The Legitimacy of EU Criminal Law 9781509919741, 9781509919772, 9781509919765

Citation preview

THE LEGITIMACY OF EU CRIMINAL LAW This book traces the history of the EU competence, EU policy discourse and EU legislation in the field of criminalisation from Maastricht until the present day. It asks ‘Why EU Criminal Law?’ looking at what rationales the Treaty, policy documents and legislation put forth when deciding whether a certain behaviour should be a criminal offence. To interpret the EU approach to criminalisation, it relies on both modern and post-modern theoretical frameworks on the legitimacy of criminal law, read jointly with the theories on the functions of EU harmonisation of national law. The book demonstrates that while EU constitutional law leans towards an effectiveness-based, enforcement-driven, understanding of criminal law, the EU has in fact in more than one instance adopted symbolic EU criminal law, ie criminal law aimed at highlighting what values are important to the EU, but which is not fit to actually deter individuals from harming such values. The book then questions whether this approach is consistent or in contradiction with the values-based constitutional identity the EU has set for itself. Volume 10 in the series Hart Studies in European Criminal Law

Hart Studies in European Criminal Law Series Editors: Professor Katalin Ligeti, University of Luxembourg; Professor Valsamis Mitsilegas, Queen Mary University of London; Professor Anne Weyembergh, Brussels Free University Since the Lisbon Treaty, European criminal law has become an increasingly important field of research and debate. Working with the European Criminal Law Academic Network (ECLAN), the series will publish works of the highest intellectual rigour and cutting edge scholarship which will be required reading for all European criminal lawyers. The series is happy to consider both edited and single authored titles. The series defines ‘European’ and ‘criminal law’ in the broadest sense, so books on European criminal law, justice and policy will be considered. The series also welcomes books which offer different methodological approaches. Volume 1: EU Criminal Law after Lisbon: Rights, Trust and the Transformation of Justice in Europe by Valsamis Mitsilegas Volume 2: Challenges in the Field of Economic and Financial Crime in Europe and the US Edited by Vanessa Franssen and Katalin Ligeti Volume 3: Chasing Criminal Money: Challenges and Perspectives On Asset Recovery in the EU Edited by Katalin Ligeti and Michele Simonato Volume 4: Limits to EU Powers: A Case Study of EU Regulatory Criminal Law by Jacob Öberg Volume 5: The Needed Balances in EU Criminal Law: Past, Present and Future Edited by Chloé Brière and Anne Weyembergh Volume 6: Redefining Organised Crime: A Challenge for the European Union? Edited by Stefania Carnevale, Serena Forlati and Orsetta Giolo Volume 7: White Collar Crime: A Comparative Perspective Edited by Katalin Ligeti and Stanislaw Tosza Volume 8: Criminal Liability of Managers in Europe: Punishing Excessive Risk Stanisław Tosza Volume 9: The European Union and Deprivation of Liberty: A Legislative and Judicial Analysis from the Perspective of the Individual Leandro Mancano

The Legitimacy of EU Criminal Law Irene Wieczorek

HART PUBLISHING Bloomsbury Publishing Plc Kemp House, Chawley Park, Cumnor Hill, Oxford, OX2 9PH, UK 1385 Broadway, New York, NY 10018, USA HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2020 Copyright © Irene Wieczorek, 2020 Irene Wieczorek has asserted her right under the Copyright, Designs and Patents Act 1988 to be identified as Author 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/, 1998–2020. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication data Names: Wieczorek, Irene, author. Title: The legitimacy of EU criminal law / Irene Wieczorek. Other titles: Legitimacy of European Union criminal law Description: Oxford ; New York : Hart, 2020.  |  Series: Hart studies in European criminal law ; volume 10  |  Based on author’s thesis (doctoral – Vrije Universiteit Brussel and Université Libre de Bruxelles, 2016) issued under title: The legitimacy of EU criminal law : what role for normative and instrumental justifications?  |  Includes bibliographical references and index. Identifiers: LCCN 2020009683 (print)  |  LCCN 2020009684 (ebook)  |  ISBN 9781509919741 (hardcover)  |  ISBN 9781509919758 (Epub) Subjects: LCSH: Law—European Union countries. Classification: LCC KJE7975 .W54 2020 (print)  |  LCC KJE7975 (ebook)  |  DDC 345.24001/1—dc23 LC record available at https://lccn.loc.gov/2020009683 LC ebook record available at https://lccn.loc.gov/2020009684 ISBN: HB: 978-1-50991-974-1 ePDF: 978-1-50991-976-5 ePub: 978-1-50991-975-8 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.

“To Danilo and to Maria and Gilda”

vi

Acknowledgements This book was written in Brussels, Cambridge and Durham. In each context, I have only found support and encouragement. My first and foremost gratitude goes to Paul De Hert for having given me a once-in-a-lifetime opportunity to pursue an intellectual career by offering the possibility to write a PhD at the Vrije Universiteit Brussels, on which this book is based, and for having guided me through the whole process with curiosity, geniality and nothing but encouragement. I would also like to thank Anne  Weyembergh, my co-supervisor, for her incomparable kindness and humanity which have been crucial in allowing me to bring this process to a good end, and for having helped me broaden my research interests and constantly find new eclectic passions. Finally, I would like to thank Ester HerlinKarnell, who has also closely followed and guided my progress during the PhD, and whose determination and ambition have always been exemplary to me. Alongside these three main advisors, I have also met some other outstanding scholars who have guided and supported me through the process and have given me many opportunities to develop. I would like to thank Gabriele Fornasari, John Spencer and Marise Cremona for having welcomed me in Trento, Cambridge and Florence respectively. In different contexts, I also benefited from the guidance of Jens Woelk, Carlo Sotis, Antje du Bois-Pedain, Alicia Hinarejos, Bruno de Witte, Claire Kilpatrick, Saskia Hufnagel and Shin Matsusawa, as well as the ECLAN Members, Sabine Gless, Robert Kert, Pedro Caeiro, Valsamis Mitsilegas, Katalin Ligeti, Rosaria Sicurella and Samuli Miettinen. Second, I would like to thank the Wiener-Anspach Foundation, which has granted me the opportunity to carry out my postdoctoral research both in Cambridge and the ULB, where the core of this book was written. In Cambridge I benefited from the outstanding guidance given by John Spencer and Nicky Padfield, who in particular re-awakened the criminal lawyer in me, helped me refine, academically and intellectually broadly speaking, and again stimulated me in discovering new passions. In Cambridge I have also had the invaluable opportunity to contribute to the Learning Together project – which has greatly enriched me intellectually – alongside amazing students and mentors, whom I also would like to thank. Third, I would like to thank Robert Schütze, Barend van Leeuwen, and Eleni Frantziou, who have warmly welcomed me in Durham, where this book was completed, for their support, encouragement and trust in my skills, which have all been greatly instrumental in the finalisation of this project.

viii  Acknowledgements Fourth, I would wholeheartedly like to thank the staff at Hart Publishing for their professionalism, kindness and enormous patience. Fifth, I would like to thank my friends Adriano Martufi, Angelo Marletta, Auke Willems, Dariusz Kloza, Radostina Primova, Mathias Holvoet, Leandro Mancano, Julia Muraskiewicz, Amy Weatherburn, Ines Armada, Marloes van Wijk, Jozefien van Caenheghen, Celine Cocq, Chloé Brière, Niovi Vavoula, Christina Peristeridou, Francesca Galli, Maria Haag, Kotomi Morigouchi, Sam Cole and Rezvan Farahibozorg, with whom I shared the happiness and sorrows of academic life, and my longtime friends Daniela, Giulia, Giorgia, Alessandro, Piera, Virginia, Renata, Marta and Ella. Finally, I would like to thank my extraordinary parents Mariapina and Paolo, who named me Irene Wieczorek because they thought it would be a good name for a writer. They are the most intellectually stimulating, wise and hilarious people I know. I would like to thank Michelino, who is always, infallibly, on my side, and Diana, who is lovingly by his, and lastly, I would like to thank my beloved Danilo, for his constant support, and especially for his unbeatable intellectual wit, humour and eye for challenges.

Contents Acknowledgements��������������������������������������������������������������������������������������vii Table of Cases���������������������������������������������������������������������������������������������xv Table of Legislation����������������������������������������������������������������������������������� xix Introduction��������������������������������������������������������������������������������������������������1 I. The Emerging Debate on the Legitimacy of EU Criminal Law..........4 II. How to Theorise on the Legitimacy of EU Criminal Law.................5 A. Research Questions and the General Purpose of the Book: An ‘Explanatory-Evaluative’ Theory of the Legitimacy of EU Criminal Law������������������������������������6 B. The Analytical Framework: A Combination of Criminalisation and Harmonisation Theories���������������������������������������������������7 C. What Indications are There from EU Constitutional Values and Principles for a Legitimate Use of EU Criminal Law?�������������������������������������������������������������������������7 D. How Does the EU Legitimise Resort to Criminal Law in Practice?������������������������������������������������������������������������������8 E. Are EU Criminalisation Choices Consistent with EU Constitutional Values and Principles?�������������������������������������10 F. Structure of the Book������������������������������������������������������������13 III. The Practical Importance of (EU Criminal Law) Theory................13 IV. The Broader Reach of the Discussion on the Legitimacy of EU Criminal Law..........................................................................15 A. EU Criminalisation and the Normative Foundations of the EU Criminal Justice Area���������������������������������������������16 B. EU Criminalisation and the EU Constitutional Identity����������17 1. Why Criminal Law? The Question of, and Models for, the Legitimacy of Criminal Law������������������������������������������������������������������������������������21 I. Why is Criminal Law Special? A Consequence-Based Approach......................................................................................22 A. The Impact on Individuals’ Liberty as a Defining Feature of Criminal Law��������������������������������������������������������������������23 B. The Impact on Individuals’ Dignity as a Defining Feature of Criminal Law��������������������������������������������������������������������25 C. Higher Procedural Safeguards as a Non-defining Feature of Criminal Law��������������������������������������������������������������������27

x  Contents II. III. IV. V.

VI. VII.

The Need to Legitimate Criminalisation Choices...........................28 A (Liberal) Deontological Approach to the Legitimacy of Criminal Law................................................................................30 A (Liberal) Utilitarian Approach to the Legitimacy of Criminal Law................................................................................33 A Joint Deontological-Utilitarian Approach to the Legitimacy of Criminal Law?........................................................35 A. Opposition and Reconciliation of Deontological and Utilitarian Approaches����������������������������������������������������35 B. The Empirical and Normative Advantages of a Joint Approach�������������������������������������������������������������������39 The Doctrinal Character of the Debate on the Legitimacy of Criminal Law...........................................................................40 Conclusions..................................................................................43

2. Why EU Criminal Law? The Question of, and the Models for, the Legitimacy of Supranational Criminal Law��������������������������������������44 I. Is the Debate on the Legitimacy of Criminal Law Relevant to the EU Legal Order?....................................................45 A. EU Definitions of Crimes and Member States’ Limited Discretion�����������������������������������������������������������������48 B. Enforcement of EU Criminal Law Obligations: Infringement Proceedings and the Bar on Direct Effect�����������51 C. Taricco and MAS and MB as a Challenge to the Bar on the Direct Effect?�������������������������������������������������������������������53 D. The Significant Impact of EU Criminalisation Choices and Their Ensuing Need for Legitimation������������������������������56 E. The Repressive Potential of EU Criminal Law and its Accentuated Need for Legitimacy������������������������������������������58 II. What Do EU Constitutional Values and Principles Tell Us about the Legitimacy of EU Criminal Law?...................................62 A. EU Values as a Foundation for a Deontological Approach to Criminalisation�������������������������������������������������62 B. EU Proportionality as a Neutral Framework to Incorporate Deontological and Utilitarian Considerations������������������������64 C. Subsidiarity: A Utilitarian Approach to EU Criminalisation�������69 III. Conclusions..................................................................................74 3. Rationales for the Harmonisation and Legitimacy of EU Criminal Law�����������������������������������������������������������������������������������76 I. A Definition of Harmonisation as a Legal Process.........................76 II. Does Harmonisation Have an Inherent, Values-Based Dimension?..............................................................78

