 House of Lords Select Committee on the Constitution’s Thirteenth Report of Session 2010–11: European Union Bill (HL 121, 2011) 37. 27 Ibid 38. 28 Government Response (n 22) 12–25. 29 European Union Act 2011, Explanatory Notes, at 71. 30 European Union Act 2011, s 7(2). 31 Ibid, s 7(4). 32 Ibid, s 8(3). 33 Ibid, s 8(4). 34 Ibid, s 8(6).

United Kingdom  785 Treaty provided an opt-out for the United Kingdom and Ireland in relation to AFSJ measures. Article 3 of this Protocol, however, stated that within three months of any proposal for an AFSJ measure having been presented to the Council, the United Kingdom and Ireland could notify their intent to take part in such a measure. It was this notification that was controlled by the 2011 Act: a Minister of the Crown could not give a notification to which section 9(1) applied unless Parliamentary approval had been given in accordance with section 9(3). The requirements for Parliamentary approval were set out in section 9(3). It provided that the Minister of the Crown must move a motion in each House of Parliament that the House approved the Government’s intention to give a notification in respect of a specified measure, and each House agreed to the motion without amendment. Section 10(1) mandated Parliamentary approval for certain EU decisions. It stated that a Minister of the Crown could not vote in favour of, or otherwise support, such decisions unless Parliamentary approval had been given. The list of EU decisions subject to this requirement covered important issues, such as decisions: amending the Statute of the Court of Justice; creating specialised courts; amending the Statute of the ECB; and increasing the number of AdvocatesGeneral. Section 10(4) required, moreover, that a Minister of the Crown could not confirm the approval by the United Kingdom of a decision under Article 218(8) TFEU for the accession of the EU to the ECHR unless Parliamentary approval had been given. Section 4 set out when a referendum was required pursuant to sections 2, 3 and 6. The drafting was exhaustive, in the sense that it was designed to catch every conceivable form of Treaty change. (1) Subject to subsection (4), a treaty or an Article 48(6) decision falls within this section if it involves one or more of the following […] (a) (b) (c) (d) (e) (f) (i) (ii) (g) (h) (i) (j) (k) (l)

(m)

the extension of the objectives of the EU as set out in Article 3 of TEU; the conferring on the EU of a new exclusive competence; the extension of an exclusive competence of the EU; the conferring on the EU of a new competence shared with the member States; the extension of any competence of the EU that is shared with the member States; the extension of the competence of the EU in relation to the co-ordination of economic and employment policies, or common foreign and security policy; the conferring on the EU of a new competence to carry out actions to support, co-ordinate or supplement the actions of member States; the extension of a supporting, co-ordinating or supplementing competence of the EU; the conferring on an EU institution or body of power to impose a requirement or obligation on the United Kingdom, or the removal of any limitation on any such power of an EU institution or body; the conferring on an EU institution or body of new or extended power to impose sanctions on the United Kingdom; any amendment of a provision listed in Schedule 1 that removes a requirement that anything should be done unanimously, by consensus or by common accord; any amendment of Article 31(2) of TEU (decisions relating to common foreign and security policy to which qualified majority voting applies) that removes or amends the provision enabling a member of the Council to oppose the adoption of a decision to be taken by qualified majority voting; any amendment of any of the provisions specified in subsection (3) that removes or amends the provision enabling a member of the Council, in relation to a draft legislative act, to ensure the suspension of the ordinary legislative procedure.

(2) Any reference in subsection (1) to the extension of a competence includes a reference to the removal of a limitation on a competence.

786  Paul Craig (3) The provisions referred to in subsection (1)(m) are (a) Article 48 of TFEU (social security), (b) Article 82(3) of TFEU (judicial co-operation in criminal matters), and (c) Article 83(3) of TFEU (particularly serious crime with a cross-border dimension). (4) A treaty or Article 48(6) decision does not fall within this section merely because it involves one or more of the following (a) the codification of practice under TEU or TFEU in relation to the previous exercise of an existing competence; (b) the making of any provision that applies only to member States other than the United Kingdom; (c) in the case of a treaty, the accession of a new member State.

The substantive criteria for holding a referendum were reinforced by procedural duties. Section 5(1) specified that if a Treaty amending the TEU or TFEU was agreed in an Intergovernmental Conference, a Minister of the Crown had to lay the required statement before Parliament within two months. Section 5(2) contained the same obligation in relation to a Treaty change made by the simplified revision procedure. The ‘required statement’ as specified by section 5(3) was a statement as to whether the Treaty change made by the ordinary or simplified revision procedure fell within section 4 of the Act. The Minister had to inform Parliament whether the Treaty change or Article 48(6) decision fell within section 4, and reasons had to be given for this statement. Section 5(4) amplified the meaning of the significance condition. If the Minister was of the opinion that an Article 48(6) TEU decision fell within section 4 only because of provision of the kind covered by section 4(1)(i) or (j), the statement had to indicate whether in the Minister’s opinion the effect of that provision in relation to the United Kingdom was significant.

B.  Constitutional Statutes Prior to the decision in HS2 the general orthodoxy was that Parliamentary sovereignty meant that Parliament had unlimited power, subject to one exception, it could not bind its successors, either as to substance, or as to the manner and form in which legislation was enacted.35 The latter proposition was contested by those who argued that manner and form provisions could bind a later Parliament, which could amend or repeal any earlier enactment but only in accord with the procedure for doing so laid down in the earlier enactment.36 Subject to this, it was generally accepted that Parliament could not only repeal or amend earlier legislation expressly, but also impliedly and that this would occur if later legislation was inconsistent with that enacted previously. There was, however, some authority for recognition of a category of constitutional statutes, which would have implications for the possibility of implied repeal or amendment. Thus in Thoburn,37 Laws LJ held that ordinary statutes were subject to the doctrine of implied repeal, but that ‘constitutional statutes’, which conditioned the legal relationship between citizen and state

35 William Wade, ‘The Basis of Legal Sovereignty’, (1955) 13 Cambridge Law Journal 172. 36 Ivor Jennings, The Law and the Constitution, 5th edn (London, University of London Press, 1959) ch 4; Robert Heuston, Essays in Constitutional Law, 2nd edn (London, Stevens, 1964) ch 1; Geoffrey Marshall, Constitutional Theory (Oxford, OUP, 1971) ch 3. 37 EWHC, Thoburn (n 10). See also, England and Wales Court of Appeal, Civil Division 05.03.2003, McWhirter v Secretary of State for Foreign and Commonwealth Affairs [2003] EWCA Civ 384 at 14; Lady Mary Arden, ‘Jurisdiction of the New United Kingdom Supreme Court’, (2004) Public Law 699.

United Kingdom  787 in some overarching manner, or which dealt with fundamental constitutional rights, were not subject to the doctrine of implied repeal, at least not in the way that it applied as between ordinary statutes.38 The repeal of such a statute, or its disapplication in a particular instance, could only occur if there were some ‘express words in the later statute, or by words so specific that the inference of an actual determination to effect the result contended for was irresistible’.39 The ECA 1972 was regarded as just such a constitutional statute. It contained provisions that ensured the supremacy of substantive EU law in the event of a clash with national law, and was not subject to implied repeal. Commentators were divided as to the impact of the case, and as to the significance of Laws LJ’s recognition of constitutional statutes.40 The existence of such a category was however, recognised, by Lord Hope in H v Lord Advocate41 and was approved, as noted earlier, by the Supreme Court in HS2. Lord Neuberger and Lord Mance echoed the disquiet felt by Lord Reed as to the possible impact on Article 9 of the Bill of Rights 1689 if EU law were held to require the kind of scrutiny of the legislative process that appeared to be suggested by Advocate General Sharpston and Advocate General Kokott.42 They accepted that pursuant to the European Communities Act 1972, UK courts acknowledged that EU law required them to treat domestic statutes, whether passed before or after the 1972 Act, as invalid if they could not be interpreted consistently with EU law.43 Lord Neuberger and Lord Mance, nonetheless, held that it would be difficult for a UK court to comply with the approach suggested by the Advocates General ‘without addressing its apparent conflict with other principles hitherto also regarded as fundamental and enshrined in the Bill of Rights’,44 since it could entail questioning and potentially impeaching Parliament’s internal proceedings, in a manner that a UK court had not hitherto done. Lord Neuberger and Lord Mance then stated,45 The United Kingdom has no written constitution, but we have a number of constitutional instruments. They include Magna Carta, the Petition of Right 1628, the Bill of Rights and (in Scotland) the Claim of Rights Act 1689, the Act of Settlement 1701 and the Act of Union 1707. The European Communities Act 1972, the Human Rights Act 1998 and the Constitutional Reform Act 2005 may now be added to this list. The common law itself also recognises certain principles as fundamental to the rule of law. It is, putting the point at its lowest, certainly arguable (and it is for United Kingdom law and courts to determine) that there may be fundamental principles, whether contained in other constitutional instruments or recognised at common law, of which Parliament when it enacted the European Communities Act 1972 did not either contemplate or authorise the abrogation.

Their Lordships did not decide unequivocally whether Article 9 would count as such a constitutional instrument, but said that the issue was too important to be decided without full argument.46 They acknowledged also the ‘penetrating discussion’ of the general issues concerning constitutional statutes by Laws LJ in Thoburn, while recognising that the HS2 case had an added dimension, given that the possible conflict was between two constitutional statutes, rather than between an ordinary statute and a constitutional statute as in Thoburn.47 38 EWHC, Thoburn (n 10) 62. 39 Ibid 63. 40 Adam Tomkins, Public Law (Oxford, OUP, 2003) 124; Geoffrey Lindell, ‘The Statutory Protection of Rights and Parliamentary Sovereignty: Guidance from the United Kingdom?’, (2006) 17 Public Law Review 188; Aileen Kavanagh, ‘Constitutional Review, the Courts and Democratic Scepticism’, (2009) 62 Current Legal Problems 293. 41 UKSC 20.12.2012, H v Lord Advocate [2012] UKSC 24. 42 UKSC, HS2 (n 3) 203 f. 43 Ibid 206, citing UKHL, Factortame (No 2) (n 1). 44 Ibid 206. 45 Ibid 207. 46 Ibid 208. 47 Ibid.

788  Paul Craig The Supreme Court’s reasoning concerning constitutional instruments is technically an obiter dictum, given that it held that there was no actual conflict with the Bill of Rights 1689, because the CJEU’s case law was interpreted so as not to demand the scrutiny of the legislative process that seemed to be required by Advocate General Sharpston and Advocate General Kokott. The Supreme Court, nonetheless, expressed an authoritative view to which all seven members subscribed, and it was therefore authoritative for UK domestic courts. It followed that a constitutional constraint on the acceptance of the supremacy of EU law in UK courts was that the ECA 1972 would not be interpreted so as to give precedence to an EU law that clashed with a UK constitutional statute. It would not be assumed that the 1972 statute contemplated or authorised the abrogation or limitation of such principles.

V.  Implementing EU Law: UK Constitutional Rules/Practice The constitutional framework for the implementation of EU law in the United Kingdom was established in the statute that provided the foundation for the United Kingdom’s entry into the EU, the European Communities Act 1972, section 2(1) of which provided: All such rights, powers, liabilities, obligations and restrictions from time to time created or arising by or under the Treaties, and all such remedies and procedures from time to time provided for by or under the Treaties, as in accordance with the Treaties are without further enactment to be given legal effect or used in the United Kingdom shall be recognized and available in law, and be enforced, allowed and followed accordingly; and the expression ‘enforceable Community right’ and similar expressions shall be read as referring to one to which this subsection applies.

Section 2(1) therefore made the concept of direct effect part of the UK legal system. It deemed law which under the EU Treaties was to be given immediate legal effect to be directly enforceable in the United Kingdom. Thus, UK courts were directed by section 2(1) to enforce any directly effective EU measures. There was no need for a fresh act of incorporation to enable UK courts to enforce each EU Treaty provision, regulation, or directive that has direct effect under EU law. Section 2(2) provided for the implementation of EU obligations, even when they were intended to replace national legislation and Acts of Parliament, by means of an Order in Council or statutory instrument rather than by primary legislation only. Section 2(4) then provided: The provision that may be made under subsection (2) above includes, subject to Schedule 2 to this Act, any such provision (of any such extent) as might be made by Act of Parliament, and any enactment passed or to be passed, other than one contained in this Part of this Act, shall be construed and have effect subject to the foregoing provisions of this section; …

The Schedule to which the provision referred set out a number of powers, such as increasing taxation or legislating retroactively, which could not be exercised by Order in Council or by delegated legislation, even if they were necessary to comply with an EU obligation. For these powers an Act of Parliament was needed. However, the part of section 2(4) which received most attention was the clause beginning ‘any enactment passed or to be passed’, which became prominent when the courts sought a way to reconcile new obligations under EU law with the traditional approach to statutory interpretation. Section 3 of the Act made CJEU decisions on EU law authoritative in UK courts. For the purposes of all legal proceedings any question as to the meaning or effect of any of the Treaties, or as to the validity, meaning or effect of any Community instrument, shall be treated as a question of law (and, if not referred to the European Court, be for determination as such in accordance with the principles laid down by and any relevant decision of the European Court or any court attached thereto).

United Kingdom  789 The preceding statutory provisions must be seen in the light of the European Union Act 2011, section 18. It provided as follows. Directly applicable or directly effective EU law (that is, the rights, powers, liabilities, obligations, restrictions, remedies and procedures referred to in section 2(1) of the European Communities Act 1972) falls to be recognised and available in law in the United Kingdom only by virtue of that Act or where it is required to be recognised and available in law by virtue of any other Act.

The rationale for section 18 was apparent from the Explanatory Notes attached to the statute, which made it clear that the major concern underlying section 18 was parliamentary sovereignty.48 This declaratory provision was included in the Act in order to address concerns that the doctrine of parliamentary sovereignty may in the future be eroded by decisions of the courts. By providing in statute that directly effective and directly applicable EU law only takes effect in the UK legal order through the will of Parliament and by virtue of the European Communities Act 1972 or where it is required to be recognised and available in law by virtue of any other Act, this will provide clear authority which can be relied upon to counter arguments that EU law constitutes a new higher autonomous legal order derived from the EU Treaties or international law and principles which has become an integral part of the UK’s legal system independent of statute.

It is important to be clear in what sense section 18 could be regarded as a ‘sovereignty clause’. It is necessary to distinguish between ‘sovereignty as dualism’ and ‘sovereignty as primacy’. Section 18 was concerned with ‘sovereignty as dualism’. It embodied in statutory form the common law concept of dualism. Section 18 stated that EU law fell to be recognised and available in the United Kingdom by virtue of an Act of Parliament. This supported the sovereignty of Parliament by making clear that executive affirmation of a Treaty would not have domestic legal effect without parliamentary ratification in a statute. This was reflected in the heading of section 18, which was ‘status of EU law dependent on continuing statutory basis’. This reading of section 18 was supported by the Explanatory Notes attached to the Act.49 The legal premise was that in a dualist country, such as the United Kingdom, there had to be an Act of Parliament that adopted or transformed the EU Treaty into UK law. Viewed from this perspective there was nothing novel about section 18. The European Communities Act 1972, and in particular section 2(1), was the gateway for EU law becoming part of UK law. There was nothing in the prior case law that undermined sovereignty in the sense of dualism. The UK cases concerning EU law and national law were predicated on dualism. The case law of the EU courts did not attack the idea that the relationship between UK law and EU law could be premised on dualism. Nor did the principle of directly applicable EU law offend sovereignty as dualism. In the European Communities Act 1972, the United Kingdom expressly agreed to the Treaties, including the idea that regulations were directly applicable. Thus, insofar as rules of EU law could have effect without the foundation of a particular statute, this was because the EU Treaty and case law thereunder affirmed that this was so, and the United Kingdom agreed to this regime when joining the EEC via the European Communities Act 1972. This was readily apparent from the express wording of sections 2(1) and 3(1) of the 1972 Act, set out above. To be sure directives require implementation and this would normally have been through UK statute or delegated legislation, but this was consonant with the very nature of directives under EU law. The United Kingdom did not constitutionally require separate national legislation as a pre-condition for the legal validity of each directly applicable EU regulation within UK law, and such a requirement would have placed the United Kingdom in systematic breach of EU law.

48 European 49 Ibid

119.

Union Act 2011, Explanatory Notes, 120.

790  Paul Craig Section 18 was not concerned with ‘sovereignty as primacy’. It told us nothing about the relation between EU law and national law in the event of a clash between the two. Indeed the Explanatory Notes stated that section 18 was not intended to affect the primacy of EU law.50 The EU courts have always taken the view that all EU law has primacy over all national law. This has not been generally accepted by national courts, which have resisted the idea that EU law takes precedence over the national constitution and/or fundamental rights. The UK courts in Factortame51 and the EOC case52 accepted that EU law could, in general, have primacy over national law in the event of a clash. Detailed discussion of this case law can be found elsewhere.53 The relevant point is that section 18 did not address this second sense of sovereignty. The preceding discussion prompts the obvious inquiry as to why section 18 was included in the 2011 Act. The previous quotation from the Explanatory Notes provides guidance in this respect. It stated that inclusion of section 18 would ‘counter arguments that EU law constitutes a new higher autonomous legal order derived from the EU Treaties or international law and principles which has become an integral part of the UK’s legal system independent of statute’.54 This concern was fuelled by argument of counsel in the Thoburn case,55 that ‘in effect, the law of the EU includes the entrenchment of its own supremacy as an autonomous legal order, and the prohibition of its abrogation by the Member States’.56 This argument was rejected by Laws LJ in Thoburn, who held that the constitutional foundation for acceptance of EU law in the United Kingdom was based on UK constitutional principle, and this was, as seen above, affirmed by the Supreme Court in the HS2 case.57 It is in any event doubtful whether EU law claims the effect argued for by counsel in the case. This begs the further inquiry as to when section 18 would have been relevant. There were only two scenarios in which this might have occurred, and both were unlikely. The first would be where the United Kingdom expressly repealed the European Communities Act 1972, with the intent of leaving the EU, but had not yet exited from the EU pursuant to Article 50 TEU. On this scenario there would be no Act of Parliament through which EU law fell to be recognised in the United Kingdom. However, until the United Kingdom left the EU it might be contended that EU law could continue to apply in the United Kingdom as an autonomous legal order even in the absence of a domestic statute. Section 18 of the European Union Act 2011 would block this line of reasoning in the UK courts. It should, nonetheless, be recognised that the United Kingdom would still have remained bound by EU law as an international Treaty obligation, and it would have been in breach of its obligations if it had failed to comply with EU law during the period after the repeal of the European Communities Act 1972, and before the United Kingdom’s withdrawal took effect in accord with the TEU. Article 50 TEU stipulates that while a state may decide to withdraw from the EU, the details of the withdrawal are to be negotiated between the state and the EU. Article 50(3) TEU provides that the Treaties only cease to apply to that Member State from the date of entry into force of the withdrawal agreement, or in the event 50 Ibid 123 f. 51 UKHL, Factortame (No 2) (n 1). 52 UKHL 03.03.1994, R v Secretary of State for Employment, ex parte Equal Opportunities Commission [1995] 1 AC 1. 53 Paul Craig, ‘Parliamentary Sovereignty of the United Kingdom Parliament after Factortame’, (1991) 11 Yearbook of European Law 221; William Wade, ‘Sovereignty – Revolution or Evolution?’, (1991) 112 Law Quarterly Review 568; Trevor Allan, ‘Parliamentary Sovereignty: Law, Politics and Revolution’, (1997) 113 Law Quarterly Review 443; Paul Craig, ‘Britain in the European Union’, in J Jowell, D Oliver and C O’Cinneide (eds), The Changing Constitution, 8th edn (Oxford, OUP, 2015) ch 4. 54 European Union Act 2011, Explanatory Notes, 120. 55 EWHC, Thoburn (n 10). 56 European Union Act 2011, Explanatory Notes, 121. 57 UKSC, HS2 (n 3) 3.

United Kingdom  791 that an agreement cannot be reached, two years after the state has notified its intent to withdraw from the EU, unless this period is extended by the European Council. The alternative scenario was one in which the United Kingdom remained in the EU, but a UK statute expressly derogated from a provision of EU law, and unequivocally excluded application of the European Communities Act 1972. If this occurred, it might be argued that there would to that extent be no Act of Parliament by virtue of which EU law was recognised and available on the relevant topic in the United Kingdom. Section 18 could then have been used to counter any argument that the provision of EU law could apply in the United Kingdom because EU law constituted an autonomous legal order. Needless to say, such an Act of Parliament would constitute breach of EU law. It would have been for the UK courts to determine whether such an Act of Parliament was compatible with the United Kingdom’s continued membership of the EU. The Supreme Court might have decided that the UK derogating statute could be given effect pursuant to the traditional theory of parliamentary sovereignty, plus section 18. It might alternatively have decided that such a statute could not be made while the United Kingdom remained within the EU. While the United Kingdom remained in the EU, such a statute was not, and should not have been enacted, in the absence of some very serious and well-founded concern about the impact of EU law on national constitutional precepts/fundamental rights, and such a step should only have been contemplated after according the EU courts the opportunity to rule on the issue via a preliminary ruling. EU membership brings benefits and burdens. If a Member State could derogate from EU law simply because the Member State disliked the outcome, this would entail inequality vis-à-vis other Member States, the denial of a level playing field, and undermine the Union. EU membership entails a diminution of the capacity for autonomous state action, but this is true of any form of collective action. If a Member State does not wish to accept the burdens of membership it should not be able to take the benefits. Lord Bridge in Factortame58 correctly noted that insofar as there has been a diminution of sovereignty flowing from EU membership this was not the result of judicial decisions, but was rather the consequence of the political decision to join.

VI.  UK and EU Law: Resulting Relationship The nature of the relationship between the United Kingdom and EU might well have been contestable. Much depends on how one interprets the preceding material and the story that one chooses to tell based on that ‘data’. A plausible interpretation is as follows. For much of the period of UK membership of the EU, it caused far fewer problems than many other Member States. It is customary to think of the United Kingdom as having been a problematic member of the EU, and there is justification for this in terms of the political relationship between the United Kingdom and EU. The picture is, however, markedly different when one considers the relationship from a constitutional and legal perspective. The UK courts caused less difficulty than those of other Member States. They accepted the political fact of membership and strove to ensure that national constitutional doctrine could accommodate that political choice. So too the UK administration, which was generally towards the top of the legal table in terms of putting into effect EU norms within the national legal order, thereby reflecting the professional ethos that political obligations that translated into hard law should be faithfully implemented.

58 UKHL,

Factortame (No 2) (n 1) 658 f.

792  Paul Craig There is much of this that continued to hold true prior to Brexit. There was, nonetheless, a shift evident. The European Union Act 2011 was indicative of the political desire to maintain strict national controls on the grant of any new power to the EU. The ruling in HS2 was indicative of the legal desire to emphasise the domestic constitutional foundations of UK membership of the EU, the corollary being that it was for the national courts to determine the limits to the impact of EU law within our own legal order. There was, finally, the political fact of Brexit, which brings us full circle from the beginning of this chapter. The United Kingdom has now exited from the EU, following the referendum held on 23 June 2016. The UK courts and UK constitutional law were intimately involved in the Brexit process. There was constitutional litigation in the first Miller decision,59 which held that the executive could not trigger the exit process under Article 50 TEU through recourse to the prerogative, and that statutory authorisation from Parliament was a condition precedent to start the exit process. Parliament duly authorised the executive to make the notification under Article 50 TEU, without any conditions being attached. The House of Commons also desisted from imposing hard-edged constraints on the executive, of the kind that had been suggested by the House of Lords, relating to the terms of the withdrawal agreement, or as to whether we leave without an agreement.60 The Supreme Court was involved yet again in Miller II, in which it held that the prerogative concerning prorogation of Parliament was subject to judicial review.61 The paradox and tragedy of Brexit is that while Brexiteers made much of the sovereignty of Parliament, and respecting the will of the people, this was conditional on this leading to the outcome that they desired. They resisted giving Parliament a real say in relation to the final deal reached with the EU, or as to whether we might leave with no deal, and they resisted, with the same fervour, any suggestion of a second referendum, precisely because they were worried that the outcome would not be in accord with what they desired.

References T Allan, ‘Parliamentary Sovereignty: Law, Politics and Revolution’, (1997) 113 Law Quarterly Review 443. Lady M Arden, ‘Jurisdiction of the New United Kingdom Supreme Court’, (2004) Public Law 699. J Buchanan and G Tullock, The Calculus of Consent: Logical Foundations of Constitutional Democracy (Ann Arbor MI, University of Michigan Press, 1965). P Craig, ‘Parliamentary Sovereignty of the United Kingdom Parliament after Factortame’, (1991) 11 Yearbook of European Law 221. P Craig, ‘Britain in the European Union’, in J Jowell, D Oliver and C O’Cinneide (eds), The Changing Constitution, 8th edn (Oxford, OUP, 2015). P Craig, ‘Miller, Structural Constitutional Review and the Limits of Prerogative Power’, (2017) Public Law Special Issue Brexit, 48. P Craig and G de Búrca, EU Law, Text, Cases and Materials, 7th edn (Oxford, OUP, 2020). R Heuston, Essays in Constitutional Law, 2nd edn (London, Stevens, 1964). I Jennings, The Law and the Constitution, 5th edn (London, University of London Press, 1959). A Kavanagh, ‘Constitutional Review, the Courts and Democratic Scepticism’, (2009) 62 Current Legal Problems 293.

59 UKSC, Miller (n 18); Paul Craig, ‘Miller, Structural Constitutional Review and the Limits of Prerogative Power’, (2017) Public Law, Special Issue Brexit, 48. 60 European Union (Withdrawal) Act 2018, s 13. 61 R (on the application of Miller) v Prime Minister, Cherry and others v Advocate General for Scotland [2019] UKSC 41.

United Kingdom  793 G Lindell, ‘The Statutory Protection of Rights and Parliamentary Sovereignty: Guidance from the United Kingdom?’, 17 (2006) Public Law Review 188. G Marshall, Constitutional Theory (Oxford, OUP, 1971). D Nicol, EC Membership and the Judicialization of British Politics (Oxford, OUP, 2001). M Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge MA, Harvard University Press, 1965). A Tomkins, Public Law (Oxford, OUP, 2003). W Wade, ‘The Basis of Legal Sovereignty’, (1955) 13 Cambridge Law Journal 172. W Wade, ‘Sovereignty – Revolution or Evolution?’, (1991) 112 Law Quarterly Review 568.