Contents  xi A. Normative v Instrumental Types of Harmonisation���������������78 B. Rationales for Harmonisation and the Legitimacy of EU Criminal Law������������������������������������������������������������������80 III. Harmonisation of National Criminal Law within the EU Legal Order.............................................................................82 IV. The ‘Values-Based’ Criminalisation Rationale for Harmonisation: Deontological EU Criminal Law...........................84 V. The ‘Justice’, ‘Free Movement’ and ‘Cooperation’ Rationales for Harmonisation: Utilitarian EU Criminal Law...........................85 A. Instrumental Rationales for Harmonisation���������������������������86 B. … Implying a Utilitarian Legitimacy for EU Criminal Law�����88 VI. The Socialising Rationale for Harmonisation: Utilitarian EU Criminal Law.........................................................91 VII. The ‘Regulatory’ Rationale for Harmonisation: Utilitarian EU Criminal Law.........................................................92 III. Conclusions..................................................................................93 V 4. EU Competences on Securitised Criminalisation: From a Utilitarian to an Integrated Approach to EU Criminal Law�������������������������������������95 I. EU Criminalisation Competences: Securitised v Functional Criminalisation.........................................................96 II. A Utilitarian Legitimacy for EU Criminal Law under the Maastricht and Amsterdam Third Pillar.........................98 A. The Position for Harmonisation in the Treaty: Evidence of a Cooperation Rationale����������������������������������� 100 B. The Objective of an Area of Freedom, Security and Justice: A Free Movement and Justice Rationale for Harmonisation?������������������������������������������������������������������ 101 III. Integrated Legitimacy for EU Criminal Law under the Treaty of Lisbon......................................................................... 106 A. EU Criminalisation Competence in the Treaty of Lisbon: An Overview����������������������������������������������������������������������� 107 B. Article 83(1) as an Example of Integrated Legitimacy for EU Criminal Law����������������������������������������������������������� 113 IV. Conclusions................................................................................ 119 5. EU Competences on Functional Criminalisation: The Route to Utilitarian EU Criminal Law������������������������������������������������������������ 121 I. The Scope of EU Functional Criminalisation............................... 122 II. The Early Case Law: Sanctioning Obligations and an Integrated Approach to Criminal Law.............................. 123 III. The Case Law in the 2000s: Criminalisation Obligations and an Integrated Approach to EU Criminal Law......................... 126

xii  Contents A. The 2005 Environmental Crime Case: The Utilitarian Dimension of EU Criminal Law������������������������������������������� 127 B. The 2005 Environmental Crime Case: The Deontological Dimension of EU Criminal Law������������������������������������������� 128 C. The Commission’s and the Parliament’s Diverging Interpretations of the Environmental Crime Judgment��������� 131 D. The 2007 Ship Source Pollution Case: A Confirmed Integrated Agenda for the Legitimacy of EU Criminal Law?�����133 IV. The Treaty of Lisbon: The Final (Utilitarian) Word...................... 135 A. Article 83(2) TFEU: The Tension between Utilitarian EU Criminal Law and EU Values����������������������������������������� 136 B. Article 83(2) TFEU: Utilitarian EU Criminal Law and Subsidiarity������������������������������������������������������������������������� 138 V. Conclusions................................................................................ 139 6. From Tampere to Stockholm: The Path towards Integrated Legitimacy�����141 I. The Utilitarian Approach to Criminal Law in Pre-Lisbon Justice and Home Affairs Programmes........................................ 142 II. The Integrated Approach in the 2009 Stockholm Justice and Home Affairs Programme........................................................... 144 III. The Silence in the 2014 Justice and Home Affairs Strategic Guidelines.................................................................................. 147 IV. The EU Criminalisation Policy Documents: The Core of the Discussion............................................................................. 148 A. The 2009 Council’s Conclusions: A Deontological, Harm-Based Approach to EU Criminal Law������������������������� 151 B. The 2011 Commission’s Communication: Hinting at an Integrated Approach������������������������������������������������������������ 154 C. The 2012 Parliament Resolution: An Integrated, Damage-Based Approach����������������������������������������������������� 161 V. The Shift from a Utilitarian to an Integrated Approach in Policy Documents....................................................................... 164 A. The Growing Importance of the Deontological Approach to the Legitimacy of Criminal Law����������������������� 165 B. The EU Policy Documents and EU Treaty Norms: Synchronies and Ruptures���������������������������������������������������� 166 VI. Conclusions................................................................................ 167 7. Legitimating EU Criminal Law in Practice: The Case of Racism and Xenophobia, Market Abuse and PIF Crimes���������������������������������� 168 I. The Patchwork Structure of EU Criminal Law and the Choice of Case Studies................................................................ 169 II. The 2008 Framework Decision on Racism and Xenophobia: Symbolic EU Criminal Law under Amsterdam............................. 172

Contents  xiii A. The Values-Based Criminalisation Rationale of the Framework Decision: EU Values of Liberty, Democracy, and Individuals’ Safety�������������������������������������� 174 B. The Cooperation Rationale of the Framework Decision: Was Transnational Racist and Hate Speech Really the Problem?����������������������������������������������������������������������� 177 C. A Complementary Social Engineering Rationale for the Framework Decision?����������������������������������������������� 179 D. The Framework Decision as Symbolic EU Criminal Law in a Utilitarian Institutional and Policy Context?������������������ 180 III. The 2014 Market Abuse Directive: Symbolic EU Criminal Law under Lisbon?...................................................................... 181 A. The EU Fight against Market Abuse: The Road to Criminal Law���������������������������������������������������������������������� 183 B. The Offences in the Market Abuse Directive: Insider Dealing, Unlawful Disclosure of Information and Market Manipulation��������������������������������������������������� 184 C. The Regulatory Rationale in the Market Abuse Directive: Three Arguments in Favour of the Effectiveness of Criminal Law���������������������������������������������������������������������� 186 D. The Weakness of the Empirical Data on the Regulatory Rationale���������������������������������������������������������������������������� 188 E. The Normative Rationale of the Market Abuse Directive: The Seriousness Threshold�������������������������������������������������� 189 F. The Market Abuse Directive and Article 83(2) TFEU: An Integrated/Symbolic Approach in a Utilitarian Institutional Context?���������������������������������������������������������� 191 IV. The 2017 PIF Directive: An Integrated Legitimacy for EU Criminal Law?................................................................. 192 A. Explicit Rationales for Criminal Law in the Field of PIF Offences: An Apparent Focus on Utilitarian Grounds����������� 194 B. Deontological, Values-Based Rationales Guiding Definitions of Offences�������������������������������������������������������� 199 C. Deontological and Utilitarian Rationales in the Definitions of Sanctions������������������������������������������������������ 205 D. The PIF Directive as (Relatively) Soundly Legitimate EU Criminal Law���������������������������������������������������������������� 208 V. Conclusions................................................................................ 208 8. Conclusions����������������������������������������������������������������������������������������� 211 I. The Long-Standing Doctrinal Debate on the Legitimacy of Criminal Law......................................................................... 211 II. The Relevance and Legal Dimensions of the Debate on the Legitimacy of EU Criminal Law........................................ 212

xiv  Contents III. IV. V. VI.

The EU Constitutional Values and Principles and the Legitimacy of EU Criminal Law...................................... 213 The Specifics of the EU Criminalisation Process........................... 214 The Main Argument of the Book: Symbolic EU Criminal Law in a Bureaucratic Criminal Law Institutional Framework....................................................... 215 The Dangers of an Expansion of Non-legitimate EU Criminal Law............................................ 218

Bibliography���������������������������������������������������������������������������������������������� 221 Index��������������������������������������������������������������������������������������������������������� 235

Table of Cases European Court of Justice Judgments Case C-29/69 Erich Stauder v City of Ulm [1969] ECR I-00419���������������������18 Case C-11/70 Internationale Handelsgeselschaft mbH v Einfuhr- und Vorratsstelle für Getreide und Futtermittel [1970] ECR-I 01125����������������18 Case C-33/76 Rewe Zentralfinanz v Lanwirtschaftskammerfür das Saarland [1979] ECR I-1989������������������������������������������������������������ 124 Case 300/98 Commission of the European Communities v Council of the European Communities [1991] ECR I-02867������������������������������� 129 CaseC-4/73 J Nold, Kohlen- und Baustoffgroßhandlung v Commission [1977] ECR I-00001���������������������������������������������������������������������������������18 Case C-50/76 Amsterdam Bulb BV v Produktschap voor Siergewassen [1977] ECR I-00137������������������������������������������������������������������������������� 123 Case C-148/78 Criminal Proceedings against Ratti [1979] ECR I-01629���������52 Case C-8/81 Becker v Finanamt Munster-Innenstadt [1982] ECR I-00053������52 Case C-240/83 ADBHU [1985] ECR I-531��������������������������������������������������� 131 Case C-80/86 Criminal Proceedings against Kolpinghuis Nijmegen BV [1987] ECRI-03969���������������������������������������������������������������������������������52 Case C-302/86 Commission of the European Communities v Kingdom of Denmark [1988] ECR I-4607������������������������������������������������������������� 131 Case C-68/88 Commission of the European Communities v Hellenic Republic [1989] ECR I-02965�������������������������97–98, 122–27, 195 CaseC-331/88 R v Minister of Agriculture, Fisheries and Food ex Parte Fedesa [1990] ECR I-4023�����������������������������������������������������������66 Joined Cases C-6/90 and 9/90, Andrea Francovich and Danila Bonifaci and Others v Italian Republic [1991] ECR I-05357�����������������������������������51 Joined Cases C-49/93 and C-48/93 Brasserie du Pêcheur SA v Bundesrepublik Deutschland and The Queen v Secretary of State for Transport ex parte Factortame Ltd and Others [1996] ECR I-01029���������������������������������������������������������������������������������������������51 Joined Cases C-74/95 and C-129/95 Criminal Proceedings against X [1996] ECR I-06609���������������������������������������������������������������������������������52 Case C-213/96 Outokumpu Oy [1998] ECR I-1777������������������������������������� 131 Case C-17/98 Emesa Sugar (Free Zone) NV v Aruba [2000] ECR I-00665�������������������������������������������������������������������������������������������11 Case C-376/98 Federal Republic of Germany v European Parliament and Council of the European Union [2000] ECR I-8419������������������������� 219

xvi  Table of Cases Case C-491/01 The Queen v Secretary of State for Health ex parte British American Tobacco (Investments) Ltd and Imperial Tobacco Ltd [2002] ECR I-11453�������������������������������������������������������������67 Case C-105/03 Criminal Proceedings against Maria Pupino [2005] ECR I-05285��������������������������������������������������������������������������� 52, 97 Case C-176/03 Commission of the European Communities v Council of the European Union [2005] ECR I-7879���������������� 97, 127, 192 Joined Cases C-387/02, C-391/02 and C-403/02 Criminal Proceedings v Silvio Berlusconi and Others [2005] ECR I-03565����������������������������������52 Case C-440/05 Commission v Council [2007] ECR I-9097�������������� 97, 127, 192 Case C-45/08 Spector Photo Group NV, Chris Van Raemdonck v Commissie voor Bank-, Financie- en Assurantiewezen (CBFA) [2009] ECR I-12073���������������������������������������������������������������������������������50 Case C-58/08 The Queen on the Application of Vodafone Ltd and Others v Secretary of State for Business, Enterprise and Regulatory Reform [2010] ECR I-04999��������������������������������������������������67 CaseC-176/09 Grand Duchy of Luxembourg v European Parliament and Council of the European Union [2011] ECR I-03727�������������������������67 Case C-539/09 European Commission v Federal Republic of Germany [2011] ECR I-11235������������������������������������������������������������������������������� 200 Case C-348/09 PI v Oberbürgermeisterin der Stadt Remscheid [2012] ECR General Court Reports������������������������������������������������������� 115 Joined Cases 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 [2012] ECR General Court Reports�������������������������������������������������������������������68 Case C-617/10 Åklagaren v Hans Åkerberg Fransson [2013] ECR General Court Reports������������������������������������������������������������������������� 200 Case C-176/12 Association de médiationsociale (AMS) [2014] ECR General Court Reports���������������������������������������������������������������������������11 Case C-105/14 Criminal Proceedings against Ivo Taricco and Others [2015] ECR Court General Reports��������������������������������������������������53, 200 Case C-105/14 Criminal Proceedings against MAS and MB [2017] ECR Court General Reports�������������������������������������������������������������������53 Case C-574/15 Criminal Proceedings against Mauro Scialdone [2017] ECR General Court Reports���������������������������������������������������������60 European Court of Human Rights Judgments Perınçek v Switzerland, App No 27510/08 (ECHtR, 15 October 2015)�����������17 Reichman v France App No 50147/11 (ECtHR, 12 July 2016)������������������������66

Table of Cases  xvii National Judgments Belgium Cour Constitutionnelle de Belgique, (Belgian Constitutional Court) Judgment of 6 December 2012, n 145/2012���������������������������������������� 41, 66 Germany Bundesverfassungsgericht (German Federal Constitutional Court), Judgment of 30 June 2009, BVerfG, 2 BvE 2/08����������������������������������������14 Italy Corte Costituzionale della Repubblica Italiana (Constitutional Court of the Italian Republic), of 13 December 1988, n 1085����������������������� 41, 66 Corte Costitutionale della Repubblica Italiana (Constitutional Court of the Italian Republic), Judgment of 23 October 2006, n 393������������������55 Corte Costitutionale della Repubblica Italiana (Constitutional Court of the Italian Republic), Judgment of 30 July 2008, n 324�������������������������55 Lithuania Lietuvos Respubliskos Konstitucinis Tesimas (Constitutional Court of the Republic of Lithuania), Judgment of 13 November 1997, n 4/97����������������������������������������������������������������������������������������������� 41, 66 Spain Tribunal Constitucional (Spanish Constitutional Tribunal), Judgment of 3 March, n 43/2003�����������������������������������������������������������������������������27 Tribunal Constitucional (Spanish Constitutional Tribunal), Judgment of 20 December, n 246/2004�������������������������������������������������������������� 27–28 UK Woolmington v DPP [1935] UKHL 1 [1935] AC 462��������������������������������������28

xviii

Table of Legislation European Union Legislation Commission Decisions Commission Decision of 28 April 1999 establishing the European Anti-fraud Office (OLAF) (notified under document number SEC (1999) 802) [1999] OJ L136/20��������������������������������������������������������������� 199 Conventions Council Act of 26 July 1995 drawing up the Convention on the protection of the European Communities’ financial interests [1995] OJ C316/48����� 192 Council Act of 26 May 1997 drawing up, on the basis of Article K.3 (2) (c) of the Treaty on European Union, the Convention on the fight against corruption involving officials of the European Communities or officials of Member States of the European Union [1997] OJ C195/1��������������������������������������������������������������������������������� 97, 100–01 Council Act of 27 September 1996 drawing up a Protocol to the Convention on the protection of the European Communities’ financial interests [1996] OJ C313/1������������������������������������������ 61, 123, 192 Council Act of 19 June 1997 drawing up the Second Protocol of the Convention on the protection of the European Communities’ financial interests [1997] OJ C221/11���������������������������������������� 61, 123, 192 Council Act of 17 June 1998 drawing up the Convention on Driving Disqualifications (98/C 216/01) [1998] OJ C216/1������������������������������������88 Council Decisions Council Decision 2009/426/JHA of 16 December 2008 on the strengthening of Eurojust and amending Decision 2002/187/JHA setting up Eurojust with a view to reinforcing the fight against serious crime [2009] OJ L138/14�������������������������������������������������������������������������16 Council Decision 2009/371/JHA of 6 April 2009 establishing the European Police Office (Europol) [2009] OJ L121/37������������������������������ 115 Directives Council Directive 91/308/EEC of 10 June 1991 on prevention of the use of the financial system for the purpose of money laundering OJ [1991] OJ L166/77�������������������������������������������������������������������������������� 182

xx  Table of Legislation Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market [2000] OJ L178/1������������������������������������������������������������������������������������50 Directive 2001/97/EC of the European Parliament and of the Council of 4 December 2001 amending Council Directive 91/308/EEC on prevention of the use of the financial system for the purpose of money laundering [2001] OJ L344/76������������������������������������������������ 182 Directive 2003/6/EC of the European Parliament and of the Council of 28 January 2003 on insider dealing and market manipulation (market abuse) [2003] L 96/16���������������������������������������������������������������� 183 Directive 2003/33/EC of the European Parliament and of the Council of 26 May 2003 on the approximation of the laws, regulations and administrative provisions of the Member States relating to the advertising and sponsorship of tobacco products [2003] OJ L152/16������ 219 Directive 2004/38/EC of the European Parliament and of the Council 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 L158/77�������������������������������������������������������������������������������� 115 Directive 2005/35/EC of the European Parliament and of the Council of 7 September 2005 on Ship Source Pollution and the Introduction of Penalties for Infringements [2005] OJ L255/11����������������������������133, 192 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] L 309/15������������������������������������������������������������������������������������� 182 Directive 2008/99/EC of the European Parliament and of the Council of 19 December 2008 on the Protection of the Environment through Criminal Law [2008] OJ L328/28����������������������������������������������������������� 128 Directive 2008/99/EU of the European Parliament and of the Council of 19 December 2008 on the Protection of the Environment through Criminal Law [2008] OJ L328/28����������������������������������������������������������� 192 Directive 2009/52/EC of the European Parliament and of the Council of 30 September 2009 providing for minimum standards on sanctions and measures against employers of illegally staying third-country nationals [2009] OJ L168/24�������������������������������������������������������������51, 134 Directive 2010/64/EU on the right to translation and interpretation in criminal proceedings [2010] OJ L280/1����������������������������������������������������14

Table of Legislation  xxi Directive 2011/36/EU of the European Parliament and of the Council of 5 April 2011 on preventing and combating trafficking in human beings and protecting its victims and replacing Council Framework Decision 2002/629/JHA [2011] OJ L101/1������������������������������������� 48, 58, 60 Directive 2011/92/EU of the European Parliament and of the Council of 13 December 2011 on combating the sexual abuse and sexual exploitation of children and child pornography, and replacing Council Framework Decision 2004/68/JHA [2011] OJ L335/1������������� 49, 60 Directive 2012/13/EU of the European Parliament and of the Council of 22 May 2012 on the right to information in criminal proceedings [2012] OJ L142/1�����������������������������������������������������������������������������51, 218 Directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012 establishing minimum standards on the rights, support and protection of victims of crime, and replacing Council Framework Decision 2001/220/JHA [2012] OJ L315/57���������������������79, 218 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 218/8�������������������������������������������������������������������������������������� 48, 59, 61 Directive 2014/41/EU of the European Parliament and of the Council of 3 April 2014 regarding the European Investigation Order in criminal matters [2014] L 130/1����������������������������������������������������������������������������87 Directive 2014/57/EU of the European Parliament and of the Council of 16 April 2014 on criminal sanctions for market abuse (market abuse directive) [2014] OJ L173/17��������������������� 10, 14–15, 48, 169, 171–72, 181–92, 208–09, 217 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, and replacing Council Framework Decision 2000/383/JHA [2014] OJ L151/1����������������� 48, 61, 193 Directive 2015/849/EU of the European Parliament and of the Council of 20 May 2015 on the prevention of the use of the financial system for the purposes of money laundering or terrorist financing, amending Regulation (EU) No 648/2012 of the European Parliament and of the Council, and repealing Directive 2005/60/EC of the European Parliament and of the Council and Commission Directive 2006/70/EC [2015] L 141/73������������������������������������������������������������������������������������� 183 Directive 2017/541/EU of the European Parliament and of the Council of 15 March 2017 on combating terrorism and replacing Council Framework Decision 2002/475/JHA and amending Council Decision 2005/671/JHA [2017] OJ L88/6����������������������������������������������������������������48

xxii  Table of Legislation Directive 2017/1371/EU of the European Parliament and of the Council of 5 July 2017 on the fight against fraud to the Union’s financial interests by means of criminal law [2017] OJ L198/29�������������10, 46, 54–55, 85, 98, 110, 153, 169, 172, 192–210, 218 Directive 2018/843/EU of the European Parliament and of the Council of 30 May 2018 amending Directive (EU) 2015/849 on the prevention of the use of the financial system for the purposes of money laundering or terrorist financing and amending Directives 2009/138/EC and 2013/36/EU [2018] OJ L156/43����������������������������������������������������������������������������������56 Directive 2018/1673/EU of the European Parliament and of the Council of 23 October 2018 on combating money laundering by criminal law [2018] OJ L284/22���������������������������������������������������������������������������48, 183 Framework Decisions Council Framework Decision 2000/383/JHAof 29 May 2000 on increasing protection by criminal penalties and other sanctions against counterfeiting in connection with the introduction of the euro [2000] OJ L140/1�����������������������������������������������������������������������������61, 193 Council Framework Decision 2001/413/JHA on combating fraud and counterfeiting of non-cash means of payment [2001] OJ L149/1��������������48 Council Framework Decision 2002/584/JHA of 13 June 2002 on the European arrest warrant and the surrender procedures between Member States [2002] OJ L190/2������������������������������������������������������� 88, 90 Council Framework Decision of 13 June 2002 on the European arrest warrant and the surrender procedures between Member States – Statements made by certain Member States on the adoption of the Framework Decision [2002] OJ L190/1��������������� 16, 87, 178 Council Framework Decision 2002/475/JHA of 13 June 2002 on combating terrorism [2002] OJ L164/3����������������������������������������������������61 Council Framework Decision on Trafficking in Human Beings Council Framework Decision 2002/629/JHA of 19 July 2002 on combating trafficking in human beings [2002] OJ L203/1������������������������������������������60 Council Framework Decision 2003/80/JHA of 27 January 2003 on the protection of the environment through criminal law [2003] OJ L29/55����������������������������������������������������������������������������������������59, 126 Council Framework Decision 2003/568/JHA of 22 July 2003 on combating corruption in the private sector [2003] OJ L192/54������������������61 Council framework Decision 2004/68/JHA of 22 December 2003 on combating the sexual exploitation of children and child pornography [2004] OJ L13/44������������������������������������������������������������������������������������60 Council Framework Decision 2004/757/JHA of 25 October 2004 laying down minimum provisions on the constituent elements of criminal acts and penalties in the field of illicit drug trafficking [2004] OJ L335/8����������������48

Table of Legislation  xxiii Council Framework Decision 2005/222/JHA of 24 February 2005 on attacks against information systems [2005] OJ L69/67������������������ 58, 61 Council Framework Decision 2005/667/JHA of 12 July 2005 to strengthen the criminal-law framework for the enforcement of the law against ship-source pollution [2005] OJ L255/164��������������������������������� 59, 126, 133 Council Framework Decision 2008/841/JHA of 24 October 2008 on the fight against organised crime [2008] OJ L35/8�������������������������������49 Council Framework Decision 2008/913/JHA of 28 November 2008 on combating certain forms and expressions of racism and xenophobia by means of criminal law [2008] OJ L328/55������������ 10, 48–49, 74, 92, 101, 169, 172–81, 199, 209–10, 216 Council Framework Decision 2008/947/JHA of 27 November 2008 on the application of the principle of mutual recognition to judgments and probation decisions with a view to the supervision of probation measures and alternative sanctions [2008] OJ L237/27�����������������������������58 Council Framework Decision 2008/919/JHA of 28 November 2008 amending Framework Decision 2002/475/JHA on combating terrorism [2008] OJ L330/1������������������������������������������������������������������������������������61 Joint Actions Joint Action/96/443/JHA of 15 July 1996 adopted by the Council on the basis of Article K.3 of the Treaty on European Union, concerning action to combat racism and xenophobia [1996] OJ L185/5��������������������������������������������������������������������������������������101, 173 Joint Action 97/154/JHA of 24 February 1997 adopted by the Council on the basis of Article K.3 of the Treaty on European Union concerning action to combat trafficking in human beings and sexual exploitation of children [1997] OJ L63/2������������������������������������������������������������������� 101 Joint Action 98/733/JHA of 21 December 1998 adopted by the Council on the basis of Article K.3 of the Treaty on European Union, on making it a criminal offence to participate in a criminal organisation in the Member States of the European Union [1998] OJ L351/1����������������������� 101 Joint Action 98/742/JHA of 22 December 1998 adopted by the Council on the basis of Article K.3 of the Treaty on European Union, on corruption in the private sector [1998] OJ L358/2������������������� 56, 58, 101 Regulations Regulation (EU, Euratom) 883/2013/EU of the European Parliament and of the Council of 11 September 2013 concerning investigations conducted by the European Anti-Fraud Office (OLAF) and repealing Regulation (EC) No 1073/1999 of the European Parliament and of the Council and Council Regulation (Euratom) No 1074/1999 [2013] OJ L248/1������������� 193

xxiv  Table of Legislation Regulation 596/2014/EU of the European Parliament and of the Council of 16 April 2014 on market abuse (market abuse regulation) and repealing Directive 2003/6/EC of the European Parliament and of the Council and Commission Directives 2003/124/EC, 2003/125/EC and 2004/72/EC [2014] OJ L173/1���������������������������������������������������������������� 184 Regulation 2016/679/EU 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 L119/1���������������������������������������������������������������������������������� 219 Regulation 2016/794/EU on the European Union Agency for Law Enforcement Cooperation (Europol) and replacing and repealing Council Decisions 2009/371/JHA, 2009/934/JHA, 2009/935/JHA, 2009/936/JHA and 2009/968/JHA [2016], OJ L135/14������������������������������16 Council Regulation 2017/1939 of 12 October 2017 implementing enhanced cooperation on the establishment of the European Public Prosecutor’s Office (‘the EPPO’) [2017] OJ L283/1���������������4, 14, 16, 45–46, 87, 111–12, 148, 193, 199–202 International Treaties Council of Europe Convention for the Protection of Human Rights and Fundamental Freedoms, Rome, 4.XI.1950 ETS No. 005������������������������������������������������41 European Convention on Mutual Legal Assistance in Criminal Matters, Strasbourg, 20.IV.1959 ETS No. 30����������������������������������������������������������87 Additional Protocol to the Convention on Cybercrime, Strasbourg, 28.I.2003, ETS No. 189������������������������������������������������������������������������� 176 European Union Treaty of the European Community [Consolidated in Amsterdam] – Protocol on the application of the principles of subsidiarity and proportionality [1997] OJ C340/105������������������������������������������������������� 71 Treaty on the Functioning of the European Union – Protocol (No 2) on the application of the principles of subsidiarity and proportionality [2008] OJ C115/206��������������������������������������������������������������������������������11 Treaty of the Functioning of the European Union – Protocol (No 3) on the Statute of the Court of Justice of the European Union [2012] OJ C326/210����������������������������������������������������������������������������������������� 124 Treaty on the Functioning of the European Union – Protocol (No 22) on the position of Denmark [2012] OJ C326/299������������������������������14, 110