794

29 Squaring the Circle – How National Constitutions and EU Law Interact STEFAN GRILLER AND LINA PAPADOPOULOU

I.  Introductory Remarks ‘National Constitutions and EU Integration’ addresses at least three legal challenges. These are, first, identifying and analysing the national constitutional foundations for participating in the European integration – with the European Union (EU) as the focus of attention; secondly, determining the significance of those national constitutional rules from the angle of the ‘EU Constitution’, ie the constitutional dimension of the founding EU instruments; thirdly, exploring the relationship between the two types of legal orders – between 27 national, on the one hand, and one supranational, on the other, including their interaction and the resolution of conflicts that arise between them. Since the beginning of the EU integration in the 1950s, these challenges have been part of the core issues of European constitutional development and scholarly debate.1 Until recently, the third challenge, and especially the resolution of conflicts, seemed to be mostly academic in nature. This was because primacy, often also called ‘supremacy’, of EU law finally always prevails and dealing with (possible) conflicts could appear as micromanaging the fundamentally benevolent dialogue between the supreme national/Constitutional Courts and the Court of Justice of the European Union (CJEU), and the cooperation between them. 1 From the flood of literature on the subject compare only (all with further references) Anneli Albi and Samo Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (The Hague, T.M.C. Asser Press, 2019); Monica Claes, Maartje de Visser, Patricia Popelier and Catherine Van de Heyning (eds), Constitutional Conversations in Europe (Cambridge/Antwerp/Portland, Intersentia, 2012); European Parliament, Policy Department C: Citizens’ Rights and Constitutional Affairs (ed), National Constitutional Law and European Integration, Study PE 432.750 (Brussels, European Parliament, 2011) 11–31, 196–226; European Parliament, Policy Department C: Citizens’ Rights and Constitutional Affairs (ed), National Constitutional Avenues for further EU Integration, Study PE 493.046, (Brussels, European Parliament, 2014); Christoph Grabenwarter, ‘National Constitutional Law Relating to the European Union’, in A von Bogdandy and J Bast (eds), Principles of European Constitutional Law, 2nd edn, (Oxford/ Munich, Hart Publishing/C.H. Beck, 2010) 83; Stefan Griller, Franz Maislinger and Andreas Reindl, Fundamentale Rechtsgrundlagen einer EG-Mitgliedschaft. Verfassungsfragen der Übernahme von EG-Recht in den bisherigen Mitgliedstaaten in vergleichender Sicht (Wien, Verlag der Österreichischen Staatsdruckerei, 1991); Alfred Kellermann/Jaap de Zwaan/Jeno Czuczai (eds), EU Enlargement – The Constitutional Impact at EU and National Level (The Hague, T.M.C. Asser Press, 2001); Franz C. Mayer and Mattias Wendel, ‘Die verfassungsrechtlichen Grundlagen des Europarechts’, in A Hatje and C Müller-Graff (eds), Europäisches Organisations- und Verfassungsrecht. Enzyklopädie Europarecht I (Baden-Baden, Nomos, 2014) 163 (§ 4); Anne-Marie Slaughter, Alec Stone Sweet, and Joseph HH Weiler (eds), The European Courts and National Courts – Doctrine and Jurisprudence. Legal Change in Its Social Context (Oxford, Hart, 1998).

796  Stefan Griller and Lina Papadopoulou This has changed, however. First, politically. Leaving the EU, as the United Kingdom did, with ‘taking back control’ or ‘reclaiming sovereignty’ as a driving momentum2 signals that this relationship no longer goes undisputed. A slightly lesser degree of discontent and challenge became visible in the controversy on respecting the fundamental EU values, not the least in Poland and Hungary.3 Second, and correspondingly, several hallmark decisions by supreme national courts went beyond details and fine-tuning and openly rejected the supremacy of EU law and, what is more, also rejected the respective CJEU judgments. This happened in the Czech Republic,4 Denmark,5 and, most recently, in Germany and Poland.6 It goes without saying that this development has the potential to destabilise the legal foundations of the EU. The core of this volume consists of 28 reports addressing the national constitutional law bases of EU integration in the Member States.7 The present analysis pulls all these studies together with the aim to present their findings and draw some conclusions. This analysis is guided by the common structure underlying the project. The first set of sections are, therefore, looking at the National Constitutional Systems and cultures (II.), the constitutional foundations of EU membership (III.) and the constitutional obstacles to further EU integration (IV.).8 Two further sections complement them, providing the inverse perspective, ie the EU law perspective on National Constitutions (V.) and some doctrinal reflections on the intricate relationship between those legal orders (VI.). Finally, some Conclusions are drawn (VII).

II.  National Constitutional Systems and Cultures A.  Forms of Constitutions Throughout Europe, the process of constitutionalisation dates back to the eighteenth century, with the result that today all EU Member States are constitutional states. Nevertheless, interruptions and discontinuity are also part of that history. In Western Europe they were partly caused by authoritarian regimes before World War II9 or by the war itself. In Central and Eastern Europe they were also caused by the much longer-lasting communist regimes. In the latter case, modern constitutions were adopted10 after the fall of communism either as entirely new texts or as restorations and continuations of older ones that had been in force before the Soviet Union gained 2 Compare the contribution of Craig in this volume; compare also Paul Craig and Gráinne de Búrca, EU Law, 7th edn (Oxford, OUP, 2020) 35. 3 Compare the contributions in András Jakab and Dimitry Kochenov (eds), The Enforcement of EU Law and Values, (Oxford, OUP, 2017); and recently Kim Lane Scheppele, Dimitry Kochenov and Barbara Grabowska-Moroz, EU Values Are Law, after All: Enforcing EU Values through Systemic Infringement Actions by the European Commission and the Member States of the European Union, (2020) 39 Yearbook of European Law 3. 4 Compare the contribution of Tichý and Dumbrovský in this volume. 5 Compare the contribution of Wind in this volume. 6 Compare below, n 103 and n 151. 7 Including the former EU Member State, the United Kingdom – whose experience as a former member is very valuable also for actual and future members. 8 The aim of these subchapters is to give a bird’s-eye view on the situation in the 28 states. For more details and references, please consult the country reports in this volume. Some of the following also draws from Stefan Griller, ‘Analytical report on the Legal background of Member States’ Positions towards Economic and Fiscal Integration’, in Stefan Griller and Elisabeth Lentsch (eds), EMU Integration and Member States’ Constitutions (Oxford, Hart, 2021) 1. 9 Eg in Austria, whose constitutional history was interrupted by the period of authoritarian rule from 1934–45. 10 Among these constitutions, the new Bulgarian Constitution was adopted on 12 July 1991; Croatia adopted its first Constitution in 1990, although it had already had one as part of the Federal Republic of Yugoslavia before it became independent.

Squaring the Circle – How National Constitutions and EU Law Interact  797 control. For these ‘new’ EU Member States the adoption of new or renewed constitutions signified the return to European constitutional ‘normality’ after several decades of authoritarianism. These states embraced constitutionalism mainly in order to maintain political pluralism, to avoid conflict with previously powerful groups and to strengthen the social cohesion of diverse groups (see Belov in this volume). Their constitutions are modern, reflect the desire to maintain liberal democracy and encapsulate elements of direct democracy and the rule of law (Constitutional Courts), as well as fundamental rights, inspired by other European constitutions as well as by the European Convention on Human Rights (ECHR). Recently, however, various shades of illiberal democracy coupled with autarchic populism have emerged in many countries across the continent. Another peculiarity can be found in the Constitution of the Republic of Cyprus, which, since 1964, has functioned under the ‘doctrine of necessity’, since it was designed to regulate the affairs of a bi-communal country but is now applied to only one of the two communities, the Greek Cypriot one. The vast majority of the EU Member States have a formal Constitution consisting of a unitary written text, with some exceptions.11 The most radical exception is the (now ex-Member State) United Kingdom, where there is no formal and codified constitutional text. Nevertheless, the United Kingdom does have both recognised constitutional principles and more recently judicially recognised constitutional statutes ‘that are regarded as especially important in fashioning the UK constitutional landscape’ (see Craig in this volume).

B.  Fundamental Constitutional Principles European constitutionalism as a political ideal comprises more than a ‘formal Constitution’. It is the set of norms in a given legal system that is more stable in terms of alteration procedures than the (subordinate) rest of the legal order. Most formal constitutions enjoy higher legal rank than ordinary legislation and primacy or ‘supremacy’ within the internal legal order. This higher legal rank results in the formation of some form of hierarchy of norms and allows in most cases for the review of constitutionality of laws. A ‘thicker’ concept of a Constitution would include12 norms defining and authorising certain organs to enact (and to enforce) law which is directly binding on the citizens, the law-making procedures, and limits to the powers of the authorised organs, especially limits flowing from (fundamental) rights of citizens or all humans, and requirements of checks and balances. European constitutionalism, in this vein, involves guarantees for democracy and human rights, combined with the social state. More specifically, constitutions feature the principles of rule of law and separation of public powers, combined with a system of checks and balances between the different branches of power, respect for fundamental rights and liberties, combined with the majoritarian principle, political pluralism and a multiparty system, as well as representation through free periodic elections. Most European constitutions consolidate a socially oriented market economy and social rights, which, however, do not exhibit the same normative density or enjoy the same normative force as civil and political rights. Explicitly or implicitly (especially in the case of the EU’s new Eastern European Member States), constitutions aim to safeguard the democratic state and prevent the rise of totalitarian regimes, although not all of them adopt a 11 The Austrian Constitution is fragmented and consists of many constitutional acts of equal rank; also the Czech, Swedish and several other constitutions consist of several constitutional acts. 12 The following is a simplification. Compare for more details only Paul Craig, ‘Constitutions, Constitutionalism, and the European Union’, (2001) 7 European Law Journal 125.

798  Stefan Griller and Lina Papadopoulou type of ‘militant democracy’.13 These combinations of procedural and substantive guarantees, on the one hand, and liberty and solidarity, on the other, form the model of European constitutional democracy. These common features are reinforced by Article 2 of the Treaty on European Union (TEU), which emphasises the respect for human rights, democracy and the rule of law as (legally binding) values common to the Member States.

C.  Organisational Diversity Many EU Member States are ‘unitary states’ where only local issues fall within the competences of elected regional and local authorities. Only three states are ‘fully fledged’ federations of states or regions (Austria, Belgium and Germany). There are also some which have either devolved certain competences to regions (Italy and Spain), and others which form federacies with autonomous regions, such as Denmark (Greenland and the Faroe Islands), Finland (Åland), France (New Caledonia, French Polynesia and Corsica), the Netherlands (the Caribbean island groups of Aruba, Curaçao and Sint Maarten), or quasi-federacies like Cyprus (with Greek and Turkish communities, according to the Constitution of 1960). Correspondingly, there is a rich variety in organisational settings, which affect, among other things, the implementation and enforcement of EU law. Some countries with a more unitary tradition (eg France) tend to implement EU law in a centralised manner, be it by legislation or subordinate measures including administrative action. The mentioned federal states and those with ‘autonomous’ regions delegate power and responsibility to constitutive parts to a significantly larger extent than unitary states. As a consequence, the reports exhibit substantive differences on how the Member States cope with the respective EU legal framework which guarantees – in the absence of specific EU measures regulating implementation and enforcement – national procedural autonomy which must nevertheless respect the requirements of ‘effectiveness’ and ‘equivalence’ vis-à-vis EU law.14 Most EU Member States are republics with an elected Head of State (President of the Republic), a minority of countries retaining a monarchy (Belgium, Denmark, Luxembourg, the Netherlands, Spain and Sweden). The parliamentary system prevails also within the latter, since the (unelected) monarchs dispose only of symbolic and ceremonial competences. Cyprus has a presidential system.15 Five EU Member States have a semi-presidential system (France, Poland, Portugal and Romania, with Lithuania also exhibiting some traits of semi-presidentialism). In some EU Member States there is a second parliamentary chamber (Austria, Germany and the Netherlands) or a Senate (Belgium, Czech Republic, France, Ireland, Italy, Poland, Romania and Spain). Italy is an example of ‘perfect bicameralism’. In some other bicameral legislatures, the more powerful house can only overrule the other on certain issues (an intermediate model between perfect and imperfect bicameralism). Interestingly enough, the EU itself, as a quasifederal system, also has a bicameral legislature, exercised by both the Council and the European Parliament, with the latter gradually gaining more powers. 13 This is, for example, the case with the German Constitution, which foresees the possibility of prohibiting undemocratic parties. 14 Compare Craig/de Búrca, EU Law (n 2) 262 ff. 15 The particularity of the Republic of Cyprus also lies in its form of consociational democracy embodied in the 1960 Constitution providing for a bi-communal state, which, however, has not functioned as such since the break-up of the government in 1964 and especially after the Turkish invasion of 1974, which consolidated the territorial and political segregation of the two communities.

Squaring the Circle – How National Constitutions and EU Law Interact  799 In parliamentary democracies, substantial power usually rests with the Government. However, in some countries (eg Denmark or Germany) there is an accentuated prevalence of Parliament with the other branches of government being dependent on more or less detailed authorisations included in Acts of Parliament. Another criterion that divides EU Member States is the degree of consensus-oriented political culture, contrasted by a tendency to polarisation and lack of compromise in some countries. There seems to be a North-South cleavage in this respect. However, the recent crises seem to have pushed into a direction of closer cooperation between political forces in the crisis-ridden southern countries.

D.  Constitutional Amendments Amendment procedures in Europe differ among states.16 The superiority of the formal Constitution is, in most cases, manifested in two ways. First, constitutional revisions involve a special procedure that is harder to fulfil than enacting or amending regular laws and is provided for in the Constitution itself. Usually there is a requirement for a reinforced majority,17 sometimes (also) for a referendum,18 and rarely for intermediate elections. The overall spectrum 16 For the amendment procedures see Xenophon Contiades (ed), Engineering Constitutional Change – A Comparative Perspective on Europe, Canada and the USA (London/New York, Routledge, 2013). 17 More specifically: in Bulgaria three-quarters of the votes of all Members of the National Assembly in three ballots on three different days (Art 155 Const), a special organ, the Grand National Assembly, can adopt a new Constitution; in Italy (Art 138 Const) amendment bills are adopted by each House after two successive debates at intervals of not less than three months, and have to be approved by an absolute majority of the members of each House in the second vote; in Lithuania (Art 148 Const) amendments must be passed at the Seimas twice by two-thirds of all members; in Luxembourg (Art 114 Const) amendments must be voted on in two successive votes, separated by an interval of at least three months, by twothirds of the Parliament; in the Netherlands (Art 137 Const) the requirement is for two-thirds of the votes cast. A three-fifths majority of all Members of Parliament is required in the Czech Republic (Art 39 Const): ‘three-fifths of all Deputies and three-fifths of all Senators present are required for the adoption of a constitutional act or for giving assent to the ratification of treaties’), in Slovakia (Art 84 Const) and in Spain (in each House, Sec 167 Const). A two-thirds majority vote of all members of the Parliament is required in the following countries: Hungary (Art S); Germany (of the Members of the Bundestag and two-thirds of the votes of the Bundesrat – Art 79 Basic Law); Croatia (Art 138 Const); Latvia (Art 76 Const), in three readings by a majority of not less than two-thirds of the members of the Saeima present; Portugal (Art 286[1] 1 Const); Romania (of each Chamber, Art 151 Const); Slovenia (Art 169 Const). In Poland (Art 235 Const) a majority of at least two-thirds of votes in the presence of at least half of the statutory number of Deputies in the Sejm, and in the Senate an absolute majority of votes in the presence of at least half of the statutory number of Senators is needed. In Austria, ‘normal amendments’ of the Constitution require the presence of at least half the members and a two-thirds majority of the votes cast in Parliament (Nationalrat) and the explicit designation of the provision as a Constitutional Act. In Croatia the Constitution may be amended upon a proposal by one-fifth of the members of Parliament, the President of the Republic and the Government. Amendments then require a two-thirds majority of MPs. 18 For example, a constitutional referendum is provided for in the following Member States: Croatia (optional, Art 87 Const); Estonia (optional, three-fifths majority of the membership of the Riigikogu is required, Art 164 Const); France (under some circumstances it can be avoided: if the President of the Republic decides to submit it to Parliament convened in Congress, any Government Bill to amend the Constitution can then be approved only if it is passed by a three-fifths majority of the votes cast: see Art 89); Ireland (any amendment bill needs to be passed by both Houses of the Oireachtas, and then submitted by referendum to the will of the people; Art 46(2) Const); Italy (Art 138: optional if a request is made by one-fifth of the members of a House or five hundred thousand voters or five Regional Councils); Latvia (Art 77 Const: only for Arts 1, 2, 3,4 6 and 77 Const). Similarly, in Lithuania (Art 148 Const) the provisions of the First Chapter (‘The State of Lithuania’) and the Fourteenth Chapter (‘Alteration of the Constitution’) may be altered only by referendum. In Poland, a confirmatory referendum after the amendment bill has already been accepted by the Senate is foreseen for some fundamental provisions (Art 235[6] Const). A confirmatory referendum is also foreseen in Romania (Art 151[3] Const). Slovenia (Art 170): ‘The National Assembly must submit a proposed constitutional amendment to voters for adoption in a referendum, if so required by at least thirty deputies’. On the other hand, Art 8(3) of the Hungarian Constitution explicitly stipulates that ‘no national referendum may be held on any matter aimed at the amendment of the Fundamental Law’.

800  Stefan Griller and Lina Papadopoulou ranges between rigidity19 and mildness.20 The most rigid constitutions are those requiring a three-stage-procedure for performing amendments: i) a proposal by Parliament; ii) parliamentary elections, and iii) completion of the revision by the next elected Parliament, combined with enhanced majorities.21 In many cases, intervals and repetitions of the voting are required in order for the procedure to be completed, or a time lapse between the two decisions.22 Secondly, several European constitutions include an eternity clause.23 A stark contrast exists in Austria where any amendment – called a ‘total revision’,24 if it affects core principles of the Constitution – is possible only after certain procedural requirements are respected (qualified majorities and mandatory referendum). This deviates from the majority of European constitutions, which are seen as guarantors of constitutional ideals and principles. In addition, and despite the formal amendment procedures and eternity clauses, amendments to European constitutions are informed by both EU law, especially the EU Treaties, and the ECHR and the respective jurisprudence of the CJEU and the European Court of Human Rights (ECtHR); at least as long as European countries remain members of the EU and of the ECHR. Member States are bound by these obligations, and accordingly their constitutions are no longer exclusively self-referential as they might have been previously. The more rigid a Constitution is, the more likely tacit or implicit modifications through the influence of EU and ECHR law can be observed. This is combined with a teleological interpretation of the Constitution in an effort to accommodate changing (social) conditions. Such constitutional evolution is often based on ‘nonclassical’ techniques of evolutionary and teleological interpretations, sometimes also by lower ranking legislation, which is sometimes labelled as ‘deconstitutionalisation’.25

In Latvia, electors, comprising not less than one-tenth of the electorate, have the right to submit a fully elaborated draft of an amendment to the Constitution or of a law to the President, who must present it to the Saeima. If the Saeima does not adopt it without any change to its content, it must then be submitted to national referendum (Art 78 Const). In Austria a referendum is obligatory only for a total revision of the Constitution. 19 The most rigid constitutions are those of Belgium and Greece, where elections must take place between the first and second stages of the accomplishment of a constitutional revision. In Belgium elections are called if Parliament decides to alter the Constitution, whereas in Greece they are called on another, constitutionally foreseen, ground. It is only the second Parliament that decides upon the new constitutional provisions. 20 The Bulgarian and Croatian Constitutions have already undergone several amendments. The Austrian Constitution has been amended more than 100 times by constitutional Acts – without referenda; the one and only mandatory referendum was performed at the occasion of EU accession, which was considered a total revision of the Constitution. 21 This system – with minor differences – exists in the following countries: Belgium (Art 195, according to which the two Houses are automatically dissolved after a decision to revise the Constitution, provided that at least two-thirds of the members who make up each House are present; and no change is adopted unless it is supported by at least twothirds of the votes cast); Denmark (Art 88), where a referendum is also needed after an amendment bill has been passed unamended by the membership of a new Folketing); Estonia (Arts 163–65, according to which the Constitution may only be amended by an Act which has been passed by two successive memberships of the Riigikogu, or a referendum; any unamended amendment bill has to be adopted by both memberships; Finland (Art 73: the following Parliament needs to adopt any amendment bill without material alterations); Greece (Art110 Const, two consecutive Parliaments, deciding with a three-fifths and an absolute majority interchangeably); the Netherlands (Art 137) (the Lower House is dissolved and after the new one has assembled, the two Houses of the States General must consider, at a second reading, the amendment bill already voted upon); Sweden (Chapter VIII, Art 15). 22 For example, in Portugal (five years, but an exception is provided for if voted by a four-fifths majority of all the Members of Parliament, Art 284(2) Const) and in Greece (five years, Art 110(6)). 23 See, for example, Art 110(1) of the Greek Constitution, Art 288 of the Portuguese Constitution and Art 79(3) German Basic Law. 24 ‘Total revision’ (Gesamtänderung), as a terminus technicus means that one or more of the fundamental constitutional principles (ie democracy, rule of law, separation of powers, liberal fundamental rights and freedoms, federal and republican character of the state) is amended or abandoned. However, until today only one such total revision was made in 1994, preceding Austria’s EU accession. 25 Belgium may serve as an example, where parts of the material Constitution are spelled-out in ‘Special Majority Acts’ or even in ordinary legislation.

Squaring the Circle – How National Constitutions and EU Law Interact  801

E.  Courts and the Constitutional Review of Statutes Based on the separation of powers principle, courts are formally independent of the executive and legislative powers. Theoretically, in all EU Member States the legislative power, exercised by the Parliament, enjoys primacy over the executive and the judiciary. Ideal-typically, this is so because the courts do not enjoy democratic legitimation through elections. This lack of democratic legitimacy makes them inappropriate topoi for the resolution of political disputes, and, therefore, they are expected to exhibit self-restraint when faced with political problems and dilemmas. This would, however, not exclude legal scrutiny of political decisions. In many EU Member States, the constitutional review of legislation, deriving from the superiority of the Constitution, is exercised by a Constitutional Court. Some states, however, do not dispose of such a court (eg Denmark,26 Finland, Greece, the Netherlands and Sweden). ‘Constitutional review’ in these countries is either excluded or exceptionally performed by ordinary courts. The stance taken by the courts, including Constitutional Courts, in different European states towards the democratically legitimated legislator varies widely from one of non-involvement to very detailed scrutiny. In some countries, judicial self-restraint is more common than in others, due to different legal cultures. It is especially in Scandinavia and the Netherlands that the courts are very reluctant to challenge decisions taken by Parliament. Under a culture of positivism, political decision-making takes priority, and the courts are expected to exhibit self-restraint. In other countries, the courts, especially the supreme ones, and most notably the Constitutional Courts, exercise control of the constitutionality of laws quite extensively; Italy and Germany may serve as examples in that regard. While constitutional issues on the relationship between EU law and national law arise everywhere, it might be fair to identify a specific flavour in those countries where Constitutional Courts are empowered to review the constitutionality of subordinate acts, including national legislation, and to nullify statutes in cases of violations. Institutionalised review of that kind accentuates the hierarchical consequences not only vis-à-vis national but also vis-à-vis EU law, ie limits flowing from constitutional authorisations for EU membership. Particularly rich experience arguably exists, in this respect, in Germany, Italy and France.27 The flipside of constitutional review, however, is the possibility for national courts to accept EU law as an integral part of their yardstick to scrutinise subordinate acts such as ordinary legislation or administrative measures. This is different from direct effect and supremacy of EU law, which, in a wider sense, might (also) be considered as a means of scrutiny of national law. Even if these two distinct features of EU law are accepted, which is everywhere the case – subject only, as mentioned, to possible limits to EU integration – this does not necessarily include integrating EU law into those clauses of the national constitutions which are applied when scrutinising national law by Constitutional or otherwise Supreme Courts. Normally, such integration is not foreseen. As a consequence, EU law cannot be enforced in the same way as national constitutional law, eg by expressly declaring void or annulling subordinate law, as is regularly done in cases of violating 26 Denmark has neither a Constitutional Court nor a constitutional council. So, in Denmark there has only been one case where a court found a statute to be unconstitutional. Nevertheless, although constitutional review is still not part of the Danish political culture, today it is more accepted than before, mainly due to the decreasing standing of the political elites. Interestingly enough, it is the executive (the Ministry of Justice) which influences the application of European law and case law in Denmark. 27 In addition to the contributions of Jestaedt, Cartabia/Guastaferro and Rainer Palmstorfer/Sermet/Vranes in this volume; compare Monica Claes, The National Courts’ Mandate in the European Constitution (Oxford, Hart, 2006) 187 ff; and many contributions in M Claes et al (eds), Constitutional Conversations (n 1).

802  Stefan Griller and Lina Papadopoulou the Constitution.28 However, more recently exceptions are emerging. They concern specifically enforcing the EU Charter of Fundamental Rights (CFR, Charter) within the national legal order. In Austria and in Belgium, the Constitutional Courts considered guarantees under the Charter as equivalent to Fundamental rights enshrined in the national Constitution. Consequently, provisions designed to enforce Fundamental Rights (of a constitutional rank) vis-à-vis the legislator (by striking down unconstitutional legislation) or the administration (by annulling administrative decisions violating fundamental rights) were extended to enforce Charter rights as part of the yardstick. In these countries, Charter rights can be enforced in the same manner as constitutional ‘law of the land’.

F.  Diverging Cultures of Interpretation Interpreting constitutions in different polities inevitably produces divergence, if not fragmentation.29 There is no such thing as a homogeneous legal world order. The same is true within the EU. Diverging history and tradition, diverging languages lead to separate communities where sometimes even one and the same expression denominates different content. Such discourse is not the least informed by the organisational setup of the respective polity. In our context, eg, the differing hurdles for constitutional change as well as the existence or lack of constitutional scrutiny by Constitutional or Supreme Courts can make a difference. All of this informs the practice of interpretation and hence the meaning of the law. This extends, among others, to reading the ‘opening clauses’ of the national constitutions allowing for EU integration. In some EU Member States, interpretation of constitutional law is more focused on ‘objective’ rather than ‘restrictive textual’ interpretation.30 This more dynamic interpretation allows for greater flexibility when facing new developments or crises. It includes increasing weight for the principle of proportionality. It might be fair to conclude that the more rigid the amendment procedures, the more flexible or teleological in nature are the methods of interpretation by the supreme courts. In doing so they compensate to a certain extent for the unavailability of amendments that would otherwise accommodate for necessary adjustments. Techniques may include a generous attitude vis-à-vis elements of ‘deconstitutionalisation’, whereby meaning is attributed to the Constitution by specific majoritarian parliamentary acts or even in ordinary legislation, or evolutionary and teleological interpretation including the emergence of unwritten constitutional principles and constitutional customs. International treaties, such as the TEU and the ECHR, are influencing such developments. Constitutional and Supreme Courts in EU Member States have been developing variations of teleological interpretations which sometimes come close to tacitly changing the Constitution.31 28 It goes without saying that this might lead to difficult questions as to what extent such exclusion would violate the EU efficiency and equivalence principle as developed by the CJEU. The exceptions in Austria and Belgium mentioned below are partly driven by the intention to avoid any violation of those principles. 29 Compare for the following the contributions in M Claes et al (eds), Constitutional Conversations (n 1); and in Georg Lienbacher (ed), Verfassungsinterpretation in Europa. Heinz Schäffer Gedächtnissymposion (Vienna, Jan Sramek Verlag, 2011); for a theoretical perspective also Stefan Griller, ‘Fragmentierungen im Öffentlichen Recht: Diskursvergleich im internationalen und nationalen Recht’, (2018) 77 VVDStRL (Berlin, de Gruyter) 237; Uwe Kischel, ‘Fragmentierungen im Öffentlichen Recht: Diskursvergleich im internationalen und nationalen Recht’, (2018) 77 VVDStRL (Berlin, de Gruyter) 285, 306f, 309f; Olivier Jouanjan, ‘Fragmentierungen im Öffentlichen Recht: Diskursvergleich im Verfassungsund Verwaltungsrecht’, 77 VVDStRL (2018) (Berlin, de Gruyter) 351; Franz Reimer, ‘Fragmentierungen im Öffentlichen Recht: Diskursvergleich im Verfassungs- und Verwaltungsrecht’ (2018) 77 VVDStRL (Berlin, de Gruyter) 413; all of them with further references. 30 See eg the report on the Czech Republic. 31 Compare for this and the following eg the reports on the Czech Republic, Belgium and Greece.

Squaring the Circle – How National Constitutions and EU Law Interact  803 This might be qualified as a parallel development if compared to the teleological interpretations adopted by the CJEU. Its allegedly sometimes exaggerated reliance on teleological interpretation (‘effet utile’) might result from the fact that any change of primary law needs unanimity among the EU Member States. This augments the temptation for the CJEU to step in as a ‘problem solver’ by means of interpretation, where the Masters of the Treaties are unable to act due to the veto right granted to every single member. Another feature deserving attention is ‘conform’, ‘consistent’ or ‘harmonious’ interpretation. It is being applied both in countries with and without constitutional scrutiny by Constitutional or Supreme Courts. They tend to read lower-ranking law – eg ordinary legislation, which is subordinate to the Constitution – to the end of avoiding conflicts which would result in a violation. Whenever possible, they interpret the lower-ranking law in conformity with a superior law, eg ordinary legislation in conformity with the Constitution. Again, there is a certain parallel to EU law and it might be that EU conform interpretation of national law is less controversial in countries where interpretation in conformity with the Constitution is common practice. Accordingly, many courts (eg in Austria, Belgium, the Czech Republic, Germany, Greece and Italy) interpret national law in accordance with EU law, thereby respecting a requirement established by the CJEU.32 There is a delicate relationship between ‘consistent interpretation’ of national law with EU law, direct applicability and supremacy in cases of violations of EU law,33 and the – as mentioned rare34 – authorisation to nullify national law. Comparing these variations, the efficiency in enforcing EU law might be almost equal. Combined with the EU law obligation to expressly repeal national law violating EU law, the second and the third options lead to a higher degree of legal certainty. Legislation and administrative measures have to be amended accordingly. Conform interpretation does not include such explicit clarification. In combination with the prejudicial force of court decisions the difference might not be too big. Consistent interpretation normally is only available in the absence of insurmountable (methodological) obstacles flowing from the meaning (wording) of the national law. The absence of any power to nullify national law violating EU law might matter: ‘Creative’ interpretation of national provisions by the courts, which brings those provisions into conformity with EU law, avoid breaking EU law in cases where nullification is not an option. As long as this is achievable, direct application of EU law and nullification of national law, both, including the verdict that EU law had been violated, can be avoided. While traditions differ between EU Member States, it is a significant feature of EU integration that approximation happens in various ways, sometimes by stealth. It has rightly been observed that, also regarding interpretation, a sort of an indirect and partly implicit dialogue35

32 Compare only, as one out of many examples, CJEU 05.10.2004 Joined Cases C-397/01 to C-403/01 (Pfeiffer and Others) ECLI:EU:C:2004:584, para 116: ‘In that context, if the application of interpretative methods recognised by national law enables, in certain circumstances, a provision of domestic law to be construed in such a way as to avoid conflict with another rule of domestic law or the scope of that provision to be restricted to that end by applying it only in so far as it is compatible with the rule concerned, the national court is bound to use those methods in order to achieve the result sought by the directive’. However, it becomes clear from the context that the CJEU goes beyond ‘directive conform’ interpretation and instead aims at comprehensive ‘EU conform interpretation’: ‘The requirement for national law to be interpreted in conformity with Community law … is inherent in the system of the Treaty, since it permits the national court, for the matters within its jurisdiction, to ensure the full effectiveness of Community law when it determines the dispute before it …’ (para 114). 33 ‘Regular’ direct applicability like in the case of EU regulations in the absence of conflicting national law is another matter. 34 See above in the text after n 28. 35 See Claes et al (eds), Constitutional Conversations (n 1).

804  Stefan Griller and Lina Papadopoulou has developed between the national legislators, the national Constitutional Courts, the CJEU and the EU Member States as the ‘Masters of the Treaties’.36 One might qualify it as an emerging system of informal cooperation between the various actors at national and EU level.37 Yet, it is rare that this dialogue is made transparent in that explicit reference is made to a concrete piece of foreign legislation or decision of another court – with the natural exception of preliminary references and infringement procedures before the CJEU and the respective reaction of national legislators or courts.

III.  Constitutional Foundations of EU Membership A.  Explicit or Implicit Authorisations for Accession In a first step, we may differentiate between EU Member States that enacted, mostly as a preparation of EU accession, specific constitutional provisions authorising EU membership, and others that did not. The latter happened mostly because in the beginning that was not deemed obligatory, or no sufficient parliamentary majority for specific constitutional provisions could be found. Consequently, this second group acceded to the EU on the grounds of general constitutional provisions dealing with international treaties and obligations.38 Countries with specific constitutional authorisations – like Austria, Bulgaria, Croatia, Germany, Greece,39 Hungary, Ireland,40 Slovakia, Sweden and Portugal – can point to explicit wording in their constitutions specifying the requirements for the transfer of powers to the EU. Normally, in their constitutional context, and partly explicitly, also possible limits are visible at first glance. In countries without such specific constitutional provisions (at least) at the time of accession – as in the Czech Republic, Denmark, Finland, France, Italy, Luxembourg, the Netherlands, Poland, Slovenia and Spain, where such limits are often not mentioned, one could even speculate on their very existence. However, the dividing lines are far from being clear-cut. In some cases, provisions were created in view of EU accession without expressly mentioning it (eg Greece), which makes the basis ‘general’. Other countries, once they became EU members, changed the constitutional foundation from a general to a specific basis (eg Germany), or complemented originally general constitutional provisions by constitutional amendments, but without creating a ‘general membership clause’ (eg France), or complemented specific, but narrow authorisations for accession by clauses covering EU Treaty amendments (eg Austria). 36 Arts 4 and 5 TEU are an example for the latter, equally not openly revealing the constitutional debate surrounding them. 37 Compare also Grabenwarter, ‘National Constitutional Law’ (n 1), especially 125 ff; Mayer/Wendel, ‘Grundlagen’ (n 1) §§ 232 ff, with further references. 38 Compare for the following also the studies cited above n 1; furthermore, Grabenwarter, ‘National Constitutional Law’ (n 1) 84 ff, 95 ff; Stefan Griller, ‘Executive Summary’, in European Parliament (ed), Study PE 432.750 (n 1) 21; Koen Lenaerts and Piet van Nuffel, Constitutional Law of the European Union, 3rd edn, (London, Sweet & Maxwell Ltd, 2011); Mayer/Wendel, ‘Grundlagen’ (n 1) paras 13 ff. 39 Greece is one among several ‘particular cases’: Art 28 Const was enacted in 1975 with a view to joining the then Communities, but without mentioning it. In 2001, an ‘interpretative clause’ was added not only to Art 28 of Const, but also to Art 80, to accommodate the participation of Greece in the EMU; compare Papadopoulou in this volume. Also, France might fall into this subgroup where originally no specific ‘EU membership provision’ existed, but constitutional amendments were later enacted in order to accommodate amendments to EU Treaties. 40 Ireland is a peculiar case where each and every Treaty change almost inevitably requires an amendment to the Irish Constitution, as well as a referendum: see below after n 41, and Barrett in this volume.

Squaring the Circle – How National Constitutions and EU Law Interact  805 Countries without specific authorisations may have to observe similar or even stricter requirements, which can be inferred from the systematic interpretation of the constitutional norms governing international obligations in general (as in Finland). It may appear that the lack of specific provisions is due to the dissent on the interpretation of existing constitutional limits and the desirability of changing them, as, arguably, in the Czech Republic and in Poland. In other countries, like Italy, the flexibility of existing constitutional provisions on international obligations was welcomed at the beginning, and only later supplemented by constitutional amendments. Against this backdrop, it does not come as a surprise that the divide between the abovementioned groups is less important than it might seem at first. The reason is that explicit constitutional provisions may be blurry and also informed by their systematic constitutional context, which makes the conditions for EU accession and EU membership a matter of thorough interpretation of the entire Constitution. Even explicit ‘EU provisions’ might not explicitly mention concrete constitutional limits to EU membership or to the further transfer of powers (like in Austria), or they might refer to such limits only indirectly through specific authorisations of treaty amendments (like in France), thereby pointing to underlying constitutional concepts like sovereignty. Consequently, the content of the constitutional authorisation and its possible limits can only be explained by systematic interpretation of the entire Constitution, and not only by pointing to the concrete provisions authorising EU membership. As a result, core values of the national constitutions establishing limits to European integration can be traced in all countries irrespective of the existence or lack of specific constitutional provisions governing EU membership. It goes without saying that, nevertheless, explicit provisions have at least a comparatively bigger potential to clarify issues which otherwise remain controversial.

B.  Dynamic and Static Constitutional Authorisations Another classification differentiates between ‘dynamic’ and ‘static’ constitutional foundations of EU membership.41 Dynamic constitutional provisions allow entering the EU and developing it further without the necessity to repeatedly change the Constitution for every Treaty amendment. These provisions, however, may require specific majorities equalling or coming near to those needed for a constitutional change. Arguably, and inevitably subject to the detailed discussion for every country, this situation today is given in most of the EU Member States: Austria, Belgium, Bulgaria, Croatia, Czech Republic, Denmark, Estonia, Germany, Greece, Italy, Latvia, Lithuania, Luxembourg, the Netherlands, Portugal, Romania, Slovakia, Spain and Sweden. Static constitutional authorisations would – almost inevitably – require introducing amendments to the national Constitution each time EU Treaty amendments are at stake. Ireland is a specific example of that. There are also ‘borderline cases’ between the two camps, where certain Treaty amendments would come under the existing authorisation,41 while others, even if not entirely novel, would not be covered. Arguably, this is so in France. The categorisation as ‘dynamic’ or ‘static’ cannot dispense from interpreting the constitutional foundations with a view to possible limits to further integration or constitutional scrutiny of secondary legislation. Per se it does not solve any hard cases either. Some constitutional texts are

41 Mayer/Wendel,

‘Grundlagen’ (n 1) paras 17 ff.

806  Stefan Griller and Lina Papadopoulou explicit in this respect – like in Germany and Sweden – others, arguably the majority of them, are silent. However, this by no means entails a carte blanche for the national legislator or the government to authorise further transfers of power. By contrast, systematic and sometimes historic interpretation reveals limitations that would flow from other provisions of the Constitution or from the position of the ‘integration clause’ in the context of the entire Constitution. Even where explicit constitutional wording, academic debate or relevant case-law is lacking, limitations might flow from the constitutional culture in the respective country not less than from an in-depth analysis of the Constitution’s text.

IV.  Constitutional Obstacles to Further EU integration A.  Limits to (Further) Transfers of Powers Through Treaty Amendments 1.  Core Values – Competences – Constitutional Identity Legal authorisations, in our case the constitutional authorisation to join the EU, include (procedural and/or substantive) conditions to take that step. These conditions very often entail or may be interpreted as limits to that authorisation, in the sense that not respecting them would violate the Constitution. Thus, it does not come as a surprise that respective debates and practice have unfolded in almost all EU Member States. The differences between various EU Member States, however, are substantive. Common elements need to be identified with caution: even identical words might convey diverging meaning in the context of the respective Constitution and in the reading of national authorities including courts.42 Core values of the Constitution and state competences can be identified as relevant for any future transfer of powers to the EU. ‘Relevance’ means that the respective transfer of powers either has to observe existing limits or would otherwise necessitate amendments of the Constitution requiring qualified majorities in Parliament, in some cases also a referendum. These majorities sometimes might be difficult to obtain. Some issues, especially relating to the ‘sovereignty’ of the country are specifically sensitive and claimed, by some commentators, to be insurmountable as long as the Constitution remains effective. Core values and competences of relevance are: 1. minimum standards regarding fundamental rights protection, democracy, and possibly other principles like the rule of law, 2. additional powers (competences) for the EU, and 3. the preservation of ‘constitutional identity’ as an unalienable core, including the preservation of statehood as such. This discussion is exemplified in the German legal debate on EU integration limits, because of both the exceptionally explicit wording in the Basic Law,43 and of the German Federal Constitutional Court’s (BVerfG) intensive control over the decades. The German situation is closely observed in other countries and the (particularly rich) German debate has been serving, over the decades, as a sort of role model for other countries.44 The jurisprudence of the BVerfG is specifically 42 This is a classic in comparing legal orders: compare above in the text close to and the references in n 29. 43 See below n 51. 44 It was driven by the BVerfG’s jurisprudence accompanied by intense academic writing, starting with the fundamental rights issue – for some ‘la querelle Allemande’ – and developed further regarding the transfer of competences

Squaring the Circle – How National Constitutions and EU Law Interact  807 outspoken, especially when it comes to defining constitutional cores and aspects of judicial scrutiny. It is addressed hereafter as a voice of the pertinent arguments of relevance for all Member States. Nuances and – sometimes crucial – differences between them exist regarding almost every major bone of contention. There is particularly rich experience and practice, documented in this volume, in countries with traditions in constitutional scrutiny of the legislative and the executive (eg Austria, Belgium, Czech Republic, France, Italy, Poland and Spain). For reasons of brevity, the important contributions to the European puzzle, and sometimes also the nuances of the arguments stemming therefrom, are not comprehensively and explicitly addressed in this chapter,45 but hopefully taken into account to a fair extent. In addition, we note that specific difficulties may arise in dealing with the lack of debate on certain issues in some of the EU Member States subject to comparison. What if all odds point in the direction that an argument might be relevant also in an EU Member State where it is (almost) non-existent in the national discourse? We contend that the spectrum of the discourse can and should benefit from comparing and not be strictly limited to the legal facts (court decisions or academic debate) that already happened. Legal discoveries (as opposed to inventions) can be part of legal comparisons. ‘Sovereignty’ might be an example, eg for the Netherlands. Giving up statehood in order to possibly integrate into a European state is not really considered. It would be tempting, therefore, to diagnose the far-reaching openness of the Dutch Constitution to take even such a step. Yet it might not only be for political reasons that the Netherlands was, alongside France, the Member State where the Constitutional Treaty failed to pass the referendum in 2005. Instead, it might also be a legal issue, even if the debate is scarce.46

2.  Fundamental Rights Protection The protection of fundamental rights and freedoms, namely safeguarding the standard of protection of the national Constitution, is an issue for all EU Member States – and at the same time a good example to demonstrate the most important issues. This is ‘how it all began’, with the CJEU, at the time the European Court of Justice (ECJ), originally denying fundamental rights protection at the Community level,47 and the BVerfG ‘responding’ by guaranteeing such protection ‘as long as’ (‘Solange’) it would not be done on Community-level.48 It might be that this beginning and the preservation of the ‘national constitutional identity’. To illustrate this by pointing only to some highlights of the academic research on the matter, and complementary to Jestaedt‘s contribution in this volume, see Rudolf Streinz, Bundesverfassungsrechtlicher Grundrechtsschutz und Europäisches Gemeinschaftsrecht (Baden-Baden, Nomos, 1989); Anne Peters, Elemente einer Theorie der Verfassung Europas (Berlin, Duncker & Humblot, 2001); and Thomas Giegerich, Europäische Verfassung und deutsche Verfassung im transnationalen Konstitutionalisierungsprozeß: Wechselseitige Rezeption, konstitutionelle Evolution und föderale Verflechtung (Berlin, Springer, 2003). This remarkable ‘lead’ of the German example is also reflected in ‘foreign’ academic writing, aiming at the broader picture. Compare eg Hjalte Rasmussen, On Law and Policy in the European Court of Justice (Dordrecht, Martinus Nijhoff, 1986) 318 ff, 400 ff; Joseph HH Weiler, The Constitution of Europe (Cambridge, Cambridge University Press, 1999) 286 ff. 45 Again, for more details we refer the reader to the country reports in this volume. We are aware of the dangers of ‘Germanocentrism’ as well as of the likely criticism. We submit that the somewhat unbalanced citation following from our approach is justified both by the mentioned role model character of the German debate and the effort to maintain this contribution readable. 46 Compare Claes in this volume or Bruno de Witte, Do Not Mention the Word: Sovereignty in Two Europhile Countries: Belgium and The Netherlands, in Neil Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 359. 47 ECJ 04.02.1959 1/58 (Stork & Cie v High Authority) ECLI:EU:C:1959:4, 26. The ECJ even misrepresented the argument of the plaintiff who did not say that the decisions in question had to be assessed from the point of view of German law, as the Court put it, but had argued that the High Authority had failed to respect fundamental rights which were ‘protected under almost all the constitutions of the member states and limit the area of application of the Treaty’ (24). 48 BVerfG 29.05.1974, 2 BvL 52/71 (Solange I) BVerfGE 37, 271 ECLI:DE:BVerfG:1974:ls19740529.2bvl005271.

808  Stefan Griller and Lina Papadopoulou contributed to the persisting conceptual confusion of limits to the transfer of powers and ultra vires control. The transfer of ‘unlimited’ powers without fundamental rights barriers would have been a violation of constitutional limits to the transfer of powers. If, by contrast, the Treaties had to be interpreted as including fundamental rights protection even without express provision from the outset – as the ECJ even at the time of Solange I had already held49 – measures disrespecting fundamental rights which would nevertheless be upheld by the ECJ could at the same time be seen as ultra vires. The BVerfG did not clearly reveal which of the two possible flaws or a combination thereof it addressed.50 Until today, there is no clear separation of these issues across all Member States. Ultra vires might concern all sorts of measures, or measures that at the same time compromise core values like fundamental rights, democracy, or the rule of law. What remained, as far as fundamental rights are concerned, from the early days is the contention that national constitutional guarantees do not only protect against national but also against EU measures, albeit in a changed environment. In the meantime, fully fledged fundamental rights control at the EU level is guaranteed by Article 6 TEU in combination with the Charter, which became binding with the Lisbon Treaty and is being enforced by the CJEU (and national courts, if applicable). Nevertheless, in some countries – like Germany, similarly but differing in detail in many others like Belgium, France, Italy, Spain and Sweden – the Constitution still requires and guarantees a standard of protection at the EU level that is ‘essentially comparable’51 to that afforded by the national Constitution. Particularly remarkable is the express Swedish constitutional requirement that the transfer of decision-making powers is subject to the condition that the level of fundamental rights protection within the EU ‘corresponds’ to that afforded under the Swedish Constitution and the ECHR.52 It was clearly inspired by the Solange jurisprudence and its subsequent ‘codification’ by the German legislator in Article 23 of the Basic Law. Clearly, this is – since the times of Solange I – a continued, and in terms of affected Member States extended, reservation against unconditional primacy of EU law as seen by the CJEU.53 49 Starting with ECJ 12.11.1969 29/69 (Stauder v City of Ulm) ECLI:EU:C:1969:57, para 7. This did not satisfy the BVerfG in Solange I. It argued that protection merely guaranteed by the ECJ would be insufficient against the background that the Community at the time lacked a ‘democratically legitimate Parliament directly elected by general suffrage which possesses legislative powers’ as well as ‘a codified catalogue of fundamental rights’. 50 This was not indispensable given that the BVerfG did not find a violation and left the possible consequences open: nullification of the German statute (Übertragungsgesetz) transferring powers without preserving fundamental rights guarantees or disregarding secondary Community measures that would violate fundamental rights. It stated however already at the time that in cases of conflict with a rule of the Basic Law relating to fundamental rights a Community rule ‘cannot be applied by the authorities or courts’ of Germany. 51 Art 23(1) Basic Law: ‘With a view to establishing a united Europe, the Federal Republic of Germany shall participate in the development of the European Union that is committed to democratic, social and federal principles, to the rule of law and to the principle of subsidiarity and that guarantees a level of protection of basic rights essentially comparable to that afforded by this Basic Law. To this end, the Federation may transfer sovereign powers by a law with the consent of the Bundesrat. The establishment of the European Union, as well as changes in its treaty foundations and comparable regulations that amend or supplement this Basic Law or make such amendments or supplements possible, shall be subject to paragraphs (2) and (3) of Article 79.’ Arts 79(2) and (3) [Amendment of the Basic Law] read as follows: ‘(2) Any such law shall be carried by two thirds of the Members of the Bundestag and two thirds of the votes of the Bundesrat. (3) Amendments to this Basic Law affecting the division of the Federation into Länder, their participation in principle in the legislative process, or the principles laid down in Articles 1 and 20 [basic rights, guarantee of a democratic and social federal state] shall be inadmissible.’ Compare www.gesetze-im-internet.de/englisch_gg/englisch_gg.pdf. 52 Ch 10, Art 6 of the Instrument of Government (Regeringsformen) reads as follows: ‘Within the framework of European Union cooperation, the Riksdag may transfer decision-making authority which does not affect the basic principles by which Sweden is governed. Such transfer presupposes that protection for rights and freedoms in the field of cooperation to which the transfer relates corresponds to that afforded under this Instrument of Government and the European Convention for the Protection of Human Rights and Fundamental Freedoms.’ Translation is taken from Nergelius’ contribution in this volume, who provides also closer analysis. 53 Compare below in the text close to n 172. At the outset, the salience of the issue arises from the respective CJEU jurisprudence on primacy. Otherwise, possible conflicts could be solved by ‘traditional means’: the Member States could

Squaring the Circle – How National Constitutions and EU Law Interact  809 Nevertheless, it should be kept in mind that the conflicts that have arisen since the Treaty of Lisbon – that made the Charter primary law – have never been provoking the contention that fundamental rights protection at the EU level would be generally deficient in the sense of ‘not essentially comparable’ to any of the national standards. Disputes always concerned individual cases – immediately provoking the disputed and across different countries differently answered question whether a singular violation could trigger the EU Member States’ ‘reserve scrutiny’ or if only a general defect at the EU level would. In some EU Member States (like Sweden), a similar compromise (‘essentially comparable’) seems possible for some fundamental rights. For others the level of protection would have to be in conformity with the national constitutional guarantees. In other EU Member States (like Finland and Poland), the notion of ‘essentially comparable’ protection is not in sight or at least not accepted. At least in Poland this is related to the scepticism vis-à-vis the primacy of EU law in general. In essence such scepticism amounts to stipulating the uncompromised relevance of the national standards at the EU level. Evidently, this amounts to either imposing the national tradition to the EU and all other EU Member States, or to jeopardise the unity of the EU legal order, which would be subject to all sorts of ‘constitutional reservations’ stemming from national constitutions.54

3.  The Transfer of Competences in the Light of Democracy a)  General Observations Nuanced differences of national ‘reservations’ if compared to fundamental rights guarantees become apparent when looking at ‘organisational principles’. We again refer to the explicit wording in Article 23(1) of the German Basic Law. Apart from fundamental rights, it specifies the democratic, social, federal and subsidiarity principles, and the rule of law as explicit barriers to participating in the EU. These principles have to be fulfilled at the EU level, and accordingly to be observed by national organs whenever they transfer competences to the EU, participate in the EU legislature, or implement EU law. The mentioned requirements take on a ‘European’ meaning: standards that are commensurate to the status and the function of the Union. What becomes clear from the above is true for all EU Member States, even if, in many of them, the wording of the Constitution as well as academic debate and jurisprudence, accordingly, is less explicit. First, in no EU Member State the national Constitution requires enforcing its own specificities at the EU level. Second, there is no obligation flowing from a national Constitution imposing congruency to the EU Constitution, as regards regulatory details of democracy, fundamental rights protection etc. In other words, EU membership is not conditional on the EU meeting the specific requirements enshrined in the national Constitution and designed for the national legal order. It is always the respective standard, the principle, more or less strictly interpreted55 that has

abstain from implementing and enforcing ‘international law’ obligations until either these obligations or the conflicting domestic provisions would be adjusted. This is excluded in a system where unconditional primacy works in favour of a supranational legal order like in the EU. 54 The CJEU certainly has a point in stressing this danger in CJEU 26.02.2013 C-399/11 (Melloni) ECLI:EU:C:2013:107, see below in the text close to n 172. 55 The extremely strict interpretation in Poland regarding fundamental rights protection (above, text near n 54) arguably cannot be, in an identical manner, transferred to other principles, eg to democracy or other organisational principles like the rule of law. Such a transfer is impossible and would mean that Poland can never join the EU if democracy has to function in an identical manner at EU level as in Poland.