Table of Legislation  xxv Treaty on the Functioning of the European Union – Protocol (No 21) on the position of the United Kingdom and Ireland in respect of the area of freedom, security and justice [2016] C 202/295����������������������14, 110 Other Protocol of San Luis concerning judicial cooperation in criminal matters (CMC 2/96) applying to MERCOSUR (Southern Common Market, Mercado Comùndel Sur) States�����������������������������������������������������������������1

xxvi

Introduction

G

lobalisation is often accused of having disrupted the strong historical tie between states’ sovereignty and criminal law, stripping nation states of their monopoly over the administration of criminal justice.1 Such characterisation is not necessarily accurate. Interesting examples of non-state criminal law jurisdiction already existed in both distant and recent history. These included Jewish and Moore communities’ penal jurisdiction over their members in the fifteenth-century Kingdom of Portugal,2 Native Americans’ Indian penal jurisdiction within the US, which still exists today,3 and the Church’s jurisdiction over canon criminal law.4 Yet, it is undeniable that the last century has seen an unprecedented proliferation of non-state penal jurisdictions outside the borders of the state, next to the pre-existing ones inside national frontiers. Transnational criminal justice actors, in the form of organisations of states, have been created at the global level (the International Criminal Court and the UN criminal justice system)5 and at the regional level in Europe (the European Union (EU) and the Council of Europe), Southeast Asia,6 Latin America7 and West and Southern Africa.8 Next to these public law bodies, transnational private actors, and especially non-governmental organisations (NGOs) like ‘WikiLeaks’ or ‘Transparency International’, have also significantly engaged in the fight against crime, particularly corruption. In this sense, these too could be

1 See, among others, M Delmas Marty, Pour un droit commun (Paris, Le Seuil, 1994). 2 P Caeiro, ‘The Relationship between European and International Criminal Law (and the Absent(?) Third)’ in V Mitsilegas, M Bergstrom and T Konstantidines (eds), Research Handbook on EU Criminal Law (Cheltenham, Edward Elgar, 2016) 580, 583. 3 ibid 584. 4 ibid 583. 5 This expression refers to the net of international criminal law conventions approved either in the remit of the United Nations Office on Drugs and Crime (UNODC) or with the involvement of any other UN stakeholders. See further R Slawomir, The United Nations Criminal Justice System in the Suppression of Transnational Crime’ in N Boister and R Currie (eds), Routledge Handbook of Transnational Criminal Law (Abingdon, Routledge, 2015) 57. 6 eg, the Association of Southeast Asian Nations (ASEAN) and the South Asian Association for Regional Cooperation (SAARC). 7 See the Protocol of San Luis concerning judicial cooperation in criminal matters (CMC 2/96) applying to MERCOSUR (Southern Common Market, Mercado Comùn del Sur) states. See S Mendes de Sousa, Cooperaçao Juridica Penal no Mercosul: novas possibilidades (Rio de Janeiro, Renovar, 2011) 193–228; R Cervini and J Tavarez, Principios de Cooperaçao Judicial Penal Internacional no Protocolo do Mercosul (Sao Paulo, Editora Revista dos Tribunais, 2000) 211. 8 See the West Africa Coast Initiative (WACI) and the Southern African Regional Police Chiefs Co-operation Organisation (SARPCCO). On regional integration on criminal matters, see V Mitsilegas, ‘Regional Organization and the Suppression of Transnational Crime’ in Boister and Currie (n 5) 73.

2  Introduction enlisted as transnational criminal justice actors.9 Lastly, Kotiswaran and Palmer also suggest that even colonial criminal law could qualify as transnational criminal law since, technically, it is an example of criminal law developed in more than one state.10 Such a proliferation of criminal law beyond the state raises a question of legitimacy on multiple dimensions. First, transnational criminal law raises a political legitimacy question. Criminal punishment has traditionally been perceived as the most elementary and obvious expression of the state’s sovereign power, except for the examples of intra-state criminal law jurisdictions mentioned above. It is generally acknowledged that the state is the only legitimate owner of the monopoly on the use of force over its citizens. Why transnational actors should also benefit from this prerogative is not self-evident. However, regulation of criminal law jurisdiction can be quite appealing for supranational entities, especially the EU, from an instrumental perspective. In light of its close connection to state sovereignty, criminal law jurisdiction is an excellent medium for a supranational entity to claim state-like features and thus to increase political legitimacy.11 However, one might question the democratic legitimacy of these bodies, enabling them to enact binding criminal law imposing obligations on citizens and limiting their fundamental rights. Second, transnational criminal law raises a social, or internal, legitimacy question. There is strong consensus among criminologists that the legitimacy of the public authority, such as the police or the judges, in the eyes of citizens is key to ensure their compliance with the law.12 Would citizens, and especially the interested sub-group of offenders, perceive non-state actors as legitimate criminal justice actors? Would they perceive the transnational criminal law they produce as fair? Finally, there is then a normative or fundamental rights legitimacy question. Criminal law is, according to most, a special regulatory tool, not only because of its closeness to sovereignty, but also because it can have an important impact on the liberty of individuals. A long-standing criminal legal theory debate exists on the use that legislators make or should make of this instrument, as will be extensively explained in Chapter 1. Yet, how do these new transnational actors

9 J Nederveen Pieterse, ‘Leaking Superpower: WikiLeaks and the Contradictions of democracy’ (2012) 33(10) Third World Quarterly 1909. See also the recent role played by Transparency International in the legal investigation of corruption scandals in Brazil and the Dominican Republic: https:// www.transparency.org/news/feature/brazil_open_data_just_made_investigating_corruption_easier. 10 P Kotiswaran and N Palmer, ‘Rethinking the “International Law of Crime”: Provocations from Transnational Legal Studies’, King’s College London Dickson Poon School of Law Legal Studies Research Paper Series, 2015–34. 11 C Sotis, ‘Criminaliser sans punir. Reflexions sur le pouvoir d’incrimination (directe et indirecte) de l’Union européènne prévu par le traité de Lisbonne’ (2010) 4 Revue de science criminelle et de droit pénal comparé 773, 774. 12 A Crawford and A Hucklesby (eds), Legitimacy and Compliance in Criminal Justice (Abingdon, Routledge, 2013).

Introduction  3 justify resort to this regulatory instrument when they have the choice between penal, administrative and civil sanctions? What is or should be a transnational crime? Do and should the same theories apply at a national level and a supranational level? A vast scholarship exists on the political, social and fundamental rights legitimacy of international criminal law, and the law of the international criminal law tribunals in particular has been the subject of considerable discussion. Normative and social legitimacy rationales for international criminal law include deterrence, accountability and providing a forum where the voices of the victims could be heard, as well as a historical record of the events,13 or even broadly protecting ‘peace, security and [the] well-being of the world’ as the fundamental values of the international community.14 The democratic credentials of international criminal justice have also been debated, with some authors criticising the International Criminal Court’s stronger attention on seeking accountability with their parent bodies, including states and the UN, rather than with their end users, such as victims.15 Moreover, interesting analysis exists on the application of criminalisation criteria at the international level.16 Conversely, the debate on the legitimacy of criminal law adopted at the EU level (EU criminal law), especially on its normative legitimacy, is less developed. This is probably due to the still relatively young age and patchwork structure of the EU Criminal Justice Area. Contrary to the international criminal justice system, which dates back to the end of the Second World War, EU intervention in criminal matters only started in 1992 with the Treaty of Maastricht. Moreover, the international criminal justice system has both prescriptive jurisdiction (definition of the law) and adjudicative jurisdiction (investigation, prosecution and conviction). Conversely, the EU mainly has the capacity to legislate on criminal

13 K Cronin-Furman and A Taub, ‘Lions and Tigers and Deterrence, Oh My: Evaluating ­Expectations of International Criminal Justice’ in WA Schabas, Y McDermott and N Hayes (eds), The Ashgate Research Companion to International Criminal Law (Farnham, Ashgate, 2013) 435; MR Damaska, ‘What is the Point of International Criminal Justice?’ (2008) 83(1) Chicago-Kent Law Review 329, 331. 14 G Werle and F Jessberger, Principles of International Criminal Law, 3rd edn (Oxford, Oxford University Press 2014) 33; O Triffterer, ‘Preamble’ in O Triffterer (ed), Commentary on the Rome Statute of the International Criminal Court, 2nd edn (Oxford, Hart Publishing, 2008) para 9; MC Bassiouni, ‘The Philosophy and Policy of International Criminal Justice’ in LC Vohrah, F Pocar, Y Featherstone, O Fourmy, MF Graham, J Hocking and N Robson (eds) Man’s Inhumanity to Man: Essays on International Law in Honour of Antonio Cassese (Leiden, Brill, 2003) 65. See also the well-articulated discussion on the purposes of international criminal law and punishment in K Ambos, Treatise on International Criminal Law. Volume 1: Foundations and General Part (Oxford, Oxford University Press 2013) 56–73. 15 For a discussion of the democratic character of international criminal law as opposed to other sources of legitimacy, see F Mégret, ‘International Criminal Justice: A Critical Research Agenda’ in C Schwöbel (ed), Critical Approaches to International Criminal Law: An Introduction (Abingdon, Routledge, 2014) 26–28. 16 M Renzo, ‘A Criticism of the International Harm Principle’ (2010) 4(3) Criminal Law and Philosophy 267.

4  Introduction justice matters, whereas it currently has a very limited enforcement capacity and then only in the field of crimes against the financial interests of the EU.17 Yet, if one considers that EU action in criminal matters occurs as part of a broader EU legal order, the need to investigate what justifies resort to supranational criminal law and whether this is consistent with the broader EU normative framework is particularly important. This book aims to complement the emerging discussion on the legitimacy EU criminal law, and especially on its normative, criminal legal theory, dimension. Section I clarifies which aspects remain underexplored in this area. Section II clarifies the specific research question that the book addresses and the relevant methodology. Section III illustrates the practical added value of enquiring into the legitimacy of EU criminal law for EU policy-making. Finally, section IV discusses the scientific added value of the research, in that it can be instructive as to the normative foundations of the whole EU Criminal Justice Area and in relation to the EU constitutional identity. I.  THE EMERGING DEBATE ON THE LEGITIMACY OF EU CRIMINAL LAW

An extensive literature exists on the political legitimacy of the EU, and especially on the question of the EU democratic deficit.18 Peristeridou,19 Muñoz de Morales Romero20 and Oberg21 have further discussed how arguments of democracy apply to the criminal law field. The question of social legitimacy of EU criminal law has also received some theoretical attention, though predominantly with respect to the norms on transfers of prisoners.22 A broader debate on the social legitimacy of the EU Criminal Justice Area is lacking. As for normative legitimacy, the policy objectives for different EU criminal law instruments have

17 The European Public Prosecutor has the competence to investigate and prosecute these crimes: Council Regulation (EU) 2017/1939 of 12 October 2017 implementing enhanced cooperation on the establishment of the European Public Prosecutor’s Office (EPPO) (hereinafter the EPPO Regulation). The EPPO is briefly discussed in ch 7. 18 For a summary of the debates on the democratic deficit, see J Weiler, ‘European Models: Polity, People and System’ in P Craig and C Harlow (eds), Lawmaking in the European Union (Boston, Kluwer, 1998) ch 1. 19 See C Peristeridou, The Principle of Legality in EU Criminal Law (Antwerp, Intersentia, 2015) ch 8. 20 M Muñoz de Morales Romero, ‘In Pursuit of Basics for a New Principle of Legal Reserve in Supranational Criminal Law’ (2012) 3(2) EU Criminal Law Review 252. 21 J Öberg, ‘EU Criminal Law, Democratic Legitimacy and Judicial Review of Union Criminal Law Legislation in the Wake of the Lisbon Treaty’ (2011) 6 Tilburg Law Review 60. 22 I Wieczorek, ‘EU Constitutional Limits to the Europeanization of Punishment: A Case Study on Offenders’ Rehabilitation’ (2019) Maastricht Journal of European and Comparative Law, https:// journals.sagepub.com/doi/metrics/10.1177/1023263X18820692; E de Wree, T Vander Beken and G Vermeulen, ‘The Transfer of Sentenced Persons in Europe’ (2009) 11(1) Punishment and Society 111; I Durnescu E Montero Perez de la Tudela and L Ravagnani, ‘Prisoner Transfer and the Importance of the “Release Effect”’ (2016) 17(4) Criminology and Criminal Justice 450.