810  Stefan Griller and Lina Papadopoulou to be observed, in the context of EU integration. This necessitates complex interpretation entailing a considerable amount of uncertainty. At the same time, transferring powers to the EU inevitably includes modifications of those principles at the national level, even if the standards are commensurate. For the sake of argument, and concentrating in the following on democracy, let us assume that law-making at the EU level would fully comply with national patterns of democratic law-making.56 Transferring powers, nevertheless, fundamentally changes democracy at the national level at least by curtailing its scope. What is transferred to the EU happens according to ‘EU democracy’ and not according to national democratic rules. Democratic majorities would be achieved among the representatives of the peoples of all EU Member States, not only by the majority in one of them. Otherwise, unanimity at the EU level would have to be mandated and majority decisions would be illegal from the outset. This inevitable tension flowing from participation in EU integration impacts on all of the abovementioned national values. It is an important reason for the difficulties of adjudication in cases of conflicts. b)  Limits Flowing from the Democracy Principle The democracy principle carries a specific weight among all aforementioned principles. Again, in Germany the Constitutional Court introduced a twofold kind of Solange formula in its Lisbon ruling. The principle of conferral and the maintenance of an ‘equal balance’ in the distribution of competences between the EU Member States and the EU assume a primary role.57 The parliament (Bundestag) must, according to the BVerfG, retain functions and powers of substantial importance,58 and the German legislature must only consent to transfers of competences and treaty amendments whose effects are foreseeable.59 Even if this approach received severe criticism in academic writing, it is crucial for the further transfer of powers through Treaty amendments in Germany. In almost all other countries, the constitutional text and respective rulings of (constitutional) courts are less detailed and less far-reaching. However, preserving democracy is of constitutional relevance everywhere. Given that all EU Member States are based on representative democratic structures where law-making is primarily done by Parliament, transferring law-making competences to the EU inevitably affects parliamentary powers and, thus, democracy. It is not crucial whether the constitutional texts allowing for the transfer of powers to the EU include an explicit reference to the preservation of democracy (like in Finland) or not (like in Austria, the Czech Republic and Poland), or refer in a more general manner to constitutional principles including democracy (like in Sweden). As a consequence, the transfer of (additional) powers to the EU everywhere comes under the respective (dynamic)60 opening clauses and their restrictions. This 56 Which it is not. See the still valid and balanced account in Peters, Elemente (n 44) 626 ff. 57 BVerfG 30.06.2009, 2 BvE 2/08 (Lisbon) BVerfGE 123, 267, ECLI:DE:BVerfG:2009:es20090630.2bve000208, para 272: ‘As long as European competences are ordered according to the principle of conferral in cooperatively shaped decisionmaking procedures, and taking into account state responsibility for integration, and as long as an equal balance between the competences of the Union and the competences of the states is retained, the democracy of the European Union cannot, and need not, be shaped in analogy to that of a state.’ To mention only one out of dozens of country specific puzzles: how this stance of the BVerfG is to be reconciled with the requirement flowing from Art 23(1) of the Basic Law that the Union is committed to the principle of democracy is not self-evident in its details. 58 BVerfG 2 BvE 2/08 (Lisbon) (n 57) para 246. 59 Ibid, para 236: ‘The integration programme of the European Union must be sufficiently precise. In so far as the people itself is not directly called upon to decide, democratic legitimation can only be achieved by means of parliamentary responsibility … A blanket empowerment for the exercise of public authority, in particular one which has a direct binding effect on the national legal system, may not be granted’. 60 Compare text above close to n 41.

Squaring the Circle – How National Constitutions and EU Law Interact  811 means that additional amendments to the Constitution are not required in each case. However, qualified majorities principally modelled after the requirements foreseen for constitutional revision are mandated. This is the case eg in Austria, Greece, Finland and Poland. Frequently, however, such additional transfer immediately triggers the debate whether or not it would respect the limits flowing from the Constitution, or whether additional amendments would be necessary, or even if they are excluded because of ‘eternity clauses’. Almost certainly some restrictions go undetected until today, in the absence of actual political relevance – as for a long time had been the case with ‘budgetary sovereignty’ (see below). Examples for an arguably augmented sensitivity – stricter constitutional limits – are Ireland and France. Ireland operates on the basis of a ‘static authorisation’ of EU membership,61 which seems to necessitate a modification of the Constitution in each case of additional competence transfer. In France, it is the notion of ‘sovereignty’, which impacts deeply on the constitutional limits to any further integration steps. Granting additional competences (to those already transferred) to the EU always touches upon the sensitive realm of ‘sovereignty’. Consequently, competences, similarly as sovereignty, are not ‘transferred’ to the EU but ‘shared’ with other EU Member States. Further amendments of the Constitution will be required for any power sharing, jeopardising the ‘fundamental conditions of the exercising of sovereignty’ either because such authorisations (i) do not relate to those already permitted in the Constitution or because (ii) there are modifications of the exercise of competences already transferred (eg, replacing unanimity in the Council by qualified majority voting). A transfer of competences, by contrast, appears to be beyond the reach of ‘normal’ constitutional amendments. A takeaway from the financial and sovereign debt crisis after 2008 is that budgetary decision-making autonomy – sometimes dubbed as parliamentary ‘budgetary sovereignty’ – forms an important element of the EU Member States’ constitutional guarantees of democracy. It is deemed a matter of national parliamentary decision-taking excluded from a transfer to the EU, at best open for ‘external’ restrictions that have to be considered when deciding on the national budget.62 Such findings were not strictly limited to EU Member States with unequivocal constitutional wording. Rather, the conclusion was drawn from a functional reading of representative democracy that would require a determination of the core aspects of national budgets by the directly elected national Parliament.63 During the crisis, challenges concerned mainly the Treaty on Stability, Coordination and Governance in the Economic and Monetary Union (TSCG) and the European Stability Mechanism (ESM). Both of these treaties include substantial restrictions to ‘budgetary sovereignty’ at the national level. Nevertheless, no breaches of the constitutional guarantees were found by national Constitutional or Supreme Courts.64 This was probably

61 Compare text above after n 41. 62 Compare for the following Griller, ‘Analytical Report’ (n 8) 15 ff, 25 ff, with further references. 63 See BVerfG 2 BvE 2/08 (Lisbon) (n 57) para 256: ‘A transfer of the right of the Bundestag to adopt the budget and control its implementation by the government which would violate the principle of democracy and the right to elect the German Bundestag in its essential content would occur if the determination of the type and amount of the levies imposed on the citizen were supranationalised to a considerable extent. The German Bundestag must decide, in an accountable manner vis-à-vis the people, on the total amount of the burdens placed on citizens. The same applies correspondingly to essential state expenditure … Budget sovereignty is where political decisions are planned to combine economic burdens with benefits granted by the state … What is decisive, however, is that the overall responsibility, with sufficient political discretion regarding revenue and expenditure, can still rest with the German Bundestag.’ 64 Eg by the High Court of Ireland IEHC 17.07.2012, [2012] IEHC 296 (Thomas Pringle v Government of Ireland, Ireland and the Attorney General) which found that the ESM did not diminish the budgetary, fiscal or economic sovereignty of Ireland (which would have required a positive referendum). In the same vein the BVerfG in BVerfG 07.09.2011, 2 BvR 987/10 (Euro-Rettungsschirm) BVerfGE 129, 124, ECLI:DE:BVerfG:2011:rs20110907.2bvr098710.

812  Stefan Griller and Lina Papadopoulou motivated by judicial self-restraint trying to avoid aggravating economic turmoil. However, eg in Austria, Estonia, Germany and Ireland, the courts, and other scrutinising bodies in countries like Finland (the Finnish Constitutional Law Committee) and the Netherlands (the Dutch Council of State) made it clear at that occasion that dismissing the claims resulted from complex reasoning based on the importance of ‘budgetary autonomy’. Procedural impediments were partly decisive. However, when it came to the substance, dismissing the action sometimes was a close decision and at any event ‘only’ a matter of degree. The invoked guarantees of the national parliaments’ budgetary decision-making autonomy and of budgetary responsibility, by contrast, were confirmed and accentuated as public goods protected by the national constitutions. It is fair to conclude that ‘budgetary sovereignty’ will remain relevant to any future EU reform scenarios and would become pertinent specifically for the creation of a Fiscal Union. c)  Compensating for the Loss of Parliamentary Powers EU primary law provides for two major mechanisms compensating for the loss of national parliamentary power following the transfer of competences to the EU. First, ‘the Union shall be founded on representative democracy’, where citizens are ‘directly represented at Union level in the European Parliament’, and EU Member States in the European Council and in the Council ‘by their governments, themselves democratically accountable either to the national Parliaments, or to their citizens’.65 This is the well-known double legitimation of EU democracy. As regards the European Parliament, the Treaties in detail spell out its composition, its powers and its procedures. All of this is pertinent, in all EU Member States, for the constitutional justification of transferring powers to the EU. It ensures that such transfer is not inevitably a drastic diminution of or even a carve-out from democratic law-making. This is crucial for securing the national principle(s) of democracy as a constitutional value also while being an EU Member State. Regulating EU democracy cannot, however, prevent that national democracy gets substantially altered in each country acceding to the EU.66 Furthermore, law-making at the EU level is seen as democratically deficient,67 giving EU democracy less compensatory potential for the loss of democracy at the national level and, consequently, reducing the openness for additional transfer of competences. This seems to be the legal nucleus of the political battle on the preferences in allocating power rather with the EU or with the Member States. Even if not explicitly, the debate is ‘loaded’ with the controversy on safeguarding democracy in a supranational Union. Furthermore, it appears to be controversial to what extent enhancing parliamentarism at the EU level could further compensate for the loss of national competencies and, thus, national parliamentary power. The argument runs that enhancing democracy at the EU level – which would mean introducing the strict ‘one (wo)man one vote’ principle and increasing law-making competences – would at the same time endanger legitimisation through national parliaments.68 65 Art 10 TEU, paras 1 and 2. 66 Compare text above close to n 56. 67 Eg and specifically voiced by the BVerfG in BVerfG 2 BvE 2/08 (Lisbon) (n 57) para 271. The Court says – in a state-centred approach – that the Council cannot be qualified as a second chamber, that the European Parliament is a representative body of the peoples which need not comply to the equal political rights to vote of all citizens, and that the Commission is not fully accountable. 68 In BVerfG 12.10.1993, 2 BvR 2134 (Maastricht) BVerfGE 89, 155, 175 ECLI:DE:BVerfG:1993:rs19931012.2bvr213492, the BVerfG conceded that the European Parliament could be strengthened if it were elected on the basis of a uniform electoral law in all member states and if its influence were to increase. However, it continued: ‘The important factor is that the democratic foundations upon which the Union is based are extended concurrent with integration, and that a living democracy is maintained in the Member States while integration proceeds. If too many functions and powers were placed

Squaring the Circle – How National Constitutions and EU Law Interact  813 This, one might conclude, would violate the inalienable guarantee of national democracy and probably even national statehood. The paradoxical result of such reasoning is that democracy at the EU level is seen as deficient, which makes it a borderline case when safeguarding democratic standards. At the same time, it is bound to remain deficient to avoid a conflict with other core elements of constitutional identity such as national democracy and statehood.69 In essence, this would mean that increased parliamentary decision-making at the EU level in principle cannot compensate for a further loss of parliamentary powers at the national level. Such development would fall short of safeguarding democracy. Surely this is an issue that in some EU Member States is cautiously avoided. Others would yet disagree and embrace enhanced parliamentarism at the EU level including extending the European Parliament’s powers as fully constitutional. The second ‘remedy’ under EU law comes from the Protocols to the Treaties. Protocol No 1 to the Treaties on the Role of National Parliaments in the EU foresees the participation of these parliaments in the EU law-making by sending reasoned opinions to draft legal legislative acts. Protocol No 2 on the Application of the Principles of Subsidiarity and Proportionality allows for the sending of reasoned opinions considering draft legislative acts allegedly being in conflict with the principle of subsidiarity. The law-making organs of the EU (mainly the European Parliament, the Council and the Commission) have to ‘take account’ of such reasoned opinions. In cases of qualified objections from national parliaments70 the obligations flowing from the protocol are satisfied if the draft is being reviewed and reasons are given either to amend, withdraw, or maintain it. No obligation to repeal is foreseen. Consequently, the national parliaments do not dispose of the power to block or stop a law-making procedure at the EU level, neither alone nor together. It is always the EU organs that have the last say. Consequently, these p ­ rotocols do not seem to figure prominently anywhere in national conversations on the constitutional justification to transfer competences to the EU. Third, in many countries (eg Austria, Denmark, Finland, Germany and Ireland) compensation for the ‘loss of national parliamentarism’ is sought in reinforced supervision and control of the conduct of the executive, notably the government, when it participates in decision-taking at the EU level, mainly in the Council or the European Council, but also in closely related environments like the ESM. These very diverse mechanisms go beyond governmental accountability vis-à-vis the Parliament and include the possibility to pass instructions binding the government in its EU activity. In addition, parliamentary consent advancing governmental action might be mandated where EU law would allow for ‘autonomous’ executive action.71 If such ‘responsibility

in the hands of the European inter-governmental community, democracy on the level of the individual States would be weakened to such an extent that the parliaments of the Member States would no longer be able to convey adequately that legitimation of the sovereign power exercised by the Union’. 69 Later, the BVerfG excluded the development of the EU into a fully-fledged democracy under the Basic Law and its Art 79(3), but left the possibility to leave that avenue nevertheless open, if only theoretically; see BVerfG 2 BvE 2/08 (Lisbon) (n 57) para 263: ‘If however, the threshold were crossed to a federal state and to the giving up of national sovereignty, this would require a free decision of the people in Germany beyond the present applicability of the Basic Law and the democratic requirements to be complied with would have to be fully consistent with the requirements for the democratic legitimation of a union of rule organised by a state. This level of legitimation could no longer be prescribed by national constitutional orders.’ 70 Whenever a quorum of one third of all parliamentary votes contends non-compliance with subsidiarity, the draft must be reviewed. If, under the ordinary legislative procedure, the non-compliance opinions represent a simple majority of all parliamentary votes, and the Commission wants nevertheless to maintain the proposal, it has to submit its reasoned opinion to the Union legislator. Both the Council (by a 55% majority of its members) and the Parliament (by simple majority) could stop the legislative procedure. 71 A prominent example is Art 48(7) and (8) TEU, the simplified Treaty revision procedures. For the sake of clarity, it shall be added that the first mentioned paragraph would not allow for increasing EU competences. Another one is the decision on the granting and the conditions of loans to beneficiary countries under the ESM.

814  Stefan Griller and Lina Papadopoulou for integration’ is effectively possible and exercised by the national parliament, otherwise unconstitutional transfers of power might be justifiable. It goes without saying that such mechanisms bring supranational (majority) decision-taking by EU organs a step back to traditional international law where international agreements need parliamentary approval before ratification. The flipside is increased participation of national parliaments entailing enhanced democratic legitimacy stemming from them. Some countries, specifically those with a tradition of a strong executive (eg France), however, do not seem to bother too much when it comes to counterbalancing the transfer of competences to the EU by internal power-sharing between Parliament and Government. Altogether, the three ‘compensatory’ elements produce a very nuanced and delicate balance of powers between national parliaments and EU democracy on the one hand, and national parliaments and national governments on the other hand. Unsurprisingly, diversity among EU Member States is enormous, not only concerning the details, but also the weight of the respective arguments in the constitutional discourse.

4.  Inalienable Core of the Constitution The preservation of democratic decision-making powers at the national level, in some countries is further specified and thereby closely intertwined with another substantive barrier: the preservation of ‘national identity’ or ‘core competences’, which includes the preservation of national statehood or sovereignty. The difference to the ‘normal constitutional barriers’ would be that conventional amendment procedures for the Constitution – what we might call ‘simple constitutional amendments’72 – would not be sufficient and could not legitimise such a change. Having said that, it might be that the creation of a European federation at the expense of national statehood would be unconstitutional – in the specified meaning – in almost all EU Member States. In many of them, however, there is not even much of a debate on this issue in the legal circles.73 For some, guarantee for statehood is included in the notion of ‘sovereignty’ (as in the Czech Republic, France, Finland, Germany, Italy and Poland), for others, the same is true even if the term is avoided (as in Austria and Sweden). Moreover, also transferring or impairing core competences, or impairing internal principles like the federal organisation of the state or the autonomy of regions might be excluded in a similar manner. It is a matter of thorough interpretation of the respective Constitution which aspects could come under such restraint.74

72 According to the provisions specified above in the text close to and after n 16. 73 Compare however BVerfG 2 BvE 2/08 (Lisbon) (n 57) para 228: ‘The Basic Law does not grant powers to bodies acting on behalf of Germany to abandon the right to self-determination of the German people in the form of Germany’s sovereignty under international law by joining a federal state.’ 74 According to the BVerfG 2 BvE 2/08 (Lisbon) (n 57) para 249 there must be a ‘sufficient space … for the political formation of the economic, cultural and social living conditions.’ This includes shaping ‘the citizens’ living conditions, in particular the private sphere of their own responsibility and of political and social security, protected by fundamental rights, as well as to political decisions … Essential areas of democratic formative action comprise, inter alia, citizenship, the civil and the military monopoly on the use of force, revenue and expenditure including external financing and all elements of encroachment that are decisive for the realisation of fundamental rights, above all in major encroachments on fundamental rights such as deprivation of liberty in the administration of criminal law or placement in an institution. These important areas also include cultural issues such as the disposition of language, the shaping of circumstances concerning the family and education, the ordering of the freedom of opinion, press and of association and the dealing with the profession of faith or ideology.’ This is spelt out in even greater detail in the judgment, which seems to be unique among all Member States.

Squaring the Circle – How National Constitutions and EU Law Interact  815

B.  Conditions for the Further Transfer of Powers As is apparent from the above, core values and competences identifiable within national constitutional law at the same time can be seen and almost everywhere are actually treated as barriers to further integration. Several possibilities have to be differentiated. First, there is a margin of discretion for the further transfer of powers that would not require constitutional amendments at all. This may even include opening up new fields of activity for the EU, but also bringing matters from unanimity to qualified majority voting in the Council.75 While this is to a certain extent possible, albeit not in all Member States, the national requirements differ considerably. A few examples shall suffice to demonstrate that the issue needs thorough interpretation in each Member State and for each case. Austria provides one of them. The Constitution allows in principle for amendments to the EU Treaties including competence transfers by specific ‘constitutional majorities’, namely a two-thirds majority in both chambers of Parliament and without amending the Constitution. Simplified treaty amendments, eg on the basis of the general bridging clause76 or specific bridging clauses would require the approval, by the same qualified majorities, of parliament not only after, but even before the (draft) decision by the European Council can be taken. Similar is the situation in Finland. In Germany, things are ‘easier’. It appears as Constitution-abiding to amend the EU Treaties including a transfer of powers to the EU within certain less ‘sensitive’ fields by ordinary legislation (Zustimmungsgesetz), meaning by consent of the two chambers of Parliament. This is specifically remarkable insofar as such transfer might affect the division of competences as foreseen by the Basic Law. Nevertheless, no explicit amendment of the text of the Basic Law is required. Simplified treaty amendments come under similar ex-ante requirements as in Austria – here on the basis of a so-called ‘integration responsibility’ of the German parliament for actions of the executive. The respective parliamentary decision, nevertheless, could still be taken by a simple majority. The situation differs between EU Member States, yet hardly in its complexity. In France, the mentioned additional transfer (‘sharing’) of competences will always require further amendments of the Constitution, as will the transition from unanimous decision to qualified majority in the Council. Abandoning national sovereignty by acceding to a new state is not even a topic – it is simply beyond the mainstream of the debate. Second, further transfers or transfers in new areas might require (previous) amendments of the Constitution, again other changes could be entirely excluded.77 This is a scenario that is not limited to the countries mentioned above, even if discussion elsewhere might be scarce. Taking the example of the Lisbon treaty: the need for constitutional change was suggested in several countries, Austria, Germany and Ireland among them. While in Austria the matter was politically settled and parliamentary approval by qualified majority achieved, it needed the BVerfG in Germany to arrive at a similar result. By contrast, Ireland made a constitutional amendment, and even rejected the Treaty by a referendum in the first attempt to pass it. Third, the issue of the ‘inalienable core’, which would need to remain untouched by any amendment to the Constitution (‘eternity clause’), is equally complex. In Austria, for example, there is no such thing. Any constitutional change could be performed by a so-called total revision of the Constitution, which would need a constitutional law passed by the two chambers of Parliament by a two-thirds majority followed by a mandatory referendum. Arguably even integrating Austria

75 Eg

based on Art 48(7) TEU, the general bridging clause. 48(7) TEU. n 74.

76 Art 77 See

816  Stefan Griller and Lina Papadopoulou into a federal European state would thereby be possible. However, the flipside is that total revisions which are not limited to joining a new state, need a super qualified amendment procedure. Even in Germany with its famous ‘eternity clause’78 things are far from being settled. While the BVerfG in its Maastricht judgment appears to have excluded any possibility to overcome the boundaries of that clause,79 the Lisbon judgment reads differently. Here, the BVerfG made a clear reference to Article 146 Basic Law which provides that the Law ‘shall cease to apply on the day on which a constitution freely adopted by the German people takes effect’. The Court, in turn, and without explicitly mentioning Article 146 Basic Law conceded that accession to ‘a federal state and … giving up of national sovereignty … would require a free decision of the people in Germany beyond the present applicability of the Basic Law.’80 Even if some consider this merely a theoretical possibility, it, nevertheless, means that the eternity clause is not as ‘eternal’ as one could assume. Without any doubt: the more important an EU Treaty change, the more central to a state’s identity the competence to be transferred or impaired, the more likely is the existence of a constitutional impediment to further integration. Differences in the respective requirements for constitutional amendments81 become crucial also regarding interpretative efforts and motivation to establish the need for a change, or to avoid it. Where, as in Bulgaria, the hurdles for constitutional amendments are deemed insurmountable, such impediment is, depending on the political estimation of a tabled proposal to change the EU Treaties, either used to discredit it as ‘unconstitutional and unfeasible’, or to restrict the scope of the burdensome amendment requirements. Seen from this angle, the difference between ‘ordinary’ constitutional amendments and overcoming ‘eternity clauses’ might lose importance. An ‘ordinary’ constitutional change clause which is, in a given country, as insurmountable as an eternity clause might produce similar effects. This might have repercussions on the ultra vires debate. It is of specific concern and regularly producing complications if for whatever reason government and Parliament decide to approve a Treaty change (or accession to the EU), while at a later stage, when dealing with secondary EU legislation, a Constitutional or Supreme Court arrives at the conclusion that an amendment of the Constitution would have been mandatory. It might not be easy to correctly differentiate in such a case: did the legislator by approving the Treaty overstep its mandate, or was it the EU, or both? Procedural differences between the Member States may also be of importance. In general, amendments to EU Treaties that arguably require constitutional amendments in an EU Member State provoke a political debate that is, in a first step, settled during the parliamentary approval procedure. In preparing such approval, governments and possibly experts make their case for or against the need for an amendment. Once Parliament decides to approve the Treaty without a constitutional amendment, ratification is performed which renders the Treaty binding provided that all other Member States equally ratify. Challenges to the constitutionality may, nevertheless, arrive later in court cases. In some countries, eg Austria and Germany, the Constitutional Courts may terminate the application of a treaty if it is found to violate the Constitution. In some countries, however, eg France and Germany, the Constitution provides also for ex-ante-scrutiny of treaties before ratifying them. Dealing with verdicts of unconstitutionality is easier at an earlier stage. If the court finds contradictions to the Constitution, as has frequently 78 Art 79(3) Basis Law, compare above n 51. 79 BVerfG 2 BvR 2134/92 (Maastricht) (n 68). 80 BVerfG 2 BvE 2/08 (Lisbon) (n 57) para 263. A few paragraphs earlier (216) it said: ‘The Basic Law … not only presumes sovereign statehood for Germany but guarantees it.’ 81 Compare above after n 16.

Squaring the Circle – How National Constitutions and EU Law Interact  817 happened in France, the duty results to either change the Constitution, or to abstain from ratifying the treaty. These differences in procedure may impact heavily on the dynamics of the national debate.

C.  Scrutiny of Secondary Legislation, Especially Ultra Vires Doctrine 1.  Introductory Remarks Ultra vires in general is an act, policy, measure or duty to act outside the powers of the acting organ or legal entity (including public authorities, states, or international organisations).82 This entails that the acting organ is overstepping its mandate. The prevailing view holds that in the absence of an institutionalised control mechanism, ultra vires measures are illegal, in case of general or individual norm of law, not binding and not to be applied. This might be mitigated or modified in the specific context. The most important options are acceptance by subsequent practice, or provisions introducing different consequences like damages. This is different in cases where legal remedies such as scrutiny by courts are available. Under such a condition, flawed, including ultra vires, measures can be nullified by the court(s), they are voidable. It is an almost theoretical but not totally excluded possibility that even in such a case ultra vires could mean void, seen from the angle of the addressees, in our case Member States of an organisation. However, the conditions differ. Establishing legal control through a court necessarily implies a margin of appreciation including possible errors for that court. Otherwise, it could not fulfil its function to settle disputes, if the parties to this dispute could, by referring to the ultra vires doctrine, continue questioning and disrespecting the obligations as if there was no binding court decision. Ultra vires in such a setting implies two errors: first, the error committed by the organ that took the measure, second, the error by the court (or the court of last instance) not to nullify the measure. The court has the ‘benefit’ of a margin of error (Fehlerkalkül). This means that in principle also flawed decisions of the court of last instance are binding, in order to meet the peacekeeping purpose of entrusting a court to decide the dispute. Only if the court oversteps even the limits of this margin of error, it might be possible to challenge its findings. Its errors must be grave and evident to bring the decision outside the margin of error. What would be at stake is claiming two measures of being void: the one scrutinised by the court, and the court decision upholding it. Arguably, this is the situation in the EU where Article 19 TEU establishes the CJEU and entrusts it with the power to ‘ensure that in the interpretation and application of the Treaties the law is observed’. It is primarily for the CJEU to decide, among other things, on the legality of secondary (and tertiary) EU law including decisions of other EU institutions and bodies. The Member States are bound by its judgments. Only if they were extremely flawed and as a result outside the margin of error, might it be arguable to question the binding force from the angle of an EU Member State.83 82 From the vast literature compare only some classic authorities: Elihu Lauterpacht, ‘The Legal Effect of Illegal Acts of International Organisations’, (1965) 64 Cambridge Essays in International Law 88; Henry Schermers and Niels Blokker, International Institutional Law, 6th edn (Nijhoff, Brill, 2018) §§ 206 ff; Karl Wengler, ‘Recours judiciaire à instituer contre les décisions d’organes internationaux, Rapport et project de Résolutions’, (1952) I Annuaire de l’Institut de Droit International 224. On one of the co-authors’ view on the issue, including methodological aspects, see Stefan Griller, Die Übertragung von Hoheitsrechten auf zwischenstaatliche Einrichtungen (Wien/New York, Springer, 1989) 449 ff with further references. 83 On alternatives see text below near n 196.

818  Stefan Griller and Lina Papadopoulou Some confusion might arise from situations where EU Member States entered into the Union or approved Treaty changes without, for whatever reason, amending their constitutions accordingly, as required. Consequently, the powers of the EU might exceed what the national Constitution allows for. In such a case, ultra vires is not the EU measure but the transfer of powers by the national government and legislator. Seen from the angle of national law, nevertheless it renders the EU measure illegal. However, the violation concerns not a disrespect of the Treaty, but a disrespect of the Constitution by concluding the Treaty. Sometimes, these issues get mixed up, which might even lead to untenable verdicts. As a matter of principle, ultra vires can be alleged against any EU measure or CJEU judgment. However, it is of specific salience in cases where the norm in question at the same time allegedly violates principles of the national Constitution. Separating the closely interwoven issues who overstepped the mandate may be a challenge. It is not excluded either that transfer of powers was beyond constitutional authorisation and enacting secondary EU law in addition overstepped the authorisation of the Treaties. These possibilities may render controversies complex. To give an example: let us assume for the sake of argument that the ECJ at the time was right that fundamental rights protection was not available under the Treaties.84 It could consequently not annul EU law (at the time EC law) for violating fundamental rights. This would have meant that entering into the EU with directly applicable law taking precedence over national law would have exempted an important and voluminous part of the legal order from fundamental rights control. This would have been a violation of the national constitutions. Access to the EU would have been forbidden. For any national court subsequently dealing with a concrete case of EU law disrespecting a fundamental right being guaranteed by the national Constitution this creates a temptation, not to say the ‘duty’ to set EU law aside.85 The reason, however, could only be that the national Constitution in such a case prevails over EU law but not that the EU overstepped its mandate. The situation is entirely different if we assume – or today can take for granted – that the CJEU has to enforce the protection of fundamental rights which are ‘essentially comparable’86 to the level of protection under national law. In such a case, the CJEU would doubtlessly have to scrutinise alleged violations.87 In such a setting, ultra vires EU action is imaginable, even if only more theoretically, namely the enactment of secondary EU legislation or decisions violating fundamental rights which would, nevertheless, get declared legal by the CJEU in overstepping the mentioned margin of error. What should have been demonstrated by taking fundamental rights as an example is valid also for other protected values, namely democracy, transfer of competences etc.

2.  Evolution within the EU a)  Introductory Remarks Alleged ultra vires acts of the EU are a traditional topic not only in the academic debate on the EU, but also in the jurisprudence of several national courts. For decades, it has been an implicit but never exercised ‘threat’ to (permanently) disapply secondary Union legislation within the 84 Above n 47. 85 If national law foresees, the application of the Treaty within the national legal order may have to be terminated before the concrete fundamental rights violation could be rectified. This termination could be done in an intermediate procedure by a Constitutional Court. 86 Above n 51. 87 In general, this would be done on the basis of Art 265 TFEU (nullification procedure) or Art 267 TFEU (preliminary ruling procedure).

Squaring the Circle – How National Constitutions and EU Law Interact  819 legal order of an EU Member State. As such, it has been exerting a certain preventive function which might have influenced daily practice. It may be fair to conclude that it was the BVerfG that explicitly started the debate at the level of constitutional or otherwise highest courts. Later, others followed its approach, specifically in the Czech Republic, Denmark and Poland.88 Partly, they ‘overtook’ the BVerfG by declaring judgments of the CJEU void. In 2020, also the BVerfG, for the first time in history, declared an European Central Bank (ECB) decision as well as the respective CJEU judgment ultra vires and consequently null and void within the German legal order.89 This judgment, controversial as it may be, could be a game-changer in the relationship between EU law and national law.90 The landscape is full of different flowers. The difference between Member States equipped with a Constitutional Court empowered with exercising ex-post review of subordinated law against the yardstick of the Constitution, and EU Member States without such a mechanism might be influential. The spelling out and concretisation of constitutional barriers is more explicit and visible in such countries like Germany, the Czech Republic, (nowadays also) France, Italy and Poland, compared to others like the Netherlands, Finland and Sweden. This is not to say that the topic is less important for the latter. However, judicial ex-post review in these countries is decentralised and traditionally limited to the correction of ‘manifest’ or ‘obvious’ errors. In addition, it is often dubious whether EU law might come under the reviewable measures as specified by the national Constitution. This renders an open conflict less likely. b)  Fundamental Rights Protection Regarding fundamental rights protection, the landscape has totally changed since the beginning of the controversy in the 1970s.91 Today, fundamental rights protection within the EU is explicitly guaranteed by the Charter, the standards of the ECHR and the constitutional traditions common to the Member States.92 Protecting all of these rights forms part of the CJEU’s core competences. It is hard to argue that on this level of the general law as it stands, fundamental rights protection in the EU would be second to any of the Member States constitutions. By contrast, individual decisions might still become controversial. This is why they reach the supreme courts. Thorough differentiation between the general standards and such individual decisions should be observed. However, this is not always the case. If fundamental rights protection within the EU generally is at least ‘essentially comparable’ to the protection guaranteed by national constitutions, the yardstick for ultra vires decisions cannot be the same. It should start from the level of protection within the EU and ask the question whether EU fundamental rights as specified above were gravely and manifestly violated beyond the margin of error accorded to the CJEU.93 If, by contrast, the constitutional duty to oblige the EU to respect the ‘essentially comparable’ level of protection should not have been met in the act of transferring powers, national courts would have to start at the national act of transfer when looking for ultra vires.

88 Compare the brief overview in Mayer/Wendel, ‘Grundlagen’ (n 1) paras 239 ff. 89 Below in the text after n 113. 90 The open refusal of the Czech Constitutional Court (below n 99) and the Danish Supreme Court (below n 136) to accept judgments of the CJEU is, in principle, of equal gravity. However, they concerned peculiar aspects of compliance with EU law, while the BVerfG’s verdict is of a very principled manner. 91 Above n 47 and the following text. 92 Art 6 TEU. 93 Above in the text after n 82.

820  Stefan Griller and Lina Papadopoulou The approach taken by the BVerfG comes close to such a position without clearly stating the difference. Constitutional complaints and submissions by courts, which challenge secondary EU law, are inadmissible from the outset, unless the grounds for a submission by a national court of justice or of a constitutional complaint which puts forward an infringement by secondary European Community Law of the fundamental rights guaranteed in the Basic Law … state in detail that the protection of fundamental rights required unconditionally by the Basic Law is not generally assured in the respective case. This requires a comparison of the protection of fundamental rights on the national and on the Community level.94

Thus, when it comes to scrutiny of individual cases, admissibility depends on demonstrating that the constitutional standards are ‘not generally assured in the respective case’. This involves two stages; first, determine whether ‘essentially comparable’ protection generally is no longer guaranteed (which appears to be difficult to prove, as mentioned) and, second, demonstrate an ultra vires violation of the EU standards. It is not entirely clear, however, whether this kind of reasoning would form two different stages, or whether the yardstick for the second one would be the same as for the first. Be that as it may, the ‘German approach’ has become popular across Europe among other supreme Member States’ courts. Several of them (in France,95 Poland96 and Spain97) embraced the approach without nullifying the EU law under scrutiny. The Danish Supreme Court98 as well as the Czech Constitutional Court99 went one step further and, without much ado, denied the binding force of CJEU judgments allegedly beyond the CJEU’s competence. It is likely that the BVerfG’s PSPP judgment of 5 May 2020 – even if it is not on fundamental rights standards but on competences – will stimulate this development, to the detriment of EU unity. The decision of the Czech Constitutional Court100 might be an example for ‘false friends’. The Court refers to the German Solange jurisprudence. However, the yardstick it applies – in a case that is dealing, amongst others, with the equality principle – hardly follows the respective standards. The court more or less dismisses the CJEU’s reasoning by replacing it by its own deliberations and preferences. There is neither any closer scrutiny on whether or not equality is protected by the CJEU in an ‘essentially comparable’ manner nor detailed reasoning on why the CJEU should have gravely overstepped its mandate. c) Competences Contrasting to the developments regarding fundamental rights protection, the general tension between safeguarding national competences and the use of transferred competences by the EU101 94 BVerfG 07.06.2000, 2 BvL 1/97 (Bananenmarktordnung) BVerfGE 102, 147, ECLI:DE:BVerfG:2000:ls20000607. 2bvl000197. 95 Conseil Constitutionel, Decision 2006–540 DC of 27 July 2006 (Droit d’auteur) ECLI:FR:CC:2006:2006.540.DC. 96 Trybunał Konstytucyjny 16.11.2011, Case No SK 45/09, see https://trybunal.gov.pl/fileadmin/content/omowienia/ SK_45_09_EN.pdf, 20 ff. 97 Tribunal Constitucional de España 13.12.2004, Declaración 1/2004 ECLI:ES:TC:2004:1D. 98 Højesteret (Supreme Court Denmark) 06.12.2016, Case 15/2014 (Ajos). An unofficial and informal translation is available at www.supremecourt.dk/., 45ff. Compare below n 136. 99 Ústavní soud ČR 31.01.2012, Pl. ÚS 5/12 (Slovak Pensions XVII) ECLI:CZ:US:2012:Pl.US.5.12.1, N 24/64 SbNU 237 (www.usoud.cz/en/decisions/2012–01–31–pl-us-5–12–slovak-pensions); the judgment ‘discards’ the judgment CJEU 22.06.2011 C-399/09 (Marie Landtová v Česká správa socialního zabezpečení) ECLI:EU:C:2011:415 and declares it irrelevant. 100 Compare the contribution of Tichý/Dumbrovský in this volume. For a very critical review compare also Georgios Anagnostaras, ‘Activation of the Ultra Vires Review: The Slovak Pensions Judgment of the Czech Constitutional Court’, (2013) 14 German Law Journal 959. 101 Above text after n 57.

Squaring the Circle – How National Constitutions and EU Law Interact  821 by definition could not decrease. Rather, it can increase, given that almost each Treaty amendment in the past entailed the transfer of additional competences to the EU. What is the same, though, is the necessity to keep, as far as legal consequences are concerned, possible errors committed by national governments and parliaments possibly not fully respecting the restrictions flowing from their Constitution, on the one hand, and alleged ultra vires measures of the EU being backed by the CJEU102 apart, on the other hand. The European landscape might have been changed by the recent PSPP judgment of the BVerfG.103 Therefore, and as an exception to the rule, it deserves closer attention.

3. The PSPP Judgment – A Game-Changer? a)  ‘Crisis Management’ by the CJEU Both in Gauweiler104 and Weiss105 the CJEU had to deal with harsh criticism of the so-called non-

traditional measures of the ECB during the crisis. This critique was voiced by the BVerfG in its questions in the first preliminary ruling procedure it ever brought to the CJEU. Put in the form of such questions the BVerfG in essence alleged the ECB to overstep its mandate, to act ultra vires. At stake were the Outright Monetary Transaction (OMT) programme106 and the PSPP Decision by the ECB, the latter creating the so-called Quantitative Easing.107 It included massive interventions in the private and public bond market by purchasing covered bonds and assetbacked securities. The questions put by the Constitutional Court, under Article 267 TFEU, suggested that the CJEU should declare the ECB measures invalid. From the outset in both cases, the controversy was a typical dispute on the boundaries of EU competences and their enforcement by the CJEU and, as a matter of last resort, by the BVerfG as a national court safeguarding at the same time national, in this case German, competences. The CJEU, however, did not consider that the ECB had overstepped its mandate and upheld both measures. It did so by confirming the line of reasoning it had started in the Pringle case, namely that the determination whether a measure falls within the area of monetary policy is principally depending upon the objectives of that measure, while the instruments employed are only of secondary relevance.108 It held ‘that a monetary policy measure cannot be treated as equivalent to an economic policy measure for the sole reason that it may have indirect effects that can also be sought in the context of economic policy’.109 On this basis the CJEU confirmed that both the OMT programme and the Quantitative Easing mechanism remained within the powers of the ECB. They were also found to be proportionate110 and respecting the prohibition to grant overdraft facilities or any other type of credit facilities of Article 123 TFEU.111

102 Above n 93. 103 The judgment was handed down during the publication process of this book. The authors decided to go more into the details of the reasoning than with any other judgment dealt with in this report. 104 CJEU 16.06.2015 C-62/14 (Gauweiler) ECLI:EU:C:2015:400. 105 CJEU 11.12.2018 C 493/17 (Weiss) ECLI:EU:C:2018:1000. 106 CJEU C-62/14 (Gauweiler) (n 104). 107 ‘PSPP’ stands for ‘Public Sector Purchase Programme’, the most important part of the ECB’s non-traditional measures to fight the economic crisis, also labelled as ‘quantitative easing’. Starting in 2015, the ECB purchased up to €80 billion per month (2016 and 2017) to support economic stability and avert deflation. 108 CJEU C-62/14 (Gauweiler) (n 104) paras 46 ff; CJEU C 493/17 (Weiss) (n 105) paras 53 ff. 109 CJEU C-62/14 (Gauweiler) (n 104) para 52; CJEU C 493/17 (Weiss) (n 105) para 61. 110 CJEU C-62/14 (Gauweiler) (n 104) paras 66 ff; CJEU C 493/17 (Weiss) (n 105) paras 71 ff. 111 CJEU C-62/14 (Gauweiler) (n 104) paras 93 ff; CJEU C 493/17 (Weiss) (n 105) paras 101 ff.

822  Stefan Griller and Lina Papadopoulou b)  The PSPP Judgment of the BVerfG While the BVerfG accepted the CJEU’s verdict in Gauweiler, it disagreed in Weiss, with serious consequences for the future relationship between EU law and national law.112 In its so-called PSPP judgment,113 the BVerfG found that the Federal Government and the Bundestag violated the complainants’ constitutional rights by failing to take steps challenging the ECB’s PSPP decision. The latter was the foundation of the ECB’s programme for purchasing Member States’ bonds via secondary markets. The BVerfG declared both the ECB’s decision and the CJEU’s judgment upholding the decision as ultra vires acts114 depriving them of their ‘binding force in Germany’. An ultra vires act of the ECB or the CJEU is, according to the BVerfG, ‘not to be applied in Germany, and has no binding effect in relation to German constitutional organs, administrative authorities and courts.’115 The BVerfG considered that the ECB measure in essence violates the limits for ECB action, which is only empowered to conduct monetary policy (but not economic policy), or at least that both institutions failed to show in an intelligible way how these limits are respected. At first glance, the BVerfG continued with its decade-old yardstick that an EU act ‘violates the principle of conferral’ only if, in accordance with Article 48 TEU or the application of the ‘evolutionary clause (Evolutivklausel)’ in Article 352 TFEU,116 it is established that the violation of competences is sufficiently qualified. This requires that the act manifestly exceeds EU competences, resulting in a structurally significant shift in the division of competences to the detriment of the Member States … This is generally the case if the exercise of the competence in question … were to require a treaty amendment.

Such ‘ultra vires review must be exercised with restraint’ respecting that ‘interpretation and application of EU law, including the determination of the applicable methodological standards, primarily falls to the CJEU’.117 Despite this starting point, the BVerfG entered into what can only be qualified as a rather detailed scrutiny of the CJEU’s reasoning and ended with harsh criticism, declaring the CJEU’s judgment as ultra vires. This opened the way for a fresh evaluation of the ECB PSPP programme based on the principle of conferred powers and proportionality. First, according to the BVerfG the CJEU should have performed its proportionality test not only regarding the exercise of the ECB’s competence, but also for delimiting the monetary and economic policies. The BVerfG considered that the CJEU ‘manifestly fails to give consideration to the importance and scope of the principle of proportionality … which also applies to the division of competences’.118 Second, the BVerfG found that the CJEU disregarded the actual effects of the PSPP and, thus, did not properly apply the proportionality test. ‘Applied in this manner, the principle of proportionality cannot fulfil its corrective function for the purposes of safeguarding the competences of the Member States … The interpretation undertaken by the CJEU essentially

112 See, however, BVerfG 29.04.2021, Order 2 BvR 1651/15, ECLI:DE:BVerfG:2021:rs20210429.2bvr165115. 113 BVerfG 05.05.2020, 2 BvR 859/15 (PSPP), ECLI:DE:BVerfG:2020:rs20200505.2bvr085915. This concludes the preliminary ruling procedure that the GFCC had initiated which had led to the CJEU’s Weiss judgment. The English version of the judgment is available at www.bundesverfassungsgericht.de/SharedDocs/Entscheidungen/EN/2020/05/ rs20200505_2bvr085915en.html. 114 Ibid, paras 165, 154 and 163, respectively. However, it did so ‘only’ in the reasons given, not in the deciding parts of the judgment. 115 Ibid, para 234. 116 BVerfG 2 BvR 859/15 (PSPP) (n 113) para 110. Compare for the previous case law especially BVerfG 2 BvE 2/08 (Lisbon) (n 57) para 240 ff; 2 BvR 2661/06, 6 July 2010, Honeywell, BVerfGE 126, 286–33. 117 BVerfG 2 BvR 859/15 (PSPP) (n 113) para 112. 118 Ibid, para 119.

Squaring the Circle – How National Constitutions and EU Law Interact  823 renders meaningless the principle of conferral’.119 The BVerfG suggested that the CJEU’s judgment was ‘informed by the notion that a generous interpretation of the specific competence conferred may, to a certain extent, be compensated by a sound proportionality assessment’.120 As for the consequences, the BVerfG held that constitutional organs ‘have a duty to take active steps against the PSPP’, based ‘on their responsibility with regard to European integration [Integrationsverantwortung]’.121 They must … actively take steps seeking to ensure adherence to the European integration agenda [Integrationsprogramm] and respect for its limits … To this end, they must take suitable action to ensure adherence to the European integration agenda … The Federal Government and the Bundestag must clearly communicate their legal view to the ECB or take other steps to ensure that conformity with the Treaties is restored.122

Generally, this ‘also applies to the Bundesbank … Following a transitional period of no more than three months … the Bundesbank may thus no longer participate in the implementation and execution’ of the PSPP Decision.123 In the aftermath of this decision, the ECB Governing Council held a meeting on 3–4 June 2020 and published the most important results. The Bundestag, on 30 June 2020 passed a motion – partly based on confidential documents from the ECB provided via the Bundesbank – stating that it considered the given information as a sufficient step satisfying the requirements of the BVerfG’s judgment. Challenging this view, an order of execution was sought with the BVerfG. To the surprise of many, the BVerfG found the application inadmissible.124 While the decision was brief, and rightly so given that the requests went beyond what had been ordered in the PSPP judgment, the BVerfG explained at length that the applications were not only inadmissible but also unfounded. In essence it explained that the Federal Government and the Bundestag dispose of a wide range of measures to fulfil the requests and that the measures taken did indeed fulfil the judgment’s requirements.125 It also found that the ECB governing Council’s assessment of the proportionality of the PSPP had been sufficient.126 c)  Some Remarks on the PSPP Judgment The BVerfG, for the first time ever, declared an EU measure and the related CJEU judgment upholding it ultra vires and, therefore, inapplicable in Germany.127 The reasons the BVerfG 119 Ibid, para 123. 120 Ibid, para 128. 121 Ibid, para 230. 122 Ibid, para 231 f. 123 Ibid, para 235. 124 BVerfG 2 BvR 1651/15 (n 112). The abovementioned documents from the ECB and the Bundestag are to be found in the text of the order. For a first comment see Martin Nettesheim, Das Ende eines epochalen Verfassungsstreits: Darlegungsanforderungen der EZB, Integrationsverantwortung und der Beschluss desBundesverfassungsgerichts vom 29. April 2021, VerfBlog, 2021/5/18, https://verfassungsblog.de/das-ende-eines-epochalen-verfassungsstreits/. 125 BVerfG 2 BvR 1651/15 (n 112) para 108: ‘With these comprehensive measures, the Federal Government and the Bundestag took action in response to the judgment of 5 May 2020 and in exercise of their “responsibility with regard to European integration”, and thus did not violate the prohibition of insufficient state action [Untermaßverbot]. The measures taken do not fall so manifestly short of the requirements arising from the Court’s judgment that they could be considered essentially equivalent to complete inaction. With regard to fulfilling the requirements arising from the aforementioned judgment, the Federal Government and the Bundestag have a broad margin of (political) appreciation, assessment and manoeuvre’. 126 Ibid, para 100. Whether these findings are convincing measured by what the Court had expressed in the PSPP judgment is not to be deepened here. 127 From the rich debate on the judgment see only Jürgen Basedow et al, ‘European integration: Quo vadis? A critical commentary on the PSPP judgment of the German Federal Constitutional Court of May 5, 2020’, (2021) 19 International

824  Stefan Griller and Lina Papadopoulou provided go to the heart of the EU economic integration agenda by stipulating methods and details for drawing the borderline between monetary and economic policies in the EU. The BVerfG held that both the ECB and the CJEU gravely overstepped their respective mandates. This stance of the BVerfG begs the question, which is also at the centre of the decades-long tension between the CJEU and the supreme national courts: is the BVerfG encroaching upon the competences that were transferred to the EU level, or is it, by contrast, only safeguarding what legally remained a national competence, against grave errors committed by EU institutions? On a material level, we tend to share the view that it is the BVerfG that erred. While we would not challenge the methodological basis of the BVerfG’s reasoning,128 it is contended that the BVerfG disrespected its own standards of ‘residual review’. Already its starting point appears to be flawed. Article 5(1) TEU reads: ‘The limits of Union competences are governed by the principle of conferral. The use of Union competences is governed by the principles of subsidiarity and proportionality.’ Contrary to what the BVerfG claimed, the proportionality review is to be performed for the use of the EU competences, not for their delimitation. And this is what the CJEU usually does. The case at issue is no exception, in contrast to what the BVerfG has suggested. With respect to the use of the ECB’s powers, the CJEU performed a proportionality test,129 reviewed the ECB’s reasoning,130 and addressed the issue that in conducting proportionality review the CJEU can only intervene in cases of ‘a manifest error of assessment’, given that proportionality places ‘broad discretion’ with the competent authority, in our case the ECB.131 On these grounds, the CJEU did not find sufficient reasons to annul the ECB’s measure, even if it did not exclude possible criticism. Moreover, it makes no big difference whether the CJEU conducts the proportionality review when it is delimiting the monetary policy or when it is scrutinising its use. In substance, the test that is used for the review is the same and so is the result. The CJEU would either uphold the ECB’s measure, or annul it. It would, however, make a big difference if, by affirming the ECB’s exclusive competence, the CJEU would in turn automatically exclude the purchase of bonds by Germany, arguing that this would encroach upon the EU’s exclusive competence in the area of monetary policy. However, given the overlap between the monetary and economic policies, such consequence can be ruled out. There is not the slightest hint that the CJEU would see this differently. Buying bonds by Member States would normally come under economic policy. Such a measure could never implement the Eurozone monetary policy. Given all these circumstances, it seems fair to conclude that the CJEU acted infra vires, even more so in the reading by the BVerfG: as long as the CJEU applies recognised methodological principles and the decision it renders is not objectively arbitrary, the Federal Constitutional Court must respect the decision of the CJEU even when it adopts a view against which weighty arguments could be made. The mandate, conferred upon the CJEU in Art. 19(1) second sentence TEU, to ensure that the law is observed in the interpretation and application of the Treaties necessarily entails that the CJEU be granted a certain margin of error.132

Journal of Constitutional Law 1; Matthias Jestaedt, ‘Europäischer Richterdialog in Zeiten der Krise’, (2020) 2 Zeitschrift für Verwaltung, 115; Franz C Mayer, ‘The Ultra Vires Ruling: Deconstructing the German Federal Constitutional Court’s PSPP decision of 5 May 2020’, (2020) 16, European Constitutional Law Review 733; Ingolf Pernice, ‘Machtspruch aus Karlsruhe: „Nicht verhältnismäßig? – Nicht verbindlich? – Nicht zu fassen …“ Zum PSPP-Urteil des BVerfG vom 5.5.2020’, (2020) 31 Europäische Zeitschrift für Wirtschaftsrecht 508; all with further references. 128 Below n 194, with further references. 129 CJEU C 493/17 (Weiss) (n 105) paras 71ff. 130 Ibid, paras 29 ff. 131 Ibid, paras 91 f. 132 BVerfG 2 BvR 859/15 (PSPP) (n 113) para 112.

Squaring the Circle – How National Constitutions and EU Law Interact  825 However, the BVerfG’s conclusions turned out to be completely different. The BVerfG characterised the CJEU’s reasoning as ‘simply not comprehensible’, ‘simply untenable’ and ‘simply not comprehensible and thus objectively arbitrary’.133 This appears as ill-founded. It is contended that it is the BVerfG that is mistaken by insisting on a proportionality review for examining the delimitation and not the use of EU competences. Moreover, its criticism concerning the quality of the test goes beyond a residual scrutiny that respects the prerogatives of the CJEU. When reading the judgment, we cannot escape the impression that the BVerfG did not voice a concern, which nevertheless informed its critique against the ECB and the CJEU. Between the lines one might read that, measured against the Basic Law and given the position of the Bundesbank, it might be unconstitutional to transfer powers to a central bank, in our case the ECB, in order for it to enact ‘non-traditional’ measures that could, under certain circumstances, deeply affect economic policy. Should this impression hold, it could support the conclusion that the BVerfG’s verdict is an example of erroneously making the ECB and the CJEU responsible for an error that had been committed by Government and Parliament in passing the approval law (Zustimmungsgesetz) on the Maastricht treaty and later the Lisbon treaty.134 Under this premise, the BVerfG would have had to declare void the respective parts of the approval law and only afterwards could have terminated the application of the ECB’s decision.135 Putting the blame entirely – but unconvincingly – on the EU organs avoided addressing the constitutional issue.136 It also avoided to observe the distinction between the obligation to respect the Constitution, which is a matter for the national organs including the Government and Parliament, and the obligation to respect the EU Treaties, which is a matter for both EU and national organs. Also, on a procedural level, the BVerfG departed from its previous jurisprudence which would have required it to confront the CJEU in a second preliminary reference procedure with its critique before entering into the deliberation to declare the ECB’s decision and the CJEU’s judgment inapplicable.137 The BVerfG’s judgment is even more fundamentally flawed in the eyes of those commentators who argue that ultra vires control by national courts may never include the option to disapply EU law. According to them, any incompatibility between EU and national constitutions could only be rectified by either changing the former or the latter, or by leaving the EU.138 This view is not shared here. The PSPP judgment can be seen as supporting those who warned against the dangers of national identity review performed by national courts.139 Undeniably, this danger includes the 133 All quotes taken from ibid, paras 116, 117 and 118. 134 We do not support the mentioned hypothetical premise that such transfer of powers to the ECB violated the Basic Law. The purpose of our comments is to highlight the danger resulting from mingling ultra vires possibilities. 135 On this two-step approach compare Streinz, Bundesverfassungsgerichtlicher (n 44) 166 f, 343. 136 The same might be true for the Danish Supreme Court’s decision in 06.12.2016, Case 15/2014 (Ajos) 45ff. It found that direct effect between individuals, was ‘not foreseen in the Law on accession’ and thus could not be applied in Denmark. However, it is hard to blame the CJEU for this specific omission in the Law of accession, which happened in order to avoid majorities for passing that law which would have been more difficult to achieve. Therefore, also in this case we might find an unjustified mingling of flaws. Sometimes, it might be easier for a court to blame the CJEU then the national legislator. 137 BVerfG 08.04.1987, 2 BvR 687/85 (Kloppenburg) BVerfGE 75, 223, ECLI:DE:BVerfG:1987:rs19870408.2bvr068785. This is, what the Italian Constitutional Court recently had done with the result that the CJEU in principle embraced the reasoning of the national court, and by doing so increased the leeway to depart from EU secondary legislation. Compare CJEU 05.12.2017 C-42/17 (M.A.S./Taricco II) ECLI:EU:C:2017:9366 (see above in the text near n 185). 138 Eg R Daniel Kelemen, ‘On the Unsustainability of Constitutional Pluralism’, (2016) 23 Maastricht Journal of European and Comparative Law 136, 149. He contends that, if Constitutional Courts would consider an EU act ultra vires or a threat to constitutional identity, the governments could be compelled ‘to either amend their constitutions, to work through the EU political process to change the EU legal norm in question or secure an opt-out, or, if necessary, to withdraw from the Union altogether.’ 139 Eg Federico Fabbrini and András Sajó, ‘The dangers of constitutional identity’, (2019) 25 European Law Journal 457. Compare also Kelemen, ‘Unsustainability’ (n 138) with further references.