How to Theorise on the Legitimacy of EU Criminal Law  5 been the subject of considerable reflection. Projects like the 1999 Corpus Iuris justified the use of criminal law by the EU with the need to protect specific EU financial interests, which Member States would neglect.23 The same arguments have been revitalised in the debate on the European Public Prosecutor Office.24 In other cases, EU definitions of crimes are justified on the basis of their greater effectiveness, when compared to national criminal law, against cross-border crime.25 Further, the need to use criminal sanctions to enforce EU policies is also listed among the possible rationales.26 Lastly, other authors have identified the willingness to show symbolic commitment to certain important values as underpinning some EU instruments.27 However, these discussions on the content of EU criminal law have not been systematised. In particular, an explicit criminal legal theory-informed approach, investigating what doctrinal understanding of criminal law underpins EU criminalisation choices, has rarely been adopted. In 2011, a group of 14 scholars from different EU countries founded the European Criminal Policy Initiative (ECPI) and called for a general reflection on the rationales guiding EU criminalisation. They openly raised the question of the legitimacy of EU criminal law from a criminal legal theory perspective, asking what interests the EU intends to protect through its criminal law.28 Yet, the impact of this project remained quite limited since the authors only analyse a number of selected EU secondary law provisions. This book aims to provide such an analysis of the normative legitimacy of EU criminal law relying on a systematic criminal legal theory framework. For the sake of simplicity, here and in the following chapters, the subject matter of the book will be referred to hereinafter as ‘the legitimacy of EU criminal law’. II.  HOW TO THEORISE ON THE LEGITIMACY OF EU CRIMINAL LAW

Theorising on transnational criminal law can be a challenging exercise. Zumbansen warns us that ‘the emerging, and quickly overwhelming, world of

23 M Delmas-Marty et al, Corpus Juris: Introducing Penal Provisions for the Purpose of the ­Financial Interests of the European Union (Paris, Economica, 1997). 24 European Commission, Staff Working Document, Impact Assessment Accompanying the Proposal for a Council Regulation on the Establishment of the European Public Prosecutor’s Office, SWD (2013) 274 final, 18. 25 I Wieczorek, ‘The Principle of Subsidiarity in EU Criminal Law’ in C Brière and A Weyembergh, The Needed Balances in EU Criminal Law: Past, Present and Future (Oxford, Hart Publishing, 2017). 26 E Herlin-Karnell, The Constitutional Dimension of EU Criminal Law (Oxford, Hart Publishing, 2012). 27 T Elholm and R Colson, ‘The Symbolic Purpose of EU Criminal Law’ in R Colson and S Field (eds), EU Criminal Justice and the Challenges of Diversity: Legal Cultures in the Area of Freedom, Security and Justice (Cambridge, Cambridge University Press 2016) 48. 28 European Criminal Policy Initiative, ‘The Manifesto on European Criminal Policy’ (2011) 1(1) European Criminal Law Review 86.

6  Introduction transnational governance needs to be treated as more than a new playground for well-worn and established rules, principles and order concepts’.29 Jurists are bound to test and question whether well-established doctrinal categories and concepts retain meaning in a transnational context of regulatory institutions, processes and ideas.30 The following sections illustrate how the book addresses these challenges when theorising on EU criminal law. Section II.A announces the three research questions and general methodology of the book. Section II.B clarifies the analytical framework of the research. Sections II.C–E detail the specific methods and sources selected in order to answer each of the three research questions. Finally, section II.F outlines the structure of the book. A.  Research Questions and the General Purpose of the Book: An ‘Explanatory-Evaluative’ Theory of the Legitimacy of EU Criminal Law The book addresses three questions: • First, what indications do EU constitutional values and principles provide on the legitimate use of criminal law by the EU? • Second, how does the EU justify resort to criminal law in practice? • Third, are such EU criminalisation choices consistent with EU constitutional values and principles? The book’s general aim is to develop an ‘explanatory-evaluative theory’, borrowing the term from Moore’s ‘theory of criminal legal theories’.31 The book does not develop an autonomous theoretical model for the legitimacy of EU ­criminal law; rather, it aims to examine which legitimacy model of criminal law the EU legal order has embraced and critically assess consistency of EU ­criminal law with EU constitutional choices. The research has both an explanatory dimension, in that it looks at the EU approach to criminalisation, and an evaluative dimension, in that it tests such an approach against the backdrop of EU constitutional law. The book thus asks: is the EU approach to criminalisation consistent with EU constitutional law? This is an internal coherence question, rather than a general question asking whether the EU approach to criminalisation is inherently valuable. Finally, the book maintains a historical dimension, looking at how the approach to criminalisation has evolved, and the degree of coherence which has been maintained with EU constitutional law from Maastricht up to the present day.

29 P Zumbansen, ‘Theorising as Activity: Transnational Legal Theory in Context’ in ­Christopher McCrudden, Upendra Baxi and Abdul Paliwala (eds), Law’s Ethical, Global and Theoretical Contexts: Essays in Honour of William Twining (Cambridge, Cambridge University Press, 2015) 280, 285. 30 Zumbansen (n 29). 31 M Moore, Placing Blame: A Theory of the Criminal Law (Oxford, Oxford University Press, 2010) 9 ff.

How to Theorise on the Legitimacy of EU Criminal Law  7 B.  The Analytical Framework: A Combination of Criminalisation and Harmonisation Theories A vast debate exists on the legitimacy of criminal law, which identifies different rationales for criminalisation. When designing the analytical framework, both the deontological approaches to criminalisation (the Anglo-Saxon harmbased approach and its continental ‘legal interests’ counterparts) and utilitarian accounts of the criminal law (Benthamite utilitarianism initially and its postmodern reformulations by law and economics scholars) will be considered.32 These theories have been developed with reference to national criminalisation processes, where a legislator adopts criminal provisions that are directly applicable to citizens. As will be further explained in Chapters 2 and 3, criminalisation at the supranational level occurs through EU harmonisation of national criminal law. The EU sets standards which Member States have to incorporate into their national criminal justice systems. Thus, national and supranational criminalisation have different objectives and might be subject to different constitutional constraints. As such, once possible models for the legitimacy of criminal law have been identified, one has to look at the objectives or the rationales for the harmonisation of definitions of crimes and assess what approach to the legitimacy of criminal law they embody. In other words, it is necessary to tailor national criminalisation theories to the transnational context. This book thus combines criminalisation debates with theories on supranational harmonisation of criminal law as developed by Weyembergh, Klip and van der Wildt, and Delmas-Marty, Pieth and Sieber.33 C.  What Indications are There from EU Constitutional Values and Principles for a Legitimate Use of EU Criminal Law? When addressing the first research question, namely what indications do EU constitutional values and principles provide for a legitimate use of EU criminal law, the book first looks at Article 2 of the Treaty on European Union (TEU). This lists the EU founding values, among which the most relevant for this analysis are those of human dignity and human freedom, as well as respect for fundamental rights. The relationship between penal law and these liberal values, and especially the legitimacy model of EU criminal law that would better protect them, is conceptualised following modern criminal legal theory tenets. More specifically, the book looks at how the Anglo-Saxon harm principle, and the German and Italian versions of Rechtsgut Theory have interpreted the relationship between criminal law and individual freedom or liberty, the two most



32 See 33 See

also ch 1 for an extensive definition of these terms. extensively ch 3.

8  Introduction relevant values when discussing criminalisation.34 The research also incorporates some reflections on how utilitarian theories of criminal law impact the freedom of individuals.35 Furthermore, it also looks at economics and social sciences literature on criminal sanctions and stigma to define the relationship between criminal law and human dignity.36 Second, the book considers other general constitutional limits to EU action, namely the principle of proportionality (Article 5(4) of the Treaty on the European Union (TEU), but see also Article 52(1) of the EU Charter of Fundamental Rights) and the principle of subsidiarity (Article 5(3) TEU). These are general principles meant to constrain EU action in all policy areas.37 Any resort to EU criminal law must naturally fit within the broader role envisaged for EU law in regulating the lives of individuals. The relationship between penal law and the EU principles of proportionality and subsidiarity is based on previous research published elsewhere,38 and other authors’ contributions to the emerging literature on the general principles of EU criminal law, including Oberg,39 and the Manifesto on European Criminal Policy.40 D.  How Does the EU Legitimise Resort to Criminal Law in Practice? When looking at the second research question, namely what approach the EU has adopted to criminalisation in practice, the book looks at primary law listing the competence to define crimes and sanctions, at policy documents containing indications on how such competences should be exercised and, finally, at secondary law containing concrete examples of EU criminalisation. As for the choice to combine criminalisation theories with harmonisation theories discussed above,41 the decision to look at these three different sources directly flows from the transnational context of the analysis and especially the multi-level structure of the EU. When analysing the rationales for criminalisation at a national level, the obvious choice is to look at actual criminal law provisions and possibly at parliamentary debates which have preceded their adoption. Yet, when looking at the EU legal order, it is interesting to adopt a broader approach, starting from EU primary law. Contrary to national constitutional law, which 34 See ch 1, s III and ch II, s II.A. 35 See ch 1, s IV and ch II, s II.A. 36 See ch I, s I,B. 37 To be precise, subsidiarity only applies to areas of shared competences between the EU and Member States (art 5(3) TEU). In any case, the Area of Freedom Security and Justice, which includes judicial cooperation in criminal matters, is among the areas of shared competences (art 4(2)(j) TFEU), so subsidiarity is relevant for this discussion. 38 Wieczorek (n 25). 39 J Öberg, ‘Subsidiarity and EU Procedural Criminal Law’ (2015) 5 European Criminal Law Review 19. 40 European Criminal Policy Initiative (n 28). 41 See s II.B.

How to Theorise on the Legitimacy of EU Criminal Law  9 normally contains only indirect references to rationales for the use of penal law,42 EU primary law can also provide substantial insights on the place of criminal law in the EU legal order. Allocating powers between the supranational level and the national level is one of the constitutional challenges that the EU faces. The need to prevent excessive infringement upon the prerogatives of Member States requires a clear delimitation of the EU’s remit of action. The EU Treaties first limit the scope of EU action by identifying EU competences. These are often shaped in a ‘purposive way’,43 namely they indicate the objectives which need to be achieved, rather than simply indicating the subject matter which belongs to either the supranational or the national level. This means that looking at EU criminal law competences can already indicate the policy areas that the EU believes deserve the introduction of criminal sanctions, and the objectives that the EU should pursue through criminal law.44 Second, EU Institutions’ declarations on EU criminal policy objectives are also of interest. EU integration in the field of justice and home affairs has been marked by the adoption of multi-annual programmes setting general objectives for this policy area, including objectives concerning the harmonisation of national criminal law. The programmes adopted so far include the Tampere Programme,45 the Hague Programme,46 the Stockholm Programme47 and the 2014 Strategic Guidelines.48 With the adoption of the Lisbon Treaty, which significantly amended the EU competences in criminal matters, the European Council,49 the European Commission50 and the European Parliament51 have also adopted ad hoc policy documents on the harmonisation of criminal law.

42 See extensively ch 1, s VI. 43 G Davies, ‘Democracy and Legitimacy in the Shadow of Purposive Competences’ (2015) 21(1) European Law Journal 2. 44 Admittedly, some federal or decentralised states’ constitutions also contain provisions similarly distributing competences among the different levels. Nonetheless, these provisions often only mention whether criminal law lies with either the higher level or the local level, without any general indication. On mentions to criminal law in the constitutions of federal states, see art 74(1) of the German Constitution, arts 10 and 15 of the Austrian Constitution, s 149 of the Spanish Constitution and art 117(2)l of the Italian Constitution. 45 European Council Conclusions of 15 and 16 October 1999, http://www.europarl.europa.eu/ summits/tam_en.htm#c. 46 European Council Conclusions, The Hague Programme: Strengthening Freedom, Security and Justice in the European Union [2005] OJ C53/1. 47 European Council Conclusions, The Stockholm Programme: an open and secure Europe serving and protecting citizens [2010] OJ C115/1. 48 European Council Conclusions, 26–27 June 2014, Brussels, Doc EUCO 79/14. 49 Council of the European Union, Council Conclusions on model provisions, guiding the Council’s criminal law deliberations, JHA Council, 30 November 2009, published in A Weyembergh and F Galli (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Éditions de l’Université de Bruxelles, 2013) 226. 50 European Commission, ‘Towards an EU Criminal Policy: ensuring the effective implementation of EU policies through criminal law’ COM (2011) 573 final. 51 European Parliament, Resolution of 22 May 2012 on an EU approach to criminal law, P7_TA(2012)0208.

10  Introduction All of these documents naturally represent precious resources for investigating the EU approach to criminalisation. Finally, the book looks at how EU criminalisation choices are justified in practice by examining three case studies among the over 30 instruments adopted by the EU including criminalisation provisions. The selected case studies are the 2008 Framework Decision on Racism and Xenophobia,52 the 2014 Market Abuse Directive53 and the 2017 Directive on the Protection of the Financial Interests of the EU (hereinafter the PIF Directive).54 As will be better explained in Chapter 7, the criteria for selecting the case studies were twofold and had the overall aim of providing an exhaustive picture of the evolution of EU criminal law: first, case studies which could show how the greatest rationales for the use of criminal law play out in practice were included; and, second, both pre-Lisbon and post-Lisbon legislation has been selected, and one of the latest adopted instruments was included, notably the PIF Directive. Each case study, as well as the analysis of EU competences and EU policy documents, assesses how each source justifies resort to criminal law and thus which model of legitimacy of criminal law they implement. E.  Are EU Criminalisation Choices Consistent with EU Constitutional Values and Principles? The last step of the analysis is to assess the coherence between the EU approach to criminalisation and the EU constitutional law framework. The question of coherence between EU criminal (binding and soft) law is understood as that of ‘embodying the same values and approach’. It should not, as a matter of principle, be understood as a question of legal compliance with hierarchically superior norms, which can be the object of judicial review. The choice between a ‘coherence’ question rather than a ‘legality’ question is dictated by the selected constitutional benchmark, the sources for the analysis of EU criminal law and the type of analysis. First, concerning the benchmarks, as a general rule, provisions on values, like Article 2 TEU, can have a double function. In specific cases, they can be legally justiciable; yet, they are normally only conceived as a ‘policy-shaping guideline’. In other words, they set normative preferences on how policies should be shaped, without actually allowing for an individual to have an act invalidated

52 Council Framework Decision 2008/913/JHA of 28 November 2008 on combating certain forms and expressions of racism and xenophobia by means of criminal law [2008] OJ L328/55. 53 Directive 2014/57/EU of the European Parliament and of the Council of 16 April 2014 on criminal sanctions for market abuse [2014] OJ L173/17. 54 Directive 2017/1371/EU of the European Parliament and of the Council on the fight against fraud to the Union’s financial interests by means of criminal law [2017] OJ L198/29.

How to Theorise on the Legitimacy of EU Criminal Law  11 in court if that specific guideline is not followed.55 Within the EU legal order, a specific mechanism to enforce EU values in Member States is envisaged in Article 7 TEU. Interestingly, this is a political mechanism, as opposed to a legal one, as it entrusts the Council of Ministers and the European Council, rather than the Court of Justice, with the task of deciding on the adequacy of protection for EU values in Member States. Admittedly, Article 2 TEU also includes fundamental rights, which are traditionally conceived to be legally justiciable, among EU founding values. However, the EU Charter explicitly marks a distinction between provisions embodying rights that need to be respected and provisions (including principles) that must be observed and promoted. In some cases, the Court of Justice has even considered that some fundamental rights provisions are not specific enough to be directly applicable to a case.56 Subsidiarity and proportionality are also conceived as legally justiciable principles. However, especially in the case of subsidiarity, the political origins of the principle, and the very light touch of the Court’s scrutiny, have raised doubts as to its nature. The issue of whether the principle is a legal or a political principle meant to become part of law-making culture has been the subject of discussion.57 Significantly, the Lisbon Treaty has introduced a new political mechanism for the enforcement of subsidiarity.58 Briefly, the constitutional benchmarks for the analysis in the book do not always lend themselves to a legal discussion. Second, whilst admitting that the relevant EU constitutional norms would set a legally justiciable benchmark, the sources selected for the analysis are not in all cases suitable for legal analysis. Definitions of competences and EU founding values and principles are all included in Treaty norms. They have the same rank in the hierarchy of norms. Thus, an inconsistency between the definition of competences and EU values would not make the former unlawful for not respecting hierarchically superior norms. The Court of Justice has only in very specific cases admitted that EU primary law should be subject to fundamental rights scrutiny.59 A question of legality is therefore hardly relevant here.

55 However, see Murswiek, who argues that art 2 is as such legally enforceable against Member States before the Court: D Murswiek, ‘The Stealthy Development of the Treaty on European Union into the Supreme European Constitution’, www.jura.uni-freiburg.de/institute/ioeffr3/forschung/ papers.php. 56 See CJEU, Association de médiation sociale (AMS) C-176/12 on art 27 of the EU Charter. 57 On the legal and political character of subsidiarity, see P Craig, ‘Subsidiarity: A Political and Legal Analysis’ (2012) 50(1) Journal of Common Market Studies 72. 58 See the Early Warning Mechanism at Consolidated Version of the Treaty on the Functioning of the European Union – PROTOCOLS – Protocol (No 2) on the application of the principles of subsidiarity and proportionality [2008] OJ C115/206. 59 In the Emesa Sugar case, the Court implicitly admitted that the statutes of the European Courts (EU primary law) should respect the right to fair trial: Case C-17/98 Emesa Sugar (Free Zone) NV v Aruba, Order of the Court of 4 February 2000, para 18.

12  Introduction Yet, there is an interest in assessing whether EU primary law is, as a whole, ‘normatively coherent’. At the very least, it would be odd to have the EU selecting liberal values as founding values and then designing EU competences that permit resort to illiberal criminal law. Having said that, conversely, there is no interest in assessing the coherence between the definition of competences and the principles of subsidiarity and proportionality. These two principles are meant to bind the exercise of EU competences rather than their definition. Thus, when discussing EU criminal law competences, the main focus will only be on their coherence with EU values and fundamental rights. The relevant chapters discussing Treaty norms will only look at whether EU primary law includes any indications on when the exercise of these competences might be legitimate. The legality question also appears to be of little interest when looking at EU policy documents. The Area of Freedom, Security and Justice multi-annual programmes, and the Institutions’ declarations on EU criminalisation are soft law instruments and are hierarchically subordinate to EU constitutional law. However, since they are non-binding, it would be of little use to reflect on their lawfulness. Conversely, it is interesting to look at how the Institutions declare EU values and principles should be applied in criminal law and how seriously the task has been undertaken. Finally, the question of legality is admittedly pertinent when discussing EU secondary law. EU framework decisions and directives are naturally meant to respect EU constitutional law norms, including EU values and fundamental rights, proportionality and subsidiarity. However, in this case, the kind of analysis requires a coherence rather than a legal compliance approach. Traditionally the choice on the functions of criminal law has been considered to be a political one falling outside the scope of legally justiciable policy choices, within the discretion of the law makers. As will be made clear in Chapter 1, despite a few exceptions, scholars have not theorised that legal obligations exist for legislators to embrace one model of criminal law legitimacy over another. Instead, they have identified desirable models, justifying the resort to criminal law, which would protect certain values or pursue certain objectives, and have critically assessed the coherence of legislative developments. Similarly, national courts have not upheld a certain approach to criminalisation, whether deontological or utilitarian, as a binding one. The judiciary has at most assessed whether or not the criminalisation choices were in compliance with general principles such as proportionality or the principle of equality.60 Briefly, in the context of criminalisation, values and fundamental rights, especially the value of liberty, have been considered and interpreted as fulfilling the first of their functions mentioned above, namely that of a policy-shaping guideline. The book will accordingly mainly focus on the legitimacy of EU criminalisation choices where legitimacy is understood as coherence with the normative premises for the use of criminal law



60 See

ch 1, s VI.

The Practical Importance of (EU Criminal Law) Theory  13 derived from EU values and general principles. It will only occasionally point out when specific issues of secondary law legal compliance with subsidiarity or proportionality arise. F.  Structure of the Book In terms of structure, the book is composed of seven chapters excluding the Introduction and Conclusions sections. Chapter 1 discusses the problem of the legitimacy of criminal law in general terms. It explains why the question ‘when is the use of criminal law legitimate?’ arises. It then illustrates different ­criminalisation theories, which provide models for the legitimacy of criminal law, and their connection to liberal values. Chapter 2 discusses why the problem of the legitimacy of criminal law is relevant to the EU. It further assesses the models for the legitimacy of EU criminal law and determines which one is most consistent with the EU’s commitment to liberal values and principles. Chapter 3 designs the analytical tool for the study on the legitimacy of EU criminal law. It discusses how criminalisation theories apply at the supranational level, where criminalisation occurs through the harmonisation of national criminal law. The following four chapters contain the bulk of the analysis, looking at the legitimacy model of criminal law that can be inferred from different sources, and to what extent it is consistent with the EU constitutional law framework. Chapters 4 and 5 look at EU criminal law competences. Chapter 6 analyses policy documents. Chapter 7 discusses the three case studies. The concluding section summarises the findings on the EU understanding of the legitimacy of criminal law and discusses the internal coherence of EU criminalisation policy choices with the EU system of values. III.  THE PRACTICAL IMPORTANCE OF (EU CRIMINAL LAW) THEORY

There are a number of reasons to embark on theoretical analysis, such as that on the EU understanding of the normative legitimacy of criminal law. For instance, Boister underlines how a taxonomy of law – an ultimate theoretical exercise – makes it possible to identify doctrinal weakness in a field of law (eg, the lack of guiding principles) and thus provides an indication on how to correct them.61 Moreover, theorising can help to improve the understanding and thus transparency of a field of law.62 The EU in particular has long struggled with the 61 N Boister, ‘Transnational Criminal Law?’ (2003) 14 European Journal of International Law 953, 958. 62 On the role of criminal legal theory in achieving transparency, see W Hassemer, ‘The Harm Principle and the Protection of “Legal Goods” (Rechtsgüterschutz): A German Perspective’ in AP Simester, A du Bois Pedain and U Neumann (eds), Liberal Criminal Theory (Oxford, Hart Publishing, 2018).

14  Introduction question of good governance and on how to better connect to its citizens, thus making its rule-making process more transparent.63 EU criminal law is a significantly contested field. Past weak democratic legitimacy, or lack thereof, has long been a ‘pathology’ of this field of law.64 Criticism has been directed quite simply at the legitimacy of the EU intervening in criminal matters. This is explicit in the German Constitutional Court’s Lisbon judgment,65 but also underpins the opt-outs from this field that Ireland, the UK and Denmark enjoy from this policy area.66 Furthermore, some of the greatest EU policy achievements in this area have been the object of significant censure, as shown by the constitutional decisions saga on the European Arrest Warrant,67 the resistance to the harmonisation of procedural rights68 or the opposition by national governments69 and parliaments70 to the European Public Prosecutor Office. EU criminal law is thus in dire need of further clarity and transparency. The connection between clarity and transparency in the policy-making and legitimacy of the EU is well epitomised in the words of the MEP Cornelis de Jong when he explained the European Parliament’s commitment to start a reflection on the directions of the EU criminal policy. He warned us that: The European Union is facing a crisis of legitimacy. In 2005, the European Constitution was voted down both in France and in the Netherlands. Nevertheless, a similar Treaty was drafted, called the Lisbon Treaty, and was ratified without additional referenda in these two countries. One of the features of the Lisbon Treaty was to abolish the right of Member States to wield a veto in the field of criminal law … it

63 For a summary of the debates on EU governance methods, see P Craig and G de Búrca, EU Law: Text, Cases and Materials (Oxford, Oxford University Press, 2015). On specific problems related to transparency, see Anaïs Berthier, Transparency in EU Law-Making (Era Forum 2016); and M Hillebrandt, D Curtin and A Meijer, ‘Transparency in the EU Council of Ministers: An Institutional Analysis’ (2014) 20(1) European Law Journal 1. 64 P de Hert, ‘Division of Competences between National and European Levels with Regard to Justice and Home Affairs’ in M Anderson and J Apap (eds), Police and Justice Cooperation in the New European Borders (Boston, Kluwer Law International, 2002) 65. 65 Judgment of 30 June 2009, Bundesverfassungsgericht, BVerfG, 2 BvE 2/08, para 249. 66 Treaty on the Functioning of the European Union – PROTOCOLS – PROTOCOL (No 21) on the position of the United Kingdom and Ireland in respect of the area of freedom, security and justice [2016] OJ C202/295; Protocol (No 22) on the position of Denmark [2012] OJ C326/299. 67 E Guild (ed), Constitutional Challenges to the European Arrest Warrant (Tilburg, Wolf Legal Publishers, 2006). 68 Back in 2004, the Commission had proposed a framework decision on certain procedural rights (COM/2004/0328 final). The proposal failed to attract support from Member States and it was only in 2010 that a first, sectorial text on the rights to translation and interpretation was adopted (Directive 2010/64/EU on the right to translation and interpretation in criminal proceedings [2010] OJ L280/1). 69 The EPPO Regulation was approved by 20 Member States, as eight Member States decided not to participate in enhanced cooperation. See recitals 8 and 9 of the EPPO Regulation (n 17). 70 National parliaments raised a ‘yellow card’ against the proposed regulation in the framework of the Early Warning Mechanism. For an overview of the objections, see I Wieczorek, ‘The EPPO Draft Regulation Passes the First Subsidiarity Test: An Analysis and Interpretation of the Commission’s Hasty Approach to National Parliaments’ Subsidiarity Arguments’ (2015) 16(5) German Law Journal 1247.

The Broader Reach of the Discussion  15 is a huge step for Member States to give up part of their sovereignty in such a sensitive area. It illustrates the gap between European policymakers and the European public. Had we asked the public for their opinion, the majority would probably have rejected the criminal law provisions of the Treaty. However, we did not do that, and we are therefore now engaged in law-making in this area without knowing for certain whether we are actually acting in accordance with the wishes of the electorate … Against this background, I considered that, now that the Lisbon Treaty has not only done away with the veto power of Member States in the field of criminal law but also created co-legislative rights for the European Parliament, we should at least create a framework for law making in this field.71

Finally, and most importantly, there are also some very practical reasons for the significance of (criminalisation) theory. As mentioned above and as will be extensively explored in Chapter 1, criminal law can potentially heavily affect the lives of citizens. Contrary to what the unprecedented spread of the criminal law might lead one to think,72 policy makers are, at least on some level, aware of this. They concretely ask themselves the question of whether the use of criminal law is justified as a policy option in specific cases. For instance, in the course of the negotiations on the EU Market Abuse Directive, Member State delegations explicitly addressed the question of the criminalisation criteria and the majority of them did ‘not consider that the distinction between the administrative offences … and the criminal offences … should be based solely on the element of intent’.73 Theoretical work in this context can play a crucial role in supporting wise policy making by providing sophisticated and sound analysis in order to address very practical questions. IV.  THE BROADER REACH OF THE DISCUSSION ON THE LEGITIMACY OF EU CRIMINAL LAW

On a more abstract level, analysing the EU approach to criminalisation can be illuminating in relation to two broader debates: first, the explanatory dimension of the book, which looks at the EU approach to criminalisation, can be instructive as to the legitimacy of the whole EU Criminal Justice Area; and, second, the evaluative part of the book, which looks at the coherence between the EU approach to criminalisation and the EU constitutional framework can provide interesting insights on the EU constitutional identity.