826  Stefan Griller and Lina Papadopoulou following considerations. The reasoning of the BVerfG might be taken as an invitation to turn it around: if the CJEU’s judgments can be ultra vires and, by consequence, inapplicable, why should the same not be true for BVerfG judgments?140 What would have hindered EU organs141 arguing that the BVerfG’s judgment is manifestly encroaching upon their competence and is consequently not only grossly violating EU law but even null and void? Despite all efforts to calm the actual controversy,142 there is the looming threat for the ‘war between judges’, with no final arbiter in sight. The ultimate sovereignty question in EU law remains open unless a third institution beyond the European and each national Constitutional Court gets explicitly vested with the competence to resolve conflicts of competence.143 Caution, in the form of judicial self-restraint, seems imperative before initiating an open conflict. In hindsight, it is much more difficult to tame its destructive force.

4.  The Controversy with Poland a)  The Rule of Law Controversy Over recent years, criticism of the Polish reform of its judiciary has been constantly increasing. At stake have been concerns that these reforms might undermine the independence of judges and thus be incompatible with Articles 2 and 19 TEU as well as Article 47 CFR.144 First, the Commission initiated a procedure under Article 7 TEU proposing that the Council should identify a clear risk of a serious breach by Poland of the rule of law145 which – foreseeably, given the required four-fifths majority – did not lead to any decision by the Council. Second, the Commission filed an action for failure to fulfil treaty obligations against Poland which led to the landmark judgment of the CJEU finding that Poland had indeed infringed Article 19 TEU by lowering the retirement age of judges and simultaneously granting the President of the Republic the discretion to extend the period of judicial activity for judges affected.146 More judgments followed,147 among others including the verdict that ‘allowing the right of courts and tribunals to 140 In this vein Mayer, ‘Ultra Vires Ruling’ (n 127) 747 ff. 141 Specifically, the European Commission, the CJEU, and, in the case at issue, the ECB, and with it the Bundesbank as an integrated organ of the European System of Central Banks (ESCB) and thus participating in its independence. On 8 June 2021 the European Commission sent a letter of formal notice under Art 258 TFEU to Germany claiming that the BVerfG by handing down its judgment on 5 May 2020 violated fundamental principles of EU law, in particular the principles of autonomy, primacy, effectiveness and uniform application of Union law, as well as the respect of the jurisdiction of the CJEU under Art 267 TFEU: https://ec.europa.eu/commission/presscorner/detail/en/inf_21_2743. 142 By dismissing the application for an order of execution (above n 124) the BVerfG made a substantive contribution, even if some find this a lack of consequence. However, the contrary is true for, nevertheless, immediately afterwards, sending a letter of formal notice to Germany. 143 Cf Joseph HH Weiler and Daniel Sarmiento, ‘The EU Judiciary After Weiss – Proposing A New Mixed Chamber of the Court of Justice’, EU Law Live, 1 June 2020. By contrast, we contend that bringing an independent Constitutional Court before the CJEU, in essence asking the Court to confirm the primacy of EU law, as the Commission is contemplating (above n 141), would rather fuel than solve the dispute. The CJEU cannot but confirm unconditional primacy. However, this would not settle the ‘final arbiter’ conflict (compare text below close to nn 196 and 203), but rather petrify the conflict by provoking explicit rulings hampering cooperation. 144 From the vast literature on the subject compare only Carlos Closa/Dimitry Kochenov (eds), Reinforcing Rule of Law Oversight in the European Union (Cambridge, CUP, 2016); and the literature cited above in n 3. 145 Reasoned Proposal in Accordance with Art 7(1) of the TEU Regarding the Rule of Law in Poland, COM (2017) 835 final, 20.12.2017. 146 In Case C-619/18, Commission v Poland (Independence of the Supreme Court), ECLI:EU:C:2019:531 the CJEU declared that this infringement consisted in lowering the retirement age of the judges of the Supreme Court and at the same time granting the President of the Republic the discretion to extend the period of (some of these) judges’ service beyond the new retirement age. Compare also the related ECHR-judgment in Case 4907/18, 7 May 2021, Xero Flor w Polsce v Poland. 147 Case C-192/18, Commission v Poland (Independence of the ordinary courts), EU:C:2019:924; Case C-791/19, Commission v Poland, ECLI:EU:C:2021:596; Case 487/19, W.Ż., 6. October 2021.

Squaring the Circle – How National Constitutions and EU Law Interact  827 submit requests for a preliminary ruling to the Court of Justice to be restricted by the possibility of triggering disciplinary proceedings’ violates Article 267 TFEU.148 In parallel we witnessed increasing reticence if not obstruction against CJEU judgments and interim measures in various fields.149 Third, European Parliament and the Council passed a Regulation allowing for the Council to suspend payments to a Member State or to impose an early repayment of loans where it is established that breaches of the principles of the rule of law in a Member State affect or seriously risk affecting the sound financial management of the Union budget or the protection of the financial interests of the Union,150 which the Commission started to apply 2021 to Poland when implementing the Next Generation EU Programme fighting the Covid 19 pandemic. b)  The Judgment of the Polish Constitutional Tribunal Against this background, on 7 October 2021 the Constitutional Tribunal, upon the Prime Minister’s application, handed down the following judgment declaring a number of inconsistencies between provisions of the TEU and the Polish Constitution:151 1. Article 1(1) and (2) in conjunction with Article 4(3) TEU was found inconsistent with Article 2, Article 8 and Article 90(1) of the Polish Constitution, ‘insofar’ as the EU creates an ever closer union among the peoples of Europe ‘in which a) the European Union authorities act outside the scope of the competences conferred upon them by the Republic of Poland in the Treaties; b) the Constitution is not the supreme law of the Republic of Poland, which takes precedence as regards its binding force and application; c) the Republic of Poland may not function as a sovereign and democratic state’; 2. Article 19(1), second subparagraph TEU was found, ‘insofar as, for the purpose of ensuring effective legal protection in the areas covered by EU law – it grants domestic courts (common courts, administrative courts, military courts, and the Supreme Court) the competence to: a) bypass the provisions of the Constitution in the course of adjudication’ – to be inconsistent with Articles 2, 7, 8(1), 90(1) and Article 178(1) of the Constitution; b) ‘adjudicate on the basis of provisions which are not binding, having been revoked by the Sejm and/or ruled by the Constitutional Tribunal to be inconsistent with the Constitution’ – to be inconsistent with Articles 2, 7, 8(1), 90(1), 178(1) and 190(1) of the Constitution. 3. Article 19(1), second subparagraph and Article 2 TEU were found – ‘insofar as, for the purpose of ensuring effective legal protection in the areas covered by EU law and ensuring 148 Case C-791/19, Commission v Poland, ECLI:EU:C:2021:596 para 234. 149 Compare only Case C-121/21 R, Czech Republic v Poland, ECLI:EU:C:2021:752 – an order by the Vice President of the CJEU imposing a penalty payment of EUR 500,000 per day for disrespecting the interim measure to stop lignite mining activities at an open cast mine in Turów lignite mine; see also Case C-204/21 R-RAP, 6 October 2021, Poland v Commission (interim measures, disciplinary regime) and the related judgment in Case P 7/20, 14 July 2021, of the Polish Constitutional Tribunal. See also, in this respect, the European Parliament resolution of 21 October 2021 on the rule of law crisis in Poland and the primacy of EU law (2021/2935(RSP)), www.europarl.europa.eu/doceo/document/TA-9–2021–0439_EN.html. 150 Regulation (EU, Euratom) 2020/2092 of 16 December 2020 on a general regime of conditionality for the protection of the Union budget, [2020] OJ L 433 I/1. Poland and Hungary filed an action with the CJEU to nullify this regulation under Art 263 TFEU. Their action has been dismissed by CJEU, C-156/21 and C-157/21, 16.02.2022, ECLI:EU:C:2022:97 and ECLI:EU:C:2022:98 respectively. 151 K 3/21, 7 October 2021, https://trybunal.gov.pl/en/hearings/judgments/art/11662-ocena-zgodnosci-z-konstytucjarp-wybranych-przepisow-traktatu-o-unii-europejskiej.

828  Stefan Griller and Lina Papadopoulou the independence of judges – they grant domestic courts (common courts, administrative courts, military courts and the Supreme Court) the competence to: a) review the legality of the procedure for appointing a judge, including the review of the legality of the act in which the President of the Republic appoints a judge’ – to be inconsistent with Articles 2, 8(1), 90(1) and 179 in conjunction with Article 144(3)(17) of the Constitution; b) ‘review the legality of the National Council of the Judiciary’s resolution to refer a request to the President of the Republic to appoint a judge’ – to be inconsistent with Articles 2, 8(1), 90(1) and 186(1) of the Constitution; c) ‘determine the defectiveness of the process of appointing a judge and, as a result, to refuse to regard a person appointed to a judicial office in accordance with Article 179 of the Constitution as a judge’ – to be inconsistent with Articles 2, 8(1), 90(1) and 179 in conjunction with Article 144(3)(17) of the Constitution. c)  Some Remarks We contend that the novelty of the Constitutional Tribunal’s verdict152 is in declaring core provisions of the TEU unconstitutional. This has never happened before. Insofar as these TEU provisions at least partly authorise EU measures, which the tribunal qualifies as unconstitutional, this judgment amounts to a verdict that Poland’s accession to the EU had been unconstitutional. The tribunal presents its findings in an opaque mixture of purported constitutional limits to EU integration, ultra vires measures of EU organs, and a combination of these two categories of flaws. Consequently, the judgment lacks clarity and confounds essential elements of legal scrutiny.153 The result is a fundamental challenge to the validity of EU law within the Polish legal order, and specifically a challenge to the primacy of EU law, voiced in a very general manner. This is a categorical difference if compared to eg the PSPP judgment of the GFCC, where the Court qualified a specific measure of the ECB and the related CJEU judgment as ultra vires. By contrast, the tribunal’s verdict is a broad invitation to Polish institutions, the legislator, the courts and the administration alike to deny the binding force of EU law both when it comes to incorporating it into national law and to directly applying it in cases of conflicts with national law. The broad wording appears to extend this ‘reservation’ beyond provisions of the Polish Constitution to lower-ranking Polish law, at least by authorising the Sejm and the Tribunal to ‘revoke’ EU provisions. In order to illustrate our point:154 first, Article 1 TEU states that the Treaty ‘marks a new stage in the process of creating an ever closer union’. This is far from authorising EU institutions to ‘act outside the scope of their competences conferred upon them’ as insinuated by the Tribunal. By confounding these issues the Tribunal seeds doubt on every EU measure which in principle might be outside the competences conferred and thus unconstitutional. Second, the TEU is, contrary to the suggestion of the Tribunal, not depriving the Constitution of its status of the ‘supreme law’ of the land. At least not as far as the legal condition of the

152 At the time of writing, no reasons for the judgment are available. They are to be published later in time. This lack of providing reasons renders our evaluation of the judgment necessarily provisional. 153 Compare above in the text after n 83. 154 A detailed analysis goes beyond the space available for our remarks.

Squaring the Circle – How National Constitutions and EU Law Interact  829 validity and applicability of EU law within the Polish legal order is concerned.155 This is to be discerned from the obligation Poland accepted as an EU member to respect the primacy of EU law.156 In other words, the Tribunal is mixing up the condition of legal validity and the hierarchy of application of norms within the Polish legal order. This is not withstanding the well-known controversies on the boundaries in detail. The Tribunal, however, is not even trying to draw this delicate line. Third, it remains utterly vague to what extent and why the TEU would jeopardise Poland’s sovereignty or democracy. Clearly, both sovereignty and democracy, for an EU Member State are changed by transferring powers to the EU as an indispensable element of EU accession.157 This, however, is a modification but not an elimination of sovereignty or democracy. By leaving this as vague as it is, the Tribunal is inviting to question the applicability of each and every EU measure. Fourth, as far as the Tribunal aims at ‘curtailing’ the meaning of Article 19 TEU, this is in open conflict with the jurisprudence of the CJEU. The latter has repeatedly been stressing that ‘it should be borne in mind that although, as the Republic of Poland points out, the organisation of justice in the Member States admittedly falls within the competence of those Member States, the fact remains that, when exercising that competence, the Member States are required to comply with their obligations deriving from EU law and, in particular, from the second subparagraph of Article 19(1) TEU’.158 Clearly, this encompasses the obligation to guarantee judicial independence whenever appointing judges or regulating their status. This reasoning of the CJEU is compelling, the Tribunal not giving any hint why it should be ultra vires. It goes beyond our tasks to scrutinise the meaning of the provisions of the Polish Constitution quoted by the Constitutional Tribunal. We express our doubts regarding the restrictive interpretation of these provisions given by the Tribunal – we refer to the analysis of Mirosław Wyrzykowski and Michał Ziółkowski in this volume. If, however, the Constitutional Tribunal were correct in identifying the many violations of the Polish Constitution, it would clearly follow from its verdict that Poland should never have acceded to the EU. The judgment of the Constitutional Tribunal combines, we submit, an overly restrictive interpretation of the Polish Constitution with the general allegation of – at least – the constant threat of ultra vires measures of EU institutions. Without declaring a specific EU measure unconstitutional or inapplicable, it can be read as a sort of precautionary measure empowering Polish authorities (the legislator, the courts, the administration) to disregard EU law whenever they qualify an EU measure to violate the Polish Constitution, to be ultra vires, or a combination thereof. In this sense, the judgment is a general invitation to disregard of EU law.

155 This is what the BVerfG stresses in its standing jurisprudence since BVerfGE 73, 339, 2 BvR 197/83 from 22.10.1986, Solange II: that the domestic priority of validity or application of Community Law (today EU law) only arises by virtue of a respective domestic legal command – but not solely on the grounds of international law including EC law. 156 Compare below in the text after n 159. 157 From the many examples of the debate on the vague concept of sovereignty compare only the contributions in Neil Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003); Neil Walker, ‘The sovereignty surplus’, (2020) 18 (2) I•CON, 370–428. For an account and the limited value of the concept for the EU and the constitutional debates within the Member States compare eg Stefan Griller, ‘The Impact of the Constitution for Europe on National Sovereignty’, in Ingolf Pernice and Jiri Zemánek (eds), A Constitution for Europe: The IGC, the Ratification Process and Beyond (Baden-Baden, Nomos, 2005) 151 ff, on Poland at 153 f. On the different notions of sovereignty at 160 ff. 158 Case C-791/19, Commission v Poland (Disciplinary Regime for Judges), ECLI:EU:C:2021:596 para 56. Previously already Case C-619/18, Commission v Poland (Independence of the Supreme Court), EU:C:2019:531, para. 52; and Case C-192/18, Commission v Poland (Independence of the ordinary courts), EU:C:2019:924, para 102; with further references.

830  Stefan Griller and Lina Papadopoulou

V.  The EU Law Perspective – Primacy Revisited A.  Continuity under the Treaty of Lisbon The longstanding jurisprudence of the CJEU concerning primacy of Community law is of continued relevance for EU law after the entry into force of the Lisbon Treaty.159 The CJEU stated already in the early 1960s that the Community constituted ‘a new legal order of international law’.160 Shortly afterwards, the CJEU changed its rhetoric by omitting the adjective ‘international’, and called the Treaty on the European Economic Community an independent source of law.161 The CJEU early-on specified the consequences of this concept for the relationship between Community law and the national law of the EU Member States, arriving at the unconditional supremacy of Community law vis-à-vis national law. Recourse to the legal rules or concepts of national law in order to judge the validity of measures adopted by the institutions of the Community would have an adverse effect on the uniformity and efficacy of Community law. The validity of such measures can only be judged in the light of Community law. In fact, the law stemming from the Treaty, an independent source of law, cannot because of its very nature be overridden by rules of national law, however framed, without being deprived of its character as Community law and without the legal basis of the community itself being called in question. Therefore the validity of a community measure or its effect within a Member state cannot be affected by allegations that it runs counter to either fundamental rights as formulated by the constitution of that state or the principles of a national constitutional structure.162

Clearly, this jurisprudence goes to the heart of any national reservation against the unconditional prevalence of Community law. The Lisbon Treaty did not bring about major changes with respect to the primacy of EU law. However, it did bring more nuances to the principle of primacy compared to the previous jurisprudence of the CJEU. The Draft Constitutional Treaty featured a so-called ‘primacy clause’.163 The Constitutional Treaty would have been the first EU Treaty to include an explicit clause governing the hierarchical position of the law adopted by the EU institutions, thereby coining the respective jurisprudence of the CJEU: ‘The Constitution and law adopted by the institutions of the Union in exercising competences conferred on it shall have primacy over the law of the Member states.’164 The Lisbon Treaty, by contrast, suppresses this clause. What is included instead is a Declaration (No 17) to the Treaties ‘concerning primacy’. It recalls that, in accordance with well settled case law of the Court of Justice of the European Union, the Treaties and the law adopted by the Union on the basis of the Treaties have primacy over the law of Member states, under the conditions laid down by the said case law.

The Intergovernmental Conference also decided to attach as an Annex to the Final Act an Opinion of the Council Legal Service. In its core part, this opinion reads as follows: ‘The fact that 159 In the following we do not intend to provide a comprehensive discussion on this fundamental and vast theme. Compare Craig/de Búrca, EU Law (n 2) 303 ff with further references. 160 ECJ 05.02.1963 26/62 (Van Gend en Loos/Administratie der Belastingen) ECLI:EU:C:1963:1, 12. 161 ECJ 15.07.1964 6/64 (Costa v E.N.E.L.) ECLI:EU:C:1964:66, 593 f. However, the CJEU never explained in which respect the treaty could be ‘independent’ from international law apart from forming a special international community. 162 ECJ 17.12.1970 11/70 (Internationale Handelsgesellschaft) ECLI:EU:C:1970:114. 163 For a more detailed discussion compare Stefan Griller, ‘Is this a Constitution? Remarks on a Contested Concept’ in S Griller and J Ziller (eds), The Lisbon Treaty. EU Constitutionalism without a Constitutional Treaty? (Vienna/New York, Springer, 2008) 21, 47 ff. 164 Art I-6 TECE. See also Declaration no 1 to the Constitutional Treaty: ‘The conference notes that Article I-6 reflects existing case law of the Court of Justice of the European Communities and of the Court of First Instance’.

Squaring the Circle – How National Constitutions and EU Law Interact  831 the principle of primacy will not be included in the future treaty shall not in any way change the existence of the principle and the existing case-law of the Court of Justice’. The place the principle of primacy was given in the Lisbon Treaty documents neither led to overturning it, nor to extending it.165 The effect of an extension could, however, be considered given that Declaration No 17 is – like the standing jurisprudence of the CJEU – unconditional and does not mention the competences of the Union as a limiting element. Nevertheless, restricting EU powers to its conferred competences by explicit Treaty provisions is clearly stated elsewhere in the Lisbon Treaty.166 Furthermore, the Lisbon Treaty reconfirms not only the limits of the Union’s competences, but also the respect for the national constitutional structures. It is this aspect which differs from the cited jurisprudence of the CJEU. The Treaty stipulates explicitly that the Union ‘shall respect the equality of Member states before the Treaties as well as their national identities, inherent in their fundamental structures, political and constitutional, inclusive of regional and local self-government. It shall respect their essential state functions’.167 Furthermore, regarding fundamental rights protection the Charter expressly states that the level of protection is that recognised not only by Union law, international law and international agreements, but also ‘by the Member states’ constitutions’.168 This could even encourage Member States’ reservations against the notion of unconditional supremacy of Union law over national law, and at first sight is not strengthening the respective CJEU’s jurisprudence.169 Accordingly, by contrast to the clear and unconditional tendency of the CJEU’s pre-Lisbon jurisprudence, the nuanced wording of the Lisbon Treaty points into the direction of a growing importance of the national constitutional foundations of EU membership, accepting their significance under EU law. These national provisions are the mentioned second cornerstone, and they would have to be qualified as such even without the cited reference in the TEU.170

B.  Nuances of Recent CJEU Jurisprudence Taking a closer look at the recent jurisprudence of the CJEU might reveal that these perspectives are being taken seriously also by the CJEU. However, and unsurprisingly so, they are based on the EU constitutional order and not on the Member States’ constitutional law. The EU constitutional order is, also when it comes to primacy, not totally rigid in nature. It has to consider that primacy has to be balanced against other principles of EU law. Some of these principles are expressly stated in Article 2 TEU. Furthermore, as a counterpart to the ‘opening clauses’ in national constitutional law, the mentioned ‘respect clause’ in Article 4(2) TEU is safeguarding the national identities of the EU Member States.171

165 In the same vein Jacques Ziller, Il nuovo Trattato europeo (Bologna, Il Mulino, 2007) 139 ff. The issue – probably for this reason – does not get much attention. 166 Arts 4(1)(2) and 5(1)(2) TEU. 167 Art 4(2) TEU. 168 Art 53 CFR. 169 This is not the place to dwell in detail on the conceptual foundations of the EU after Lisbon. However, it might still be defendable to characterise the Union by a statement dating back before the major reforms of the last decade, if ‘EC law’ is being replaced by ‘EU law’: ‘The principles of direct effect and supremacy, as presently formulated and accepted, continue to confirm the nature of EC law as that of a branch of international law, albeit a branch with some unusual, quasi-federal, blossoms.’ (Bruno de Witte, ‘Direct Effect, Supremacy and the Nature of the Legal Order’ in P Craig / G de Búrca [eds], The Evolution of EU Law (Oxford, OUP, 1999), 177, 210). 170 Compare below n 193. 171 This clause was created by the Treaty of Maastricht (Art F[1] TEU), possibly in an effort to counterbalance the more ‘centralist’ tendencies in that treaty, which created not only the 2nd and 3rd pillar, but also Union citizenship.

832  Stefan Griller and Lina Papadopoulou On the one hand, there is jurisprudence emphasising the continuity of unconditional primacy. In Melloni,172 the CJEU rejected the contention that EU fundamental rights protection – and consequently, primacy of EU law – could be subject to reservations that a higher national standard of protection could prevail. It found, inter alia by referring to Internationale Handelsgesellschaft, that such an interpretation would undermine the principle of the primacy of EU law inasmuch as it would allow a Member State to disapply EU legal rules which are fully in compliance with the Charter where they infringe the fundamental rights guaranteed by that State’s constitution.173

The CJEU stressed that all secondary EU legislation can be reviewed against the standards of EU law. However, it rejected the possibility of casting ‘doubt on the uniformity of the standard of protection of fundamental rights’ as defined in Union law, for this ‘would undermine the principles of mutual trust and recognition’.174 In Åkerberg Fransson, the CJEU confirmed a broad reading of its exclusive power to interpret the applicability limitation of the Charter related to the implementation of EU law.175 In essence, it found based on previous jurisprudence that the Charter is applicable in all situations coming under the scope of EU law. In that case, the situation related to tax penalties and criminal proceedings concerning value added tax (VAT) based on the VAT Directive, and the treaty obligation to counter illegal activities affecting the financial interests of the EU. On the other hand, there are judgments balancing these obligations against others with the potential of creating tension or conflict between the two types of obligations. In Sayn-Wittgenstein,176 the CJEU accepted the protection of national identity under Article 4(2) TEU177 as a justification for the restriction on the freedom of movement and residence for Union citizens that resulted from a prohibition for Austrian citizens to use titles of nobility or noble elements in their name, like ‘Fürst’ or ‘von’. It found that Article 21 TFEU must be interpreted as not precluding such a restriction in a situation where the surname includes a title of nobility, which is not permitted under the constitutional law of a Member State, provided that the measures adopted are proportionate (which the court found fulfilled). In this context, the CJEU emphasised that the respective content of public policy ‘may vary from one Member State to another and from one era to another’178 – and it did not bother that the prohibition to use titles does not form part of the highest norms in the Austrian constitutional hierarchy. In Taricco I179 and Taricco II,180 the CJEU reconfirmed the duty flowing from Article 325(1) and (2) TFEU for national courts, in criminal proceedings for VAT infringements, to dis-apply national provisions on limitation, forming part of national substantive law, under two circumstances. First, if their use would prevent the application of effective and deterrent criminal penalties in a significant number of cases of serious fraud affecting the financial interests of the EU, or ‘if they lay down shorter limitation periods for cases of serious fraud affecting those interests than for those affecting the financial interests of the Member State concerned’.181 172 CJEU C-399/11 (Melloni) (n 54) paras 55 ff. 173 Ibid, para 58. 174 Ibid, paras 63 ff. 175 Art 51(1) CFR. See CJEU 07.05.2013 C-671/10 (Åkerberg Fransson) ECLI:EU:C:2013:105, paras 16 ff. 176 CJEU 22.12.2010 C-208/09 (Sayn-Wittgenstein) ECLI:EU:C:2010:806. 177 For a deeper discussion especially on the attitude of several highest national courts discovering ‘national identity’, compare Monica Claes, ‘Negotiating Constitutional Identity or Whose Identity Is It Anyway?’ in M Claes et al (eds), Constitutional Conversations (n 1) 205, 213 ff; several contributions in Alejandro Saiz Arnaiz, Carina Alcoberro Llivina (eds), National Constitutional Identity and European Integration (Cambridge/Antwerp/Portand, Intersentia, 2013). 178 CJEU C-208/09 (Sayn-Wittgenstein) (n 161) para 87; compare also para 91. 179 CJEU 08.09.2015 C-105/14 (Taricco I) ECLI:EU:C:2015:555. 180 CJEU C-42/17 (M.A.S./Taricco II) (n 137). 181 Ibid, paras 29 ff and verdict.

Squaring the Circle – How National Constitutions and EU Law Interact  833 However, the CJEU agreed that there are important exceptions to that duty if that disapplication would entail a breach of the principle that offences and penalties must be defined by law because of the lack of precision of the applicable law, or if the retroactive application of legislation imposing conditions of criminal liability would be stricter than those in force at the time the infringement was committed.182

The CJEU stressed that the respective national legislation had only been partially harmonised by the EU. Consequently Italy was, at the time, free to secure legal protection, under the condition that the level of protection provided for by the Charter ‘and the primacy, unity and effectiveness of EU law are not thereby compromised’.183 The Court accepted the importance, both in the EU legal order and in national legal systems, of the principle that offences and penalties must be defined by law, which includes the requirements of foreseeability, precision and nonretroactivity.184 The important, and at least to a certain extent novel, conclusion was: If the national court were thus to come to the view that the obligation to disapply the provisions of the Criminal Code at issue conflicts with the principle that offences and penalties must be defined by law, it would not be obliged to comply with that obligation, even if compliance with the obligation allowed a national situation incompatible with EU law to be remedied … It will then be for the national legislature to take the necessary measures.185

More generally speaking, the duty to respect primacy of EU law may be subordinate to the duty to respect the (EU and national law) principle that offences and penalties must be defined by law, as specified by the (particular) national Constitution.186 Even if we cannot qualify these rulings as numerous and unequivocal in every respect, they provide for a nuanced view on primacy. The CJEU seems to accept that general principles of law that are guaranteed in EU constitutional law might imply exceptions to the primacy of EU law, even primary law that would otherwise require to disapply national law. The Court also seems to accept that these opposing general principles might create such effect through their specific expression within national constitutional law.

VI.  Doctrinal Solutions to the Potential Clash of Constitutions? A.  Plurality of Theories The national reports reveal an unsurprising, yet impressive variety of approaches of national constitutions vis-à-vis EU law. The respective ‘answers’ from national law inform all aspects our reports took a closer look on: limits to further integration, scrutiny of secondary EU law, implementation of the latter within the national legal system. Much has been written on the theoretical dimension of the issue, not the least in an effort to ‘solve’ it by giving answers regarding legal hierarchy or on the relationship between legal orders.

182 Ibid, para 62. 183 Ibid, para 47. 184 Ibid, para 51 ff. 185 Ibid, para 61 (emphasis added). 186 Closer scrutiny, also of important issues that are omitted here, in Clara Rauchegger, ‘National constitutional rights and the primacy of EU law: M.A.S.’, (2018) 55 CML Rev 1521, 1529 ff. On the balancing of EU constitutional principles see also Takis Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford European Union Law Library, 2006) 5 ff.

834  Stefan Griller and Lina Papadopoulou Below, we provide a brief account of such ‘theories’ without going into great detail.187 Instead, we submit that a sober look on the foundations both of the EU Constitution and the Member States’ constitutional foundations to participate in EU integration reveal why none of these theories had the power to lay the potential conflict to rest. 1.

Monism and Dualism: this is the most ‘traditional’ pattern explaining (or designing) the relationship between two or more legal orders.188 The decisive factors are the legal foundation (validity-basis) and hierarchy. • Monism postulates that • all law has its foundation in the international legal order (as the delegating legal system), or that • all law has its foundation in statal legal orders (as the delegating legal systems) – which arguably results in ‘pluralism’. • Dualism postulates two different legal foundations (delegating norms) for the validity of international law as opposed to statal law, consequently creating two (in fact: many) different legal orders each determining the relationship autonomously; in the absence of such norms incorporating norms stemming from another legal order, the latter is irrelevant.

2.

Constitutional Pluralism:189 according to this approach, the interplay between two or more legal orders should function on the basis of liberal democratic constitutionalism and the balancing between concurring principles, such as the rule of law, subsidiarity, democracy and the protection of fundamental rights. Conflicts should not be decided on the grounds of hierarchical constructs but by optimising principles according to the ‘best fit’.

187 Compare Griller, ‘Fragmentierungen’ (n 29) 250 ff. 188 Even if, originally, this has been developed for the relationship between international law and national law, it can be applied, mutatis mutandis, to the relationship between EU law and national law as well. From the flood of literature compare only the following mixture of classics and recent contributions: Dionisio Anzilotti, Lehrbuch des Völkerrechts (Leipzig, Walter de Gruyter & Co., 1929) 36 ff; Rosalyn Higgins, Problems and Process. International Law and How We Use it (Oxford, Clarendon Press, 1994) 205 ff; Helen Keller, Rezeption des Völkerrechts (Berlin, Springer, 2003); Hans Kelsen, Das Problem der Souveränität und die Theorie des Völkerrechts (Tübingen, J.C.B. Mohr (Paul Siebeck), 1920) 102 ff, 120 ff, 151 ff, 204 ff; Heinrich Triepel, Völkerrecht und Landesrecht (Leipzig, Hirschfeld, 1899) 111 ff; Alfred Verdross, Die Einheit des rechtlichen Weltbildes auf der Grundlage der Völkerrechtsverfassung (Tübingen, Mohr, 1923) 36 ff, 162; Alfred Verdross, Die Verfassung der Völkerrechtsgemeinschaft (Vienna, Springer, 1926) 33 ff. For the relevance for EU integration see Claes, Courts’ Mandate (n 27) 166 ff; Stefan Griller, ‘Völkerrecht und Landesrecht – unter Berücksichtigung des Europarechts’, in R Walter/C Jabloner/K Zeleny (eds), Hans Kelsen und das Völkerrecht (Vienna, Manz, 2004) 83. For the relevance regarding the relationship between EU law and international law see Graínne De Búrca, ‘The ECJ and the international legal order: a re-evaluation’, in G De Búrca/J Weiler (eds), The Worlds of European Constitutionalism (Cambridge, Cambridge University Press, 2012) 105, 118 ff, 119: ‘robustly dualist’; Bardo Fassbender, ‘Triepel in Luxemburg. Die dualistische Sicht des Verhältnisses zwischen Europa- und Völkerrecht in der „Kadi-Rechtsprechung“ des EuGH als Problem des Selbstverständnisses der Europäischen Union’, (2010) 8 Die Öffentliche Verwaltung, 333. 189 Compare eg Julio Baquero Cruz, ‘The Legacy of the Maastricht Urteil and the Pluralist Movement’, (2008) 14 European Law Journal 389; Armin von Bogdandy, ‘Pluralism, direct effect, and the ultimate say: On the relationship between international and domestic constitutional law’, (2008) 6 International Journal of Constitutional Law 397; Matthias Kumm, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy in Europe before and after the Constitutional Treaty’, (2005) 11 European Law Journal, 262; Franz C Mayer, ‘Verfassung im Nationalstaat: Von der Gesamtordnung zur europäischen Teilordnung?’, (2016) 75 VVDStRL 7, 28 ff; Miguel Poiares Maduro, ‘Contrapunctal Law: Europe’s Constitutional Pluralism in Action’, in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 501; Neil Walker, ‘Constitutional Pluralism Revisited’ (2016) 22 European Law Journal 333; Lina Papadopoulou, ‘Vom Prinzip des „Vorrangs“ zur pluralistischen Betrachtung der Rechtsquellen’ Lecture at the Distance-Learning University Hagen, DE, 01.03.2010, https://ub-deposit.fernuni-hagen.de/receive/mir_mods_00000387. Meanwhile, Julio Baquero Cruz has become rather critical, compare: ‘Another Look at Constitutional Pluralism in the European Union’, (2016) 22 European Law Journal 356, 368 ff.

Squaring the Circle – How National Constitutions and EU Law Interact  835 3.

Multilevel Constitutionalism: ‘This perspective views the Member States’ constitutions and the treaties constituting the European Union, despite their formal distinction, as a unity in substance and as a coherent institutional system, within which competence for action, public authority or, as one may also say, the power to exercise sovereign rights is divided among two or more levels.’190 However, primacy of EU law should govern all possible conflicts. It does not only flow from the CJEU jurisprudence but it is framed as founded in the common decision of the peoples of the EU Member States.191

B.  Theories Versus Highest Courts’ Authority The ‘weakness’ of the abovementioned theories is their level of abstraction combined with normative elements, as far as there are any. While academic analysis and propositions are performed from ‘outside observers’, courts, including Supreme and Constitutional Courts as well as the CJEU, are ‘inside’. Their authority and, hence, the limits to their authority, flow from the legal order they have been created by, and in which they operate. It is always this legal order, which, from the point of view of the courts created by that order, determines the validity of norms they are authorised to base their decision on, including the relevance of norms stemming from another legal system. Supreme courts in all Member States take explicit or implicit authorisations to participate in the EU integration at the same time as conflict governing provisions. They have to respect and enforce limits resulting from interpreting these clauses in the context of the national Constitution. In other words, ‘opening up’ the national legal order to EU law is at the same time seen as putting limits to integration. One essential element for such reasoning is that all of these provisions have to be interpreted in the context of the respective Constitution. All of these constitutions are putting limits to amendments, procedural and substantive ones. Some include eternity clauses.192 Provisions on EU accession are affecting such limits given that the procedures for enacting EU law claiming primacy over national law is deviating from democratic national procedures – to mention only the most important aspect. Given that joining the EU undisputedly did not include ending statehood, balancing the mentioned opening clauses against constitutional guarantees for certain features of the national system is indispensable. This difference in perspective (outside versus inside) between academic reasoning and adjudication must be taken into account.193 It explains why theories cannot ‘create’ unity or conflict settling norms. It equally explains the diversity of national discourses on the subject. Legal reasoning can explore the potential and the limits of the respective legal bases. It also offers guidance on

190 Ingolf Pernice, ‘Multilevel Constitutionalism and the Treaty of Amsterdam: European Constitution-Making Revisited’, (1999) 36 CML Rev 703, 706. Further details of this influential approach, which Pernice calls ‘Europäischer Verfassungsverbund’ (‘European Constitutional Compound’) eg in Ingolf Pernice, ‘Europäisches und nationales Verfassungsrecht’, (2001) 60 VVDStRL, 148, 163 ff; in-depth discussion in Christian Calliess, ‘Zum Denken im europäischen Staaten- und Verfassungsverbund’, in C Calliess (ed), Verfassungswandel im europäischen Staaten- und Verfassungsverbund (Tübingen, Mohr Siebeck, 2007) 187. 191 Pernice, ‘Multilevel Constitutionalism’ (n 175) 719; and Pernice 60 VVDStRL 182 (at 185 nevertheless combined with the contention that this would not include any hierarchy). 192 Compare text above after n 16. 193 This aspect is – with some variations – also stressed in Robert Alexy, Begriff und Geltung des Rechts, 3rd edn (Freiburg, Alber, 2011) 47 f (observer and participant perspectives); Claes, Courts’ Mandate (n 27) 212; Bernd Rüthers/Christian Fischer/Axel Birk, Rechtstheorie mit Juristischer Methodenlehre, 9th edn (Munich, C.H. Beck, 2016) margin no 22; H.L.A. Hart, The Concept of Law, 3rd edn (Oxford, OUP 2012) 88 ff; Matthias Jestaedt, Das mag in der Theorie richtig sein …, (Tübingen, Mohr Siebeck, 2006) 17 f.

836  Stefan Griller and Lina Papadopoulou informed decision-taking. However, it cannot change the legal foundations for the creation and authorisation of (the supreme) courts. It goes without saying that the picture would be different if legal analysis could reveal already existing unity or clear-cut dispute settlement between legal orders, or layers of one legal order, in our case: EU law and national law. However, our country reports reveal that this has not been the case, or at least, that none of the courts involved accepted the homogeneous unity of the legal system or any unconditional subordination to the CJEU or inversely (for the CJEU) to the supreme national courts. We do not deem it indispensable to take, in the framework of this contribution, sides with one of the above-mentioned ‘theories’ or ‘models’ on the relationship between various legal orders. Their analysis did not reveal a stable base for unity or for the solution of conflicts that could have been embraced by the deciding courts. The latter were remarkably silent in this respect, and understandably so. Courts are not supposed to write textbooks or theories. As far as the above ‘theories’ include normative elements proposing an allegedly preferable future design for the relationship between EU law and national law, this is certainly worthwhile pondering. However, setting it into work would necessarily involve the legislators, in our case constitutional amendments in many of the EU Member States as well as a corresponding Treaty change at the EU level. This, in our view is a political agenda transcending the limits of our academic undertaking. As a result, we can present neither a homogeneous picture of a uniform relationship between EU law and Member States’ constitutions, nor a compelling ‘theory’ how to arrive there. Rather, our country reports reveal a rich mosaic of variations. We submit that such diversity is neither surprising nor fundamentally flawed. It is a consequence of the tension between continuing statehood of the Member States and the deep integration within the EU they agreed upon. A consequence which can be anchored in legal analysis. Exceptionally – certainly as opposed to ‘everyday business’ – this includes the possibility of unresolved disputes where EU law, based on the Treaties, direct applicability, and supremacy requires unconditional respect while a national Constitution requires the respect for conflicting national law that would outlaw implementing or enforcing the conflicting EU rules. We cannot offer a ‘theory based’ or methodologically indisputable solution to such an exceptional conflict. We contend that the scrutiny of CJEU judgments by national courts can against the background of Article 19 TEU only be the rare exception, in cases of (grave and evident) errors transcending even the margin of error in favour of the CJEU. But we also share the stance that it would be ill founded under any of the national constitutions to claim that by allowing for EU membership any court scrutiny against massive violations of the national Constitution would have been abandoned – either when transferring powers to the EU or when implementing and enforcing EU law. It is convincing to claim a sort of ‘residual competence’ to protect the national Constitution both against unconstitutional transfers of power and grave errors committed by the EU.194 194 At this level of generalisation, we share the principled characterisation of the issue by the BVerfG, which confirmed in the (otherwise flawed) PSPP case BVerfG 2 BvR 859/15 (PSPP) (n 113): ‘the Treaties confer upon the CJEU the mandate to interpret and apply the Treaties and to ensure uniformity and coherence of EU law (cf. Art. 19[1] subpara. 2 TEU, Art. 267 TFEU); it is imperative that the respective judicial mandates be exercised in a coordinated manner. If any Member State could readily invoke the authority to decide, through its own courts, on the validity of EU acts, this could undermine the precedence of application accorded to EU law and jeopardise its uniform application. Yet if the Member States were to completely refrain from conducting any kind of ultra vires review, they would grant EU organs exclusive authority over the Treaties even in cases where the EU adopts a legal interpretation that would essentially amount to a treaty amendment or an expansion of its competences. Though cases in which institutions, bodies, offices and agencies of the European Union exceed their competences are exceptionally possible, it is to be expected that these instances remain rare due to the institutional and procedural safeguards enshrined in EU law. Nevertheless, where they do occur, the constitutional

Squaring the Circle – How National Constitutions and EU Law Interact  837 Having said this, it does not mean that each judgment referring to such foundation necessarily leads to well-founded results.195 We acknowledge that different solutions are being contended. An important and influential one claims that it must inevitably be the CJEU that has the ‘last word’. This would, for any EU Member State disagreeing on the basis of both a dissent on the interpretation of EU law and insurmountable difficulties flowing from its Constitution, only leave the choice to initiate amendments to EU law or to terminate its EU membership.196 Such a consequence would be disproportional.197 It would also deprive Constitutional and Supreme Courts of the power to fulfil one of their core competences, namely protecting the Constitution against violations not only by the executive but also by the legislator. Neither amendments of the Constitution198 nor those of EU law, nor the decision to leave the Union come under the competence of the courts. The contention in questions implies that acceding to the EU would at the same time terminate a major achievement of the constitutional development in Europe during the last century, namely that the respect for the Constitution can be enforced by the courts. This can be considered a core element of the rule of law. Unsurprisingly, in the absence of clear wording in the Treaties, national courts cannot accept this. Moreover, if the EU Treaties would indeed include unconditional primacy in the above sense, it would almost certainly violate the core of the Constitution in many Member States. This would bring us back to square one: such transfer of powers would have been unconstitutional.

VII.  A Modest Conclusion: Cooperation is a Common Constitutional Principle While we cannot offer a ‘theory’ that would solve each conflict between EU law and national law, we can confirm that the theoretical difficulties to reach such a solution at the same time build a strong argument for cooperation as a constitutional principle common to the EU and its Member States governing also conflicts of that kind. Article 4(3) TEU mandates ‘the principle of sincere cooperation’, pursuant to which ‘the Union and the Member States shall, in full mutual respect, assist each other in carrying out tasks which flow from the Treaties’. Despite its ambiguity, it has a

perspective might not perfectly match the perspective of EU law given that, even under the Lisbon Treaty, the Member States remain the ‘Masters of the Treaties’ and the EU has not evolved into a federal state … In principle, certain tensions are thus inherent in the design of the European Union; they must be resolved in a cooperative manner, in keeping with the spirit of European integration, and mitigated through mutual respect and understanding. This reflects the nature of the European Union as a union based on the multi-level cooperation of sovereign states, constitutions, administrations and courts (Staaten-, Verfassungs-, Verwaltungs- und Rechtsprechungsverbund)’. 195 As many others, we contend that the BVerfG erred in the PSPP case where the above quote is taken from; compare text above after n 127. 196 Eg Kelemen, ‘Unsustainability’ (n 138) 149: If Constitutional Courts would consider an EU act ultra vires or a threat to constitutional identity, the governments could be compelled ‘to either amend their constitutions, to work through the EU political process to change the EU legal norm in question or secure an opt-out, or, if necessary, to withdraw from the Union altogether.’ 197 Would the legal arguments for a constitutional duty to leave the EU have been stronger than rendering the PSPP judgment (even if it were not flawed)? And how should this duty have been enforced by the BVerfG? Should it instead have annulled Germany’s EU membership, and on what constitutional basis? We are not talking of the possible political motivation to make the hurdle as high as possible in favour of EU law: why leaving for one specific issue? But this is not a legal argument. 198 With the exception of EU Member States with layered constitutions that make unconstitutional constitutional law possible, like in Austria, the Czech Republic and Germany. However, this does not change the point we intend to make in the above text.

838  Stefan Griller and Lina Papadopoulou guiding force also in reducing open conflicts between EU law and national law.199 Arguably, there is no stronger legal force than this one currently available. It goes without saying that an amendment of the EU Treaties could change this.200 However, such an amendment would, as the contributions in this volume and the above deliberations reveal, almost certainly necessitate constitutional amendments to most of the Member States’ constitutions. It is beyond the purpose of this book to explore such possibilities closer. The principle of cooperation evidently cannot avoid every conflict. However, it is not without teeth. It calls for cooperative solutions and excludes the mentioned ‘extreme’ positions that there must be and there is always a ‘final arbiter’ on one of the two sides. Cooperation, by contrast, calls for cooperation also between the CJEU and Supreme Member States’ Courts. It would also be mistaken to take pathology for a principle. The contrary is true. Open conflict is the rare exception, not the rule. We have been witnessing ‘unity in diversity’ over decades meaning that all EU Member States are acting on the basis of (implicit or explicit) constitutional authorisations to participate in the EU integration, encompassing the constitutional specificities of the Union, including direct effect and supremacy of EU law. All of the EU Member States allow for developing the EU integration further, if only against the backdrop of a broad spectrum of diverging constitutional conditions. All of the EU Member States dispose of mechanisms for implementing and enforcing EU law including CJEU judgments. The clash of legal orders consequently is not an everyday business. However, it is not only theoretically challenging, but a looming threat that may from one day to the other become practical. This concerns both potential stumbling blocks for reforms as well as refusals to implement and enforce EU law. Mitigating elements are, on the one hand, the ‘opening clauses’ of the national constitutions interpreted in their context and their telos legitimising participation in EU integration as a ‘process of creating an ever-closer union among the peoples of Europe’.201 This context cannot immunise EU law against reinforcing national constitutional limits. However, it can serve as an obstacle against overly strict and ‘EU-sceptical’ interpretation of any opening clause. On the other hand, the duty of the EU to respect the ‘national identities’ of its Member States, as ‘inherent in their fundamental structures, political and constitutional’ and including ‘their essential State functions’,202 is, among others, and jointly with the duty of sincere cooperation, an indisputable basis for the need to proportionally balance the benefit of efficient and uniform EU governance against specific reservations of independence at the Member States’ level. Against this backdrop, ‘cooperation’,203 as a guiding principle of interaction between supreme national courts and the CJEU, can be qualified as a means to reduce, as far as possible, open conflicts between national law and EU law, a legal principle calling for conflict avoiding

199 Compare Marcus Klamert, The Principle of Loyalty in EU Law (Oxford, OUP, 2014) especially at 47 ff and 211 ff, with further references. 200 Some of the proposals are being recalled by Mayer, ‘Ultra Vires Ruling’ (n 127) 765. See also above n 143. 201 Quote taken from the preamble of the TEU. This clause has been ratified by all Member States and has become a symbol of the dynamics that not the least inform legal reasoning. 202 Art 4(2) TEU. 203 From the abundant academic debate compare, in addition to Klamert, Loyalty (n 184), Thomas von Danwitz, ‘Gerichtlicher Schutz der Grundrechte’, in C Grabenwarter (ed), Europäischer Grundrechteschutz (Baden-Baden, Nomos, 2014) § 6 marginal nos 11 ff; Rudolf Streinz, ‘Streit um den Grundrechtsschutz? Zum Grundrechtsschutz in der Europäischen Union nach den Urteilen des EuGH in den Fällen Åkerberg Fransson und Melloni und des BVerfG zur Antiterrordatei’, in D Heid/R Stotz/A Verny (eds), Festschrift für Manfred A. Dauses zum 70. Geburtstag (Munich, C.H. Beck, 2014) 429, 441 ff; Daniel Thym, ‘Separation versus Fusion – or: How to Accommodate National Autonomy and the Charter? Diverging Visions of the German Constitutional Court and the European Court of Justice’, (2013) 9 European Constitutional Law Review 391.

Squaring the Circle – How National Constitutions and EU Law Interact  839 interpretation to the largest extent possible. In the same vein, it appears untenable to identify the national supreme courts as the ‘last arbiters’ in any such disputes.204 Be that as it may, ‘cooperation’ as such cannot establish a clear hierarchy. It can only reduce the likelihood of conflicts resulting from the absence of hierarchy. Having said that, we admit that once inconsistencies between national Constitutions and EU law transcend punctual controversies and develop into a systemic conflict hampering the proper interaction of EU and national law, ‘cooperation’ by definition cannot provide for a solution. In such a case we only see the option to either amend the national Constitution, convince all other 26 Member States to change the Treaties according to the requirements of the national Constitution, or to leave the EU. Expulsion is not foreseen in the Treaties. Neither is enforcement of EU law by coercion which the EU does not dispose of. The ‘sanctions’ available to the EU205 are too weak to coerce a Member State defying to respect the fundamental values of the EU.

References A Albi and S Bardutzky (eds), National Constitutions in European and Global Governance: Democracy, Rights, the Rule of Law (The Hague, T.M.C. Asser Press, 2019). R Alexy, Begriff und Geltung des Rechts, 3rd edn (Freiburg, Alber, 2011). G Anagnostaras, ‘Activation of the Ultra Vires Review: The Slovak Pensions Judgment of the Czech Constitutional Court’, (2013) 14 German Law Journal 959. D Anzilotti, Lehrbuch des Völkerrechts (Leipzig, Walter de Gruyter & Co., 1929). J Baquero Cruz, ‘The Legacy of the Maastricht Urteil and the Pluralist Movement’, (2008) 14 European Law Journal 389. J Baquero Cruz, ‘Another Look at Constitutional Pluralism in the European Union’, (2016) 22 European Law Journal 356. J Basedow, J Dietze, S Griller, M Kellerbauer, M Klamert, L Malferrari, T Scharf, D Schnichels, D Thym, J Tomkin, ‘European integration: Quo vadis? A critical commentary on the PSPP judgment of the German Federal Constitutional Court of May 5, 2020’, (2021) 19 International Journal of Constitutional Law 1. A von Bogdandy, ‘Pluralism, direct effect, and the ultimate say: On the relationship between international and domestic constitutional law’, (2008) 6 International Journal of Constitutional Law 397. C Calliess, ‘Zum Denken im europäischen Staaten- und Verfassungsverbund’, in C Calliess (ed), Verfassungswandel im europäischen Staaten- und Verfassungsverbund (Tübingen, Mohr Siebeck, 2007) 187. M Claes, The National Courts’ Mandate in the European Constitution (Oxford, Hart, 2006). M Claes, ‘Negotiating Constitutional Identity or Whose Identity Is It Anyway?’ in M Claes et al (eds), Constitutional Conversations in Europe (Cambridge/Antwerp/Portland, Intersentia, 2012) 205. M Claes, M de Visser, P Popelier and C Van de Heyning (eds), Constitutional Conversations in Europe (Cambridge/Antwerp/Portland, Intersentia, 2012). C Closa/D Kochenov (eds), Reinforcing Rule of Law Oversight in the European Union (Cambridge, CUP, 2016). X Contiades (ed), Engineering Constitutional Change – A Comparative Perspective on Europe, Canada and the USA (London/New York, Routledge, 2013). 204 Eg Theodor Schilling, ‘The Autonomy of the Community Legal Order: An Analysis of Possible Foundations’, (1996) 37 Harvard International Law Journal 389, 408: ‘It is basically up to the Member States to choose how they exercise the final word on interpreting the European Treaties.’ Compare, already at the time, the harsh critique in Joseph H.H. Weiler/Ulrich R. Haltern, ‘The Autonomy of the Community Legal Order – Through the Looking Glass’, (1996) 37 Harvard International Law Journal 411, qualifying, however the CJEU as the ‘final umpire’ which comes at least close to the abovementioned stance which arguably is not ‘cooperation’ either. 205 Compare above in the text after n 144.