71 C de Jong, ‘The European Parliament Resolution of 22 May 2012 on an EU Approach to Criminal Law’ in Weyembergh and Galli (n 49) 37, 38. 72 The problem of overcriminalisation is particularly accentuated in the Anglo-Saxon world; see A Ashworth, ‘Is the criminal law a lost cause?’ (2000) 116 LQR 225; and Douglas Husak, Overcriminalisation (Oxford, Oxford University Press, 2008). 73 Note of the Council Presidency to the Delegations, Proposal for a Directive of the European Parliament and of the Council on criminal sanctions for insider dealing and market manipulation – outstanding issues, Doc 12089/12, pt 4.

16  Introduction A.  EU Criminalisation and the Normative Foundations of the EU Criminal Justice Area Traditionally, the values that lie at the heart of a criminal justice system, be they liberal or authoritarian, shape both the substance of the criminal law and its procedural norms. For instance, the Nazi reforms of the German criminal justice system, which undermined the core of the pre-existing Weimar liberal system, both restricted procedural liberties (such as freedom from unlawful arrest, search and seizure) and widely enlarged the range of substantive criminal law, introducing severe punishments against political, racial and religious minorities.74 In the same way, Eastern European regimes exiting the communist era jointly addressed the reform of substantive and procedural criminal law, purging both of repressive elements, as well as enshrining liberal values in national constitutions.75 More recently, it was argued that the legal and economic movement has had a significant impact on the criminal justice system, influencing both substantive criminal law (eg, implying an increase in the use of fixed penalties) and procedural criminal law (eg, greater use of summary trials).76 The EU did not build a fully fledged criminal justice system. However, the harmonisation of the definition of crimes does not exhaust the kinds of EU intervention in criminal matters. The EU has also enacted legislation harmonising procedural criminal law,77 creating frameworks for judicial cooperation among Member States,78 and setting up EU criminal justice and intelligence agencies, such as Eurojust,79 Europol80 and the European Public Prosecutor.81 Investigating the normative foundations of the EU definition of crimes provides insight into the legitimacy of the whole system. If one ascribes to the assumption that the EU Criminal Justice Area is also a normatively coherent whole, the examination of a part (EU substantive criminal law) offers an insight on the philosophy underpinning the whole. Admittedly, the assumption of normative coherence between various branches of a criminal justice system is not always valid.82 Even so, there are still technical reasons which make an enquiry into the legitimacy of criminalisation a 74 F Hoefer, ‘The Nazi Penal System – I’ (1944–45) 35 Journal of Criminal Law & Criminology 385. 75 See R Alogna and S Riondato (eds), Studies on the Criminal Law Reform in the Post-Soviet Countries (Padua, Padua University Press, 2013). 76 L Zedner and A Ashworth, ‘Defending the Criminal Law: Reflections on the Changing Character of Crime, Procedure, and Sanctions’ (2008) 2 Journal of Criminal Law and Philosophy 21. 77 See n 68 above. 78 eg, Framework Decision 584/2002/JHA, the European arrest warrant and the surrender procedures between Member States [2002] OJ L190/1. 79 Decision 2009/426/JHA on the strengthening of Eurojust and amending Decision 2002/187/JHA setting up Eurojust with a view to reinforcing the fight against serious crime [2009] OJ L138/14. 80 Regulation 2016/794/EU on the European Union Agency for Law Enforcement Cooperation (Europol) and replacing and repealing Council Decisions 2009/371/JHA, 2009/934/JHA, 2009/935/ JHA, 2009/936/JHA and 2009/968/JHA [2016] OJ L135/14. 81 EPPO Regulation (n 17). 82 Interestingly, for instance, after 1945, Italy amended its code of criminal procedure, but maintained a fascist illiberal code of criminal law for much longer. On fascist Italian criminal law,

The Broader Reach of the Discussion  17 key one. This is because the legitimacy of procedural criminal law is necessarily tangled up with that of substantive criminal law. In his passionate plea in favour of the study of substantive criminal law, Kadish explains that ‘there can be no authorized police inquiry, arrest, prosecution, conviction, or sentence which is not based on the rules and doctrines of the substantive criminal law. This is the law which finally legitimates or fail[s] to legitimate what is done by the agencies of criminal justice’.83 In short, substantive criminal law triggers the application of procedural criminal law. If there is no crime, there can obviously be no trial. This subordinate relationship has arguably specific consequences in terms of the interrelation between the legitimacy of the two branches of criminal law. This is best exemplified when looking at the relationship between criminal law and fundamental rights. In this context, one could indeed speak of an ‘upstream’ interaction between the definitions of crimes and fundamental rights, and derivative interactions between criminal procedure norms and fundamental rights. For instance, considering of hate speech as a crime instantly compresses freedom of expression.84 Yet, it also determines that if the prescription is violated and a criminal sanction is imposed, compression of other possible fundamental rights might occur. These might be the result of criminal proceedings (eg, the right to privacy being compressed by wiretapping) and of the execution of a criminal sentence (eg, the right to liberty and/or the right to property being compressed by imprisonment and/or fines). As a logical consequence, the justification for compressing citizens’ liberty via criminal law provisions provide a necessary, although obviously not sufficient, upstream legitimation for the compression of their liberties by means of criminal procedural law. Thus, the legitimacy of criminal procedural norms partially rests on the legitimacy of substantive criminal law. Enquiring into the latter necessarily provides initial answers in relation to the former. Briefly, when looking into the legitimacy of EU substantive criminal law, that is, the definition of crimes, one gains an insight into a necessary component for the legitimacy of EU procedural criminal law, be this in the form of judicial cooperation or harmonisation of procedural criminal law. B.  EU Criminalisation and the EU Constitutional Identity Finally, if one bears in mind the connection between criminal law and fundamental rights, looking at the legitimacy of EU criminal law can be instructive

see Paul Garfinkel, Criminal Law in Liberal and Fascist Italy (Cambridge, Cambridge University Press, 2016). On the current Italian criminal law code, see Stephen Skinner, ‘Tainted Law? The ­Italian Penal Code, Fascism and Democracy’ (2011) 7(4) International Journal of Law in Context 423. 83 Sanford H Kadish ‘Why Substantive Criminal Law – A Dialogue’ (1980) 29 Cleveland State Law Review 1, 5. 84 See Perınçek v Switzerland, App No 27510/08 (ECHtR, 15 October 2015), where the Court went so far as to consider criminal prohibitions of denying the Armenian genocide to be disproportionate.

18  Introduction not only as to the criminal justice policy area, but also for broader debates on EU constitutional identity. The connection between the two can be best understood when recalling the role of the protection of fundamental rights in the process of EU integration. Protection of fundamental rights was not initially incorporated into the EC Treaties. The project for a European Political Community, inherently connected to the European Defence Community, had included a role for the Community to review fundamental rights protection within Member States.85 The political context was the original one where the horrors of the two world wars and totalitarian regimes were still very much alive, and where political cooperation was seen as a means to achieve peace.86 However, when the project of establishing the European Political and Defence Communities failed, the inclusion of the protection of fundamental rights within the EC Treaties was also temporarily abandoned. It was only 10 years later, in the late 1960s, that the Court of Justice re-introduced the idea of a supranational protection of fundamental rights. National courts (especially the German and Italian ones) were resistant to the invasiveness of EU law in the national legal orders. In particular, they wanted to retain their jurisdiction to check the compatibility of EU law with national constitutional standards.87 In response, the Court introduced the protection of fundamental rights as a general principle of EC (at the time) law. There was thus no need for a national fundamental rights check on EU law, since this was preliminarily carried out by the EC (at the time) institutions when assessing EU law compliance with the general principles of EC (at the time) law.88 Ensuring supranational protection of fundamental rights was thus a way for the EC to gain bargaining power with respect to national courts when requiring the primacy of EC law over national law.89 When compared with the initial approach in the Treaty on a European Political Community, one can easily appreciate a shift in focus. While initially the protection of fundamental rights had been introduced as an supranational standard with which Member States had to comply, when the Court of Justice finally acknowledged it as a general principle, this was to include an supranational double-check on its own action. Moreover, contrary to the initial project, the EU commitment to fundamental rights was manifestly an instrumental one, 85 G de Búrca, ‘The Evolution of EU Human Rights Law’ in P Craig and G de Búrca (eds), The Evolution of EU Law (Oxford, Oxford University Press, 2011) 465, 467–71. 86 ibid 471. 87 ibid 478–79. 88 Case 29/69 Stauder v City of Ulm [1969] ECR I-00419; Case 4/73 J Nold, Kohlen- und Baustoffgroßhandlung v Commission [1977] ECR I-00001; Case C-11/70 Internationale Handelsgeselschaft mbH v Einfuhr- und Vorratsstelle für Getreide und Futtermittel [1970] ECR-I 01125. 89 Azoulai speaks of protection of fundamental rights as ‘means of actions’ for the Court to expand the EU field of action: L Azoulai, The Case of Fundamental Rights: A State of Ambivalence’ in H-W Micklitz and B de Witte (eds), The European Court of Justice and the Autonomy of the Member States (Antwerp, Intersentia, 2012) 207.

The Broader Reach of the Discussion  19 aimed at securing the integrity and uniform application of EU law. Yet, the implications of identifying fundamental rights as a general principle of EU law rapidly went further than this initial instrumental goal. Protection of fundamental rights grew to contribute not only to the legitimacy of the EU integration project as a whole, but also to the definition of its identity.90 In 1973, the European Council declaration on European identity included respect for fundamental rights as an intrinsic element of the European identity.91 Moreover, the 1978 Copenhagen Declaration listed respect for fundamental rights as a condition for EC (at the time) membership. The Maastricht Treaty then gave Treaty recognition to the obligation to respect fundamental rights.92 The Treaty of Amsterdam listed respect for fundamental rights together with other values such as freedom, democracy and the rule of law as EU founding principles.93 This clause was arguably inserted to enshrine the Copenhagen political criteria for accession within the Treaty.94 The Treaty of Amsterdam would also explicitly mention these criteria in the provision for accession.95 The listed principles thus attributed the EU identity to outsiders. As mentioned above, the Treaty of Amsterdam also introduced into Article 7 TEU a procedure to suspend voting rights in the Council for those Member States that were found to be in consistent and serious breach of the principles listed in Article 6 TEU. This brought attention to the initial function of EU fundamental rights in the European Political Community project as a monitoring instrument for the situation in Member States. The harshness of the sanction and the political character of the procedure have rendered its use quite difficult in practice.96 However, on a symbolic level, Article 7 TEU highlighted the high importance within the EU legal order of the protection of liberal values, including the protection of fundamental rights. Failure to respect them potentially entailed partial ‘exclusion from the club’ through the suspension of voting rights. In this sense, the EU founding principles also clarified the EU identity to its members. 90 G de Búrca, ‘The Language of Rights and European Integration’ in J Show and G More (eds), New Legal Dynamics of European Union (Oxford, Clarendon Press, 1995) 43. 91 European Council, Declaration on European Identity, Bulletin of the European Communities, December 1973, No 12. Luxembourg: Office for Official Publications of the European Communities, 118–22. 92 Article 6 TEU, consolidated in Maastricht [1992] OJ C191/1. 93 Article 6 TEU, consolidated in Amsterdam [1997] OJ C340/145. 94 But see contra D Murswiek, ‘The Stealthy Development of the Treaty on European Union into the Supreme European Constitution’ (2009) Paper and Preprints of the Institute of Public Law 2 and the references mentioned therein – available at: https://www.jura.uni-freiburg.de/de/institute/ioeffr3/ forschung/papers/murswiek/IOER_The_stealthy_development_of_the_TEU_preprint.pdf. 95 See art 49 TEU, consolidated in Amsterdam [1997] OJ C340/145. 96 So far, the procedure has not been activated against any Member State. However, the ­European Parliament has been asked by the EU Council to consider whether to activate it against Hungary for the human rights and rule of law situation in the country; see https://www.euronews. com/2018/09/17/rule-of-law-in-poland-or-lack-thereof-7-questions-for-the-article-7-hearing-view. Moreover, the Commission has requested the Council to consider whether to activate it in response to the alleged dangers to the rule of law in Poland; see http://europa.eu/rapid/press-release_ IP-18-4341_en.htm#_ftn1.