840  Stefan Griller and Lina Papadopoulou P Craig, ‘Constitutions, Constitutionalism, and the European Union’, (2001) 7 European Law Journal 125. P Craig and G de Búrca, EU Law, 7th edn (Oxford, OUP, 2020). T von Danwitz, ‘Gerichtlicher Schutz der Grundrechte’, in C Grabenwarter (ed), Europäischer Grundrechteschutz (Baden-Baden, Nomos, 2014) § 6. B de Witte, ‘Direct Effect, Supremacy and the Nature of the Legal Order’ in P Craig and G de Búrca (eds), The Evolution of EU Law (Oxford, OUP, 1999) 177. B de Witte, Do Not Mention the Word: Sovereignty in Two Europhile Countries: Belgium and The Netherlands, in Neil Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 359. European Parliament, Policy Department C: Citizens’ Rights and Constitutional Affairs (ed), National Constitutional Law and European Integration, Study PE 432.750 (Brussels, European Parliament, 2011). European Parliament, Policy Department C: Citizens’ Rights and Constitutional Affairs (ed), National Constitutional Avenues for further EU Integration, Study PE 493.046, (Brussels, European Parliament, 2014). F Fabbrini and A Sajó, ‘The dangers of constitutional identity’, (2019) 25 European Law Journal 457. B Fassbender, ‘Triepel in Luxemburg. Die dualistische Sicht des Verhältnisses zwischen Europa- und Völkerrecht in der „Kadi-Rechtsprechung“ des EuGH als Problem des Selbstverständnisses der Europäischen Union’, (2010) 8 Die Öffentliche Verwaltung, 333. T Giegerich, Europäische Verfassung und deutsche Verfassung im transnationalen Konstitutionalisierungsprozeß: Wechselseitige Rezeption, konstitutionelle Evolution und föderale Verflechtung (Berlin, Springer, 2003). C Grabenwarter, ‘National Constitutional Law Relating to the European Union’, in A von Bogdandy and J Bast (eds), Principles of European Constitutional Law, 2nd edn, (Oxford/Munich, Hart Publishing/ C.H. Beck, 2010) 83. S Griller, Die Übertragung von Hoheitsrechten auf zwischenstaatliche Einrichtungen (Wien/New York, Springer, 1989). S Griller, ‘Völkerrecht und Landesrecht – unter Berücksichtigung des Europarechts’, in R Walter/ C Jabloner/K Zeleny (eds), Hans Kelsen und das Völkerrecht (Vienna, Manz, 2004) 83. S Griller, ‘The Impact of the Constitution for Europe on National Sovereignty’, in Ingolf Pernice and Jiri Zemánek (eds), A Constitution for Europe: The IGC, the Ratification Process and Beyond (Baden-Baden, Nomos, 2005) 151. S Griller, ‘Is this a Constitution? Remarks on a Contested Concept’ in S Griller and J Ziller (eds), The Lisbon Treaty. EU Constitutionalism without a Constitutional Treaty? (Vienna/New York, Springer, 2008) 21. S Griller, ‘Executive Summary’, in European Parliament, Policy Department C: Citizens’ Rights and Constitutional Affairs (ed), National Constitutional Law and European Integration, Study PE 432.750 (Brussels, European Parliament, 2011) 21. S Griller, ‘Fragmentierungen im Öffentlichen Recht: Diskursvergleich im internationalen und nationalen Recht’, (2018) 77 VVDStRL, 237. S Griller, ‘Analytical report on the Legal background of Member States’ Positions towards Economic and Fiscal Integration’, in Stefan Griller and Elisabeth Lentsch (eds), EMU Integration and Member States’ Constitutions (Oxford, Hart, 2021) 1. S Griller, F Maislinger and A Reindl, Fundamentale Rechtsgrundlagen einer EG-Mitgliedschaft. Verfassungsfragen der Übernahme von EG-Recht in den bisherigen Mitgliedstaaten in vergleichender Sicht (Wien, Verlag der Österreichischen Staatsdruckerei, 1991). HLA Hart, The Concept of Law, 3rd edn (Oxford, OUP 2012). A Jakab and D Kochenov (eds), The Enforcement of EU Law and Values (Oxford, OUP, 2017). R Higgins, Problems and Process. International Law and How We Use it (Oxford, Clarendon Press, 1994). M Jestaedt, Das mag in der Theorie richtig sein …, (Tübingen, Mohr Siebeck, 2006). M Jestaedt, ‘Europäischer Richterdialog in Zeiten der Krise’, (2020) Zeitschrift für Verwaltung, 115. O Jouanjan, ‘Fragmentierungen im Öffentlichen Recht: Diskursvergleich im Verfassungs- und Verwaltungsrecht’, (2018) 77 VVDStRL, 351. RD Kelemen, ‘On the Unsustainability of Constitutional Pluralism’, (2016) 23 Maastricht Journal of European and Comparative Law 136. H Keller, Rezeption des Völkerrechts (Berlin, Springer, 2003).

Squaring the Circle – How National Constitutions and EU Law Interact  841 A Kellermann/J de Zwaan/J Czuczai (eds), EU Enlargement – The Constitutional Impact at EU and National Level (The Hague, T.M.C. Asser Press, 2001). H Kelsen, Das Problem der Souveränität und die Theorie des Völkerrechts (Tübingen, J.C.B. Mohr [Paul Siebeck], 1920). U Kischel, ‘Fragmentierungen im Öffentlichen Recht: Diskursvergleich im internationalen und nationalen Recht’, (2018) 77 VVDStRL, 285. M Klamert, The Principle of Loyalty in EU Law (Oxford, OUP, 2014). M Kumm, ‘The Jurisprudence of Constitutional Conflict: Constitutional Supremacy in Europe before and after the Constitutional Treaty’, (2005) 11 European Law Journal, 262. E Lauterpacht, ‘The Legal Effect of Illegal Acts of International Organisations’, (1965) 64 Cambridge Essays in International Law 88. K Lenaerts and P van Nuffel, Constitutional Law of the European Union, 3rd edn, (London, Sweet & Maxwell Ltd, 2011). G Lienbacher (ed), Verfassungsinterpretation in Europa. Heinz Schäffer Gedächtnissymposion (Vienna, Jan Sramek Verlag, 2011). FC Mayer, ‘Verfassung im Nationalstaat: Von der Gesamtordnung zur europäischen Teilordnung?’, (2016) 75 VVDStRL, 7. FC Mayer, ‘The Ultra Vires Ruling: Deconstructing the German Federal Constitutional Court’s PSPP decision of 5 May 2020’, (2020) 16 European Constitutional Law Review 733. FC Mayer and M Wendel, ‘Die verfassungsrechtlichen Grundlagen des Europarechts’, in A Hatje and C Müller-Graff (eds), Europäisches Organisations- und Verfassungsrecht. Enzyklopädie Europarecht I (Baden-Baden, Nomos, 2014) 163. L Papadopoulou, ‘Vom Prinzip des „Vorrangs“ zur pluralistischen Betrachtung der Rechtsquellen’ Lecture at the Distance-Learning University Hagen, DE, 01.03.2010, https://ub-deposit.fernuni-hagen.de/receive/ mir_mods_00000387. I Pernice, ‘Multilevel Constitutionalism and the Treaty of Amsterdam: European Constitution-Making Revisited’, (1999) 36 CML Rev 703. I Pernice, ‘Europäisches und nationales Verfassungsrecht’, (2001) 60 VVDStRL 148. I Pernice, ‘Machtspruch aus Karlsruhe: „Nicht verhältnismäßig? – Nicht verbindlich? – Nicht zu fassen …“ Zum PSPP-Urteil des BVerfG vom 5.5.2020’, (2020) 31 Europäische Zeitschrift für Wirtschaftsrecht 508. A Peters, Elemente einer Theorie der Verfassung Europas (Berlin, Duncker & Humblot, 2001). M Poiares Maduro, ‘Contrapunctal Law: Europe’s Constitutional Pluralism in Action’, in N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003) 501. H Rasmussen, On Law and Policy in the European Court of Justice (Dordrecht, Martinus Nijhoff, 1986). C Rauchegger, ‘National constitutional rights and the primacy of EU law: M.A.S.’, (2018) 55 CML Rev 1521. F Reimer, ‘Fragmentierungen im Öffentlichen Recht: Diskursvergleich im Verfassungs- und Verwaltungsrecht’, (2018) 77 VVDStRL, 413. B Rüthers/C Fischer/A Birk, Rechtstheorie mit Juristischer Methodenlehre, 9th edn (Munich, C.H. Beck, 2016). A Saiz Arnaiz/C Alcoberro Llivina (eds), National Constitutional Identity and European Integration (Cambridge/Antwerp/Portand, Intersentia, 2013). KL Scheppele, D Kochenov and B Grabowska-Moroz, ‘EU Values Are Law, after All: Enforcing EU Values through Systemic Infringement Actions by the European Commission and the Member States of the European Union’, (2020) 39 Yearbook of European Law 3. H Schermers and N Blokker, International Institutional Law, 6th edn (Nijhoff, Brill, 2018). T Schilling, ‘The Autonomy of the Community Legal Order: An Analysis of Possible Foundations’, (1996) 37 Harvard International Law Journal 389. A-M Slaughter/A Stone Sweet/JHH Weiler (eds), The European Courts and National Courts – Doctrine and Jurisprudence. Legal Change in Its Social Context (Oxford, Hart, 1998). R Streinz, Bundesverfassungsrechtlicher Grundrechtsschutz und Europäisches Gemeinschaftsrecht (BadenBaden, Nomos, 1989).

842  Stefan Griller and Lina Papadopoulou R Streinz, ‘Streit um den Grundrechtsschutz? Zum Grundrechtsschutz in der Europäischen Union nach den Urteilen des EuGH in den Fällen Åkerberg Fransson und Melloni und des BVerfG zur Antiterrordatei’, in D Heid/R Stotz/A Verny (eds), Festschrift für Manfred A. Dauses zum 70. Geburtstag (Munich, C.H. Beck, 2014) 429. D Thym, ‘Separation versus Fusion – or: How to Accommodate National Autonomy and the Charter? Diverging Visions of the German Constitutional Court and the European Court of Justice’, (2013) 9 European Constitutional Law Review 391. T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford European Union Law Library, 2006). H Triepel, Völkerrecht und Landesrecht (Leipzig, Hirschfeld, 1899). A Verdross, Die Einheit des rechtlichen Weltbildes auf der Grundlage der Völkerrechtsverfassung (Tübingen, Mohr, 1923). A Verdross, Die Verfassung der Völkerrechtsgemeinschaft (Vienna, Springer, 1926). N Walker, ‘The Idea of Constitutional Pluralism’, (2002) 65 Modern Law Review 317. N Walker (ed), Sovereignty in Transition (Oxford, Hart, 2003). N Walker, ‘The sovereignty surplus’, (2020) 18 (2) I•CON 370. J Weiler, The Constitution of Europe (Cambridge, Cambridge University Press, 1999). J Weiler/U Haltern, ‘The Autonomy of the Community Legal Order – Through the Looking Glass’, (1996) 37 Harvard International Law Journal 411. K Wengler, ‘Recours judiciaire à instituer contre les décisions d’organes internationaux, Rapport et project de Résolutions’, (1952) I Annuaire de l’Institut de Droit International 224. J Ziller, Il nuovo Trattato europeo (Bologna, Il Mulino, 2007).

INDEX Austria constitutional amendments  566–7 constitutional culture  555–6 constitutional system  553–5 Courts  570–2 double obligation  572–4 ESM (European Stability Mechanism) treaty  561, 567 EU integration  560–8 EU law constitutional review  574–80 implementation  568–80 and national law  561–3 secondary legislation, scrutiny of  563–4 EU membership  556–61 Fiscal Compact  561, 567 judiciary  554–5 neutrality  564–7 Parliament  568–70 ultra vires doctrine  563–4 Belgium constitutional amendments  1, 5–6, 20–1 constitutional characteristics  3–5 constitutional culture  5–6 Council of Ministers (EU)  21–2 ESM (European Stability Mechanism) treaty  10 EU integration  9–10, 13–21 EU Law constitutional amendments  20–1 constitutional hierarchy  12–13 constitutional requirements  17–19 drafting  22–4 implementation  21–9 and national law  29–30 EU membership  6–13 federalisation  1–3 and international law  10–12 international organisations  8–10 national identity doctrine  17–19, 30 national sovereignty  7–9 state reforms  1–3 treaties  13–15 Bologna Process  200 Bulgaria BGCC (Constitutional Court)  41–5 constitutional amendments  41–8

constitutional characteristics  34–7 Constitutional Court  41–5 constitutional culture  37–9 courts  54 EU integration  46–52 EU law implementation  52–4 and national law  55–7 and national legal order  48 EU membership  39–48 constitutional amendments  41–6 sanction for  39–41 judiciary  36–7 ‘just international order’ concept  47–8 multilevel constitutionalism  56 multilevel judicial dialogue  54 open statehood  46–8 parliamentary system  34–6, 52–3 Subsidiarity Protocol  53 Charter of Fundamental Rights (CFR) Croatia  351–2 Czech Republic  82–3 Finland  740 Greece  278–9 Ireland  243–4 Luxembourg  477 Malta  518, 530 Spain  297–300 CJEU see European Union, Court of Justice competences, transfer of  809–14 constitutional amendments  799–800 constitutional pluralism  834 constitutional principles  797–8, 837–9 constitutional review of statutes  801–2 constitutional systems, and cultures  796–804 constitutions clashes of  833–7 forms of  796–7 cooperation as a constitutional principle  837–9 core competences  814 Croatia Charter of Fundamental Rights (EU)  351–2 constitutional culture  337–8 constitutional identity  344–5 constitutional system  333–7 EU citizenship rights  352–3

844  Index EU integration  343–53 EU law  345–52 implementation  353–4 and national law  351–2 and EU legal order  354–5 EU membership  338–42 judicial review  345–6 treaty amendments  341–2 cultures of interpretation  802–4 Cyprus constitutional culture  388–9 constitutional system  385–8 ESM (European Stability Mechanism) treaty  395 EU integration  390–5 EU law implementation  396–8 and national law  398 EU membership  389–90 European Financial Stability Facility (EFSF)  395 Fiscal Compact  395 Czech Republic Charter of Fundamental Rights (EU)  83 Committee for the EU, binding decisions  80 Constitution  62–5 Constitutional Court  60–1, 66 and EU law  82–3 European Doctrine  70–4 preliminary question procedure  74–5 constitutional culture  62–6 constitutional system  59–60 EU integration  76–9 EU law and constitutional system  85–6 drafting  81 implementation  79–85 secondary legislation, scrutiny of  78–9 transposition of  81–3 EU membership  67–76 Fiscal Compact  76 international law  69–70 Parliament  79–81 Sugar Quotas III case  70–1, 74–5, 77 transfer of powers to EU  76–8 treaties  69–70 ultra vires doctrine  78–9 Denmark constitutional system  90–4 EU integration  103–13 EU law  113–17 EU membership  98–103 judicial system  92–4 Lisbon case  106–7 Maastricht case  104–6 Ministry of Justice  115–17 opt-outs  107–8 Parliament  94–8

political system  91–2 ultra vires doctrine  108–12 dualism  834 EFSF (European Financial Stability Facility)  395, 471 ESM (European Stability Mechanism) treaty Austria  561, 567 Belgium  10 Cyprus  395 Estonia  189–93 Greece  260 Hungary  484 Luxembourg  471 Malta  519 Portugal  643–4 Estonia Bologna Process  200 constitutional changes  178–82 constitutional culture  176–7 constitutional review  199–200 constitutional system  175–6 EKRE (Estonian Conservative People’s Party)  202 ESM (European Stability Mechanism) treaty  189–93 EU integration  182–6 EU law implementation of  186–8 and national law  189–204 primacy of  182–6 EU membership  178–82 extradition rules  196–7 farming  203 judicial protection  195 legitimate expectations, principle of  196, 199 publication of laws  196 Rechtsstaat constitutionalism  193, 194–6 rule of law crisis  202–4 rule of law safeguards  193–202 sugar stocks cases  196 EU see European Union European Financial Stability Facility (EFSF)  395, 471 European Stability Mechanism treaty see ESM (European Stability Mechanism) treaty European Union accession to  804–5 Charter of Fundamental Rights Croatia  351–2 Czech Republic  82–3 Finland  740 Greece  278–9 Ireland  243–4 Luxembourg  477 Spain  297–300

Index  845 Court of Justice (CJEU) jurisprudence  831–3 obligation of trust  197–8 and Portuguese Constitutional Court  648–9 power to develop law  143–4 primacy in German Basic Law  144–6 integration  806–30 Austria  560–8 Belgium  9–10, 13–21 Bulgaria  46–8, 49–52 Croatia  343–53 Cyprus  390–5 Czech Republic  76–9 Denmark  103–13 Estonia  182–6 Finland  736–41 France  318–28 Germany  129–66 Greece  251–2, 260–75 Hungary  494–6 Ireland  225–32 Italy  367–73 Latvia  408–17 Lithuania  446–53 Luxembourg  469–75 Malta  519–28 Netherlands  544–5 Poland  606–11 Portugal  633–9, 640–4 Romania  664–8 Slovakia  713–21 Slovenia  687–9 Spain  290–1 Sweden  770–1 law Austria  568–80 and national law  561–3 scrutiny  563–4 Belgium constitutional amendments  20–1 constitutional hierarchy  12–13 constitutional requirements  17–19 drafting  22–4 implementation  21–9 and national law  29–30 Bulgaria  48 implementation  52–4 and national law  55–7 Croatia  345–52 implementation  353–4 and national law  351–2 Cyprus implementation  396–8 and national law  398 Czech Republic Constitutional Court  82–3 implementation  79–85

Denmark implementation  113–17 preliminary references  113–15 Estonia implementation of  186–8 and national law  189–204 primacy of  182–6 Finland  741–5 France  324–8 Germany Federal Constitutional Court  148–51 openness of Basic Law to  148, 149–50 Greece  262–79 Hungary  496–503 Ireland implementation of  232–44 and national law  244–5 Italy implementation  373–7 and national law  377–82 supremacy of  367–73 Latvia  418–25 Lithuania  442–5, 452–6 Luxembourg  473–9 Malta  525–31 Netherlands  546–9 Poland  604–13 Portugal  644–51 Romania  668–76 secondary legislation, scrutiny of  817–29 Slovakia  711–13, 716–24 Slovenia  689–97 Spain  291–301 supranational character of  141–3 Sweden  761–70, 771–4 United Kingdom  788–91 Lisbon, Treaty of  830–1 Maastricht treaty  104–6, 132–5, 308–9, 469–70 membership of  804–6 Austria  556–61 Belgium  6–13 Bulgaria  39–48 constitutional authorisations  805–6 Croatia  338–42 Cyprus  389–90 Czech Republic  67–76 Denmark  98–103 Estonia  178–82 Finland  731–6 France  308–18 Greece  252–60 Hungary  489–94 Ireland  216–25 Italy  364–7 Latvia  404–8 Lithuania  439–47

846  Index Luxembourg  466–71 Malta  514–19 Netherlands  538–44 Poland  601–6 Portugal  635–6 Romania  660–4 Slovakia  707–13 Slovenia  685–7 Spain  287–90 Sweden  756–70 United Kingdom  778–82 opt-outs  107–8 Stability and Growth Pact  770 transfer of powers  806–17 Czech Republic  76–8 France  318–24 Greece  261–2 Ireland  225–6 Lithuania  452–3 Luxembourg  472–3 Malta  519–25 Portugal  640–2 Finland Charter of Fundamental Rights  740 constitutional amendments  745–6 constitutional culture  729–31 constitutional identity  740–1 constitutional system  727–9 EU integration  736–41 EU law implementation  741–4 and national law  744–5 EU membership  731–6 human rights  739–40 Fiscal Compact Austria  561, 567 Cyprus  395 Czech Republic  76 Greece  260, 275 Lithuania  446–7 Luxembourg  471 Malta  519 Portugal  643–4 Sweden  770 France Constitutional Council  324–7 constitutional system and culture  305–7 Council of State  327–8 EU integration  318–28 EU law, secondary legislation, scrutiny of  324–8 EU membership  308–18 Maastricht Treaty  308–9 national identity  323 transfer of powers to EU  318–24 ultra vires doctrine  324–8

fundamental constitutional principles  797–8 fundamental rights protection  807–9 see also human rights Germany Basic Law  125–35 openness to European law  148, 149–50 primacy of EU Court of Justice  144–6 Bundestag  158–9 constitutional culture  127–9 constitutional identity  153–8 constitutional jurisprudence  123–4 constitutional system  125–7 electoral citizens  159–61 EU integration  129–66, 158–9 Federal Constitutional Court (BVerfG)  123–4, 135–66 arguments limiting integration  151–61 arguments promoting integration  141–51 and EU law  148–51 and European Court of Justice (ECJ)  148–9 judicial review  159–61 jurisprudence  135–41, 161–6 Solange jurisprudence  132, 136, 146–8, 807–8 ultra vires review  151–3 Maastricht treaty  132–5 Parliament  158–9 PSPP judgment  821–6 responsibility for integration  148, 150–1 Greece Charter of Fundamental Rights (EU)  278–9 constitutional culture  250–2 constitutional system  247–50 courts  250 EU integration  251–2, 260–75 EU law court enforcement  278–9 implementation  275–9 and national law  279 primacy of  266–75 secondary legislation, scrutiny of  262–6 EU membership  252–60 European Stability Mechanism (ESM) treaty  260 Fiscal Compact  260, 275 international law  252–4 judiciary  248–50 Parliament  250, 276–7 subsidiarity, principle of  276–7 transfer of powers to EU  261–2 Hungary Constitutional Court  493–4 constitutional culture  486–9 constitutional identity  484–5 constitutional system  485–6 ESM (European Stability Mechanism) treaty  484 EU integration  494–6

Index  847 EU law  496–503 EU membership  489–94 Fundamental Law  483–6, 492–3 sovereignty  484–5 Ireland Charter of Fundamental Rights  243–4 Citizens’ Assembly  215–16 constitutional amendments  227 constitutional culture  213–16 constitutional identity  227–9 constitutional system  211–13 Crotty v An Taoiseach  219 EU integration  225–32 EU law court enforcement  243–4 implementation of  232–44 and national law  244–5 secondary legislation, scrutiny of  231–2 EU membership  216–25 and European Commission  239 and European Council  239–40 and Eurozone  239–40 human rights  231 international law  224–5 judicial control  221–4 judicial system  213 Parliament (Oireachtas)  235–40 EU Affairs Committee  236–7 permanent representative in Brussels  239 sectoral committees  237–8 sovereignty  229–31 subsidiarity, principle of  238–9 transfer of powers to EU  225–6 treaty amendments  219–20, 225–6 ultra vires doctrine  231–2 Italy constitutional culture  362–4 constitutional reform  361–2 constitutional system  357–62 counter-limits doctrine  371–3 EU integration  367–73 EU law implementation  373–7 and national law  377–82 supremacy of  367–73 EU membership  364–7 Parliament  375–7 separation of powers  359–61 sovereignty  368–9 Latvia constitutional core  413–15 Constitutional Court  420 constitutional culture  403–4 constitutional system  401–3 courts  420–3

EU integration  408–17 EU law  418–25 EU membership  404–8 human rights  410–13 legitimate expectations, principle of  196, 199 Lithuania constitution, structure of  431–3 constitutional amendments  434–6 constitutional culture  436–9 constitutional review  429–31 constitutional system  427–36 EU integration  446–53 EU law implementation  453–5 and national law  455–6 secondary legislation, scrutiny of  452–3 status of  442–5 EU membership  439–47 Fiscal Compact  446–7 human rights  451–2 international law  433–4 judiciary  427–8 land ownership  439–41 local self-government  442 national bank  445–6 national identity  450–1, 451–2 sovereignty  449–50 transfer of competences to EU  452–3 Luxembourg Chambre  461–2, 476 Charter of Fundamental Rights  477 constitutional amendments  469–71, 479–80 constitutional culture  463–6 constitutional system  459–63 ESM (European Stability Mechanism) treaty  471 EU integration  469–75 EU law  473–9 court enforcement  477 implementation  475–7 relationship with national law  477–9 secondary legislation, scrutiny of  473–5 EU membership  466–71 European Financial Stability Facility (EFSF)  471 Fiscal Compact  471 judiciary  462–3 Maastricht Treaty  469–70 monarchy  461 Parliament  461–2, 476 transfer of powers to EU  472–3 treaties  467–8 ultra vires doctrine  473–5 Maastricht treaty  104–6, 132–5, 308–9, 469–70 Malta abortion  513 Charter of Fundamental Rights  518, 530 constitutional culture  510–13

848  Index constitutional system  507–10 culture  512 economic freedoms  513 ESM (European Stability Mechanism) treaty  519 EU integration  519–28 EU law  525–31 court enforcement  530 implementation  528–30 judicial review  525–8 relationship with national law  530–1 secondary legislation, scrutiny of  527–8 EU membership  514–19 Fiscal Compact  519 human rights  512–13 mutual loyalty  524–5 national identity  511–13 neutrality  524 social rights  513 subsidiarity, principle of  529 transfer of powers to EU  519–25 monism  834 multilevel constitutionalism  56, 835 multilevel judicial dialogue  54 national identity  323, 451–2, 511–13, 516, 616–18, 814 Netherlands constitutional culture  194, 536–8 constitutional system  533–6 EU integration  544–5 EU law  546–9 EU membership  538–44 treaties  538–9 organisational diversity  798–9 Poland constitutional crisis  596–600, 826–9 constitutional culture  596–600 constitutional identity  615–19 constitutional reform  613–14 constitutional review  606–9 constitutional system  585–601 Constitutional Tribunal  592–6, 827–9 EU integration  606–11, 614–15 EU law conflicts of authority  611–13 constitutionality of  609–11 implementation  611–19 and national law  619–22 status of  604–6 EU membership  601–6 government  588–92 international law  604–5 judiciary  596–600, 826–9 national identity  616–18 political pluralism  591–2

rule of law  619–22, 826–7 separation and balancing of powers  592 Social Market Economy  592 sovereignty  588–90 Portugal constitutional amendments  633–5 Constitutional Court  632–3, 648–51 constitutional culture  631–3 constitutional system  627–31 ESM (European Stability Mechanism) treaty  643–4 EU integration  633–9, 640–4 EU law court enforcement  646–7 implementation  644–7 and national law  647–51 secondary legislation, scrutiny of  642–3 EU membership  635–6 Fiscal Compact  643–4 international law  637–9 judicial control  636–7 judiciary  630–1 legitimate expectations, principle of  199 separation of powers  630 sovereign debt crisis  649–51 TESM (Treaty on the European Stability Mechanism)  643–4 transfer of powers to EU  640–2 ultra vires doctrine  642–3 post-totalitarian constitutions  193–4 PSPP judgment (Germany)  821–6 Rechtsstaat constitutionalism  173–4, 193–6, 204–5 see also rule of law Romania constitutional culture  658–60 constitutional system  655–8 EU integration  664–8 EU law  668–76 EU membership  660–4 human rights  657 rule of law  193–202, 202–4, 619–22 SGP (Stability and Growth Pact)  770 Slovakia Constitutional Court  701–5 constitutional culture  705–6 constitutional system  699–705 EU integration  713–21 EU law  711–13 implementation  721–4 and national law  724 secondary legislation, scrutiny of  716–21 EU membership  707–13 international law  709–11 Slovenia constitutional amendments  685–7 Constitutional Court  692–5

Index  849 constitutional culture  682–4 constitutional system  679–82 courts  692–5 EU integration  687–9 EU law  689–97 court enforcement  691–2 implementation  689–95 and national law  695–7 EU membership  685–7 international law  687 National Assembly  689–90 treaties  687 Solange jurisprudence  132, 136, 146–8, 807–8 Spain Autonomous Communities  291–7 Charter of Fundamental Rights (EU)  297–300 constitutional culture  286–7 constitutional system  283–5 EU integration  290–1 EU law court enforcement  297–300 implementation of  291–300 and national law  301 transposition of  294–7 EU membership  287–90 human rights  285 Parliament  291–7 subsidiarity, principle of  293–4 Stability and Growth Pact (SGP)  770 subsidiarity, principle of  53, 238–9, 276–7, 293–4, 529 Sugar Quotas III case  70–1, 74–5, 77 Sweden constitutional culture  752–6 constitutional system  749–52 EU integration  770–1

EU law implementation  771–4 primacy of  761–70 EU membership  756–70 Fiscal Compact  770 judicial review  754–5 legislation  755–6 Stability and Growth Pact (SGP)  770 Treaty on Stability, Coordination and Governance (TSCG)  770 TESM (Treaty on the European Stability Mechanism) see ESM (European Stability Mechanism) treaty Treaty on Stability, Coordination and Governance (TSCG) see Fiscal Compact ultra vires doctrine  817–29 Austria  563–4 Czech Republic  78–9 Denmark  108–12 France  324–8 Germany  151–3 Ireland  231–2 Luxembourg  473–5 Portugal  642–3 United Kingdom constitution  777–8 constitutional culture  194, 778 constitutional statutes  786–8 EU law  788–92 EU membership constitutional foundations  778–82 constitutional limits  783–8 European Union Act 2011  783–6

850