20  Introduction In the Treaty of Lisbon, the fundamental rights dimension acquired even more visibility.97 The EU identity-related principles are mentioned in Article 2 TEU and are referred to as founding values rather than principles, arguably giving them a more ‘sacred’ dimension. Moreover, Article 6 TEU attributed a binding value to the EU Charter of Fundamental Rights and provided a legal basis for the EU accession to the European Convention on Human Rights. Briefly, fundamental rights, and EU values in general, became over time more than just one constitutional benchmark for EU secondary law among others, like subsidiarity and proportionality. They acquired a further role of defining the identity of the EU project, thus providing it with greater normative legitimacy (to refer back to the introduction’s distinction among different kinds of political social and normative legitimacy). In this, the EU is part of a broader trend which, since the Second World War, has granted human rights a ‘symbolic pre-eminence’ as an instrument for polity legitimation.98 In the context of European integration, such a symbolic legitimating function works particularly well because fundamental rights are perceived as a specific European common heritage, something ‘archetypically’ European and thus part of the EU identity.99 As a consequence, complying with fundamental rights, or more broadly with what the EU lists as its founding values, goes beyond a mere rule of law question as it is a matter of being consistent with the EU self-appointed identity. However, the extent to which the EU has taken the commitment to fundamental rights and liberal values seriously is contested. Criticism has been raised as to whether the EU incorporated fundamental rights considerations into its own policies, with criminal justice being one of the key policy fields considered to be lacking in this respect.100 Alongside this, a vivid debate also exists on the alleged ‘hypocrisy’ of the EU, which uses less strict standards when reviewing respect of fundamental rights within Member States than when assessing the performance of candidate countries.101 In light of the above, looking at whether EU criminalisation choices are coherent with EU values can have some interesting implications which go beyond a mere investigation of criminal law theory. It can contribute to this broad debate by providing empirical insights as to whether or not the EU is actually acting in compliance with its own, self-appointed identity. 97 De Búrca speaks of a coming of age of fundamental rights in this period: de Búrca (n 85) 481. 98 D Chalmers, G Davies and G Monti, European Union Law (Cambridge, Cambridge University Press, 2010) 233. 99 On this, see A Williams, EU Human Rights Policies: A Study in Irony (Oxford, Oxford U ­ niversity Press, 2004) 139. Williams also illustrates how the early declarations and the preamble spoke of regaining a pre-existing commonality between the peoples of Europe which had been lost due to the war (see especially the European Coal and Steal Community Treaty’s preamble at 169). 100 See P de Hert, ‘EU Criminal Law and Fundamental Rights’ in Mitsilegas, Bergström and Konstantinides (n 2) 105. 101 The critique is that the EU arguably allows more leeway to its Member States to depart from its core, identity-defining norms than it does to third states who want to cooperate with the EU. For a very detailed analysis of internal identity discourse and the external identity discourse, and a criticism of the EU lack of internal commitment to the protection of fundamental rights, see Williams (n 99) ch 7.

1 Why Criminal Law? The Question of, and Models for, the Legitimacy of Criminal Law

T

he question of the legitimate content of criminal law, namely of what behaviours are or should be a crime, is among the most ­discussed in criminal legal scholarship. The debate has lasted for at least two ­centuries1 and has sparked global interest, simultaneously developing in Continental Europe (Germany and Italy) and in the Anglo-Saxon world (the UK and the US).2 Interesting enquiries also exist into the Australian3 or African criminal justice systems.4 Moreover, the debate has a strong multidisciplinary character. While being a core criminal legal theory question, discussions on the content of criminal law have implications for moral philosophy,5 criminology,6 law and economics,7 constitutional and fundamental rights law,8 and political theory.9 Without any ambition to exhaust this debate in all its geographical extension and multidisciplinarity, this chapter provides an overview of this discussion, with the aim of delineating the theoretical background for the book’s discussion on EU criminal law. In other words, before turning to the question ‘why EU criminal law?’, this chapter explains the interest in asking ‘why criminal law?’ in the first place and identifies the analytical tools to address the question. 1 See the discussion in sections II–IV. 2 See the discussion in sections II–IV. 3 C Lauterwein, The Limits of Criminal Law: A Comparative Analysis of Approaches to Legal Theorizing (Farnham, Ashgate, 2010). 4 E Alemika et al, The Theory and Practice of Criminal Justice in Africa (Tswhane, Institute for Security Studies, 2016). 5 SP Green, Lying, Cheating, and Stealing: A Moral Theory of White-Collar Crime (Oxford, Oxford University Press, 2005). 6 JQ Wilson and RJ Herrnstein, Crime and Human Nature (New York, Simon & Schuster, 1985). 7 See, inter alia, G Becker, ‘Crime and Punishment: An Economic Approach’ in G Becker and W Landes (eds), Essays in the Economics of Crime and Punishment (Cambridge, MA, NBER, 1974) 45. 8 On the relationship between criminal legal theory and constitutional law, see, inter alia, A du Bois Pedain, ‘The Place of Criminal Law Theory in the Constitutional State’ in AP Simester, A du Bois Pedain and U Neumann, Liberal Criminal Theory (Oxford, Hart Publishing, 2018). 9 P Pettit, ‘Republican Theory and Criminal Punishment’ (1997) 9 Utilitas 59.

22  Why Criminal Law? Section I first explains the logical premises of this debate, namely that criminal law is a special field of law which can have a significant impact on individual liberty and dignity. Resort to criminal sanctions thus begs for justification. Section II illustrates the historical origins of the discussion. During the eighteenth century, the Enlightenment intellectual debates postulated the need to put limits to state power to preserve individuals’ liberty and dignity. This required a selective approach to criminal law and triggered a debate on when its use was legitimate. Sections III and IV go to the core of the debate, sketching two approaches to the legitimacy of criminal law – deontological and utilitarian – that have developed over the years and have traditionally been opposed to one another. Seciton V explains the advantages of jointly considering deontological and utilitarian rationales for legitimising criminalisation and advocates this joint approach. Section VI discusses the character (legal or not) of the obligation on national legislators to follow this joint approach to the legitimacy of criminal law. Section VII concludes. I.  WHY IS CRIMINAL LAW SPECIAL? A CONSEQUENCE-BASED APPROACH

In his famous 1995 article ‘The Definition of Crime’, Glanville Williams states: [W]e have rejected all definitions purporting to distinguish between crimes and other wrongs by reference to the sort of thing that is done or the sort of physical, economic or social consequences that follow from it. Only one possibility now remains. A crime must be defined by reference to the legal consequences of the act. We must distinguish, primarily, not between crimes and civil wrongs but between criminal and civil proceedings. A crime then becomes an act that is capable of being followed by criminal proceedings, having one of the types of outcome (punishment, and so forth) known to follow these proceedings.10

In this passage, Williams highlights the difficulty in identifying a special content of the criminal law and opts for a procedural-only understanding of crimes. The accuracy of this definition, more recently re-proposed by Lipman and Molan,11 can be debated. Due to the different influence of religion on state law, homosexuality is still a crime, in some cases subject to the death penalty in other parts of the world,12 whereas same-sex marriage is being progressively legalised in a large number of Western countries. Similarly, differences in corporate culture explain why the US only criminalised insider trading late in the twentieth

10 G Williams, ‘The Definition of Crime’ (1955) 8(1) Current Legal Problems 107, 123. 11 M Molan, Cases and Materials on Criminal Law (Abingdon, Routledge-Cavendish, 2009) 1; M Lippman, Contemporary Criminal Law: Concepts, Cases and Controversies (London, SAGE Publishing, 2013) 3. 12 2017 State-sponsored Homophobia Report, available at: https://www.ilga.org/downloads/2017/ ILGA_State_Sponsored_Homophobia_2017_WEB.pdf.

Why is Criminal Law Special? A Consequence-Based Approach  23 century when this practice was criminalised much earlier in other corporate cultures.13 Remarkably, a uniform definition of rape, an intuitively harmful behaviour, is lacking even in rather culturally homogeneous contexts like the EU.14 It appears that the countours of criminal law are to a large extent historically and geographically contingent, depending on the government in power.15 Still, a number of forms of harmful conduct which one can intuitively recognise as crimes, such as theft and assault, are traditionally found in the vast majority of criminal codes. Interestingly, Interpol has been functioning for almost a century with a scope of application defined with reference to ‘ordinary crimes’.16 This is arguably evidence of a certain international consensus on what is a crime. In fact, criminal law scholarship has oscillated over time between formal and content-based definitions of criminal law, in what Horder defines as waves of reformation and counter-reformation of criminal legal theory.17 Be that as it may, even without embracing Glanville Williams’ total rejection of a content-based understanding of criminal law, his definition is a precious one, in that it captures one, even if not the only, specific feature of criminal law: the consequences flowing from a transgression of a criminal norm are significantly different from those flowing from violations of administrative or civil norms. As the following sections will demonstrate, they have a stronger bearing on fundamental rights. This is why criminalisation choices beg justification. Glanville Williams speaks of the kind of proceedings and the kind of outcomes. The following discussion addresses these two aspects, but reverses the order by first discussing the impact of criminal sanctions on liberty (section I.A) and dignity (section I.B), and later the kinds of criminal proceedings (section I.C). This is because, as is better illustrated below, the type of proceedings logically descends from the kind of outcome. A.  The Impact on Individuals’ Liberty as a Defining Feature of Criminal Law Any provision of law which prescribes a certain behaviour might affect a specific fundamental right, for example, freedom of expression or in general the ‘freedom to act as one pleases’.18 Moreover, when a sanction like the payment of a 13 M Findlay, The Globalization of Crime: Understanding Transitional Relationships in Context (Cambridge, Cambridge University Press, 1999) 36. 14 On differences in definitions of crimes within Europe, including with regard to corruption and rape, see J-M Jehle and S Harrendorf (eds), Defining and Registering Criminal Offences and Measures (Göttingen, Universitätsverlag Göttingen, 2010). 15 A Ashworth, ‘Is the Criminal Law a Lost Cause?’ (2000) 116 LQR 225, 226. 16 Article 2 of the Constitution of the ICPO-INTERPOL adopted by the UN General Assembly at its 25th session (Vienna, 1956). 17 J Horder, ‘Bureaucratic “Criminal” Law: Too Much of a Bad Thing?’ (2014) LSE Working Papers 1/2014, 2–6. 18 As protected by art 2(1) the German Constitution. See du Bois Pedain (n 8) 305, 310.

24  Why Criminal Law? fine is imposed, the individual’s right to property is affected. This can hold true for criminal law, civil law and administrative law. Yet, with minor exceptions illustrated below, it is only criminal sanctions which can affect individuals’ right to liberty, where liberty is intended as freedom of movement. Offenders can be materially deprived of their own physical liberty as a result of criminal proceedings, first and foremost when imprisonment is imposed as a sanction following a criminal conviction.19 However, limitations of individuals’ liberty can also occur while criminal proceedings are still in progress and no conviction has yet been imposed. Suspects can be subject to preventive detention, which similarly limits their freedom of movement. Moreover, they can also be subject to a number of investigative measures which can be similarly prejudicial for one’s liberty, like body and home searches or coercive transfer before the judicial authorities for questioning.20 Admittedly, the relationship between imprisonment and criminal law is not necessarily a perfect equation. In some cases individuals can be subject to administrative detention without being convicted of a crime. These include immigration detention (for the purposes of expulsion or in the context of an asylum claim)21 and suspect terrorists’ administrative deprivation of liberty.22 Moreover, the English legal system includes a specific example of civil law detention in the case of contempt of court.23 Nonetheless, detention as a sanction attached to administrative offences remains more the exception than the rule, and material deprivation of one’s physical liberty is normally associated with criminal law proceedings. 19 For a comparative overview of the use of imprisonment in Europe, see H-J Albrecht, ‘Imprisonment and Alternatives to Prisons: Changes and Prospects in a Comparative Perspective’ (2006) 3(6) Revista Académica – Facultad de Derecho de la Universidad la Salle 27. 20 See extensively with reference to the Italian legal system V Manes, Il principio di offensività. Canone di politica criminale, criterio ermeneutico, parametro di ragionevolezza (Turin, Giappichelli, 2005) 142. For an overview of the criminal procedures in several national legal systems and the presence of similar investigative measures, see M Delmas-Marty and JR Spencer (eds), European Criminal Procedures (Cambridge, Cambridge University Press, 2005). 21 D Wilsher, ‘The Administrative Detention of Non-nationals Pursuant to Immigration Control: International and Constitutional Law Perspectives’ (2004) 53(4) International and Comparative Law Quarterly 897. 22 See, eg, MC Waxman, ‘Administrative Detention of Terrorists: Why Detain, and Detain Whom?’ (2009) 3(1) Journal of National Security Law and Policy 1. 23 This can occur when a civil injunction or other restriction order is adopted. Breach of this order amounts to civil contempt of court, for which a judge might make various orders, including imprisonment of up to two years. On this, see D Eady and ATH Smith, Arlidge, Eady and Smith on Contempt (London, Sweet & Maxwell, 2015) ch 3. Also, the anti-social behaviour order (ASBO) introduced in England by s 1 of the Crime and Disorder Act 1998 represented an expression of a hybrid approach between civil and criminal procedures, which could arguably be construed as introducing detention sanctions as a consequence of a breach of a civil law norm. An ASBO could restrict a person’s behaviour in specified ways. Breach of such order without any reasonable excuse was a criminal law offence with a maximum penalty of five years’ imprisonment. Technically here the imprisonment sanction was connected to a criminal offence; however, the punished behaviour in substance constituted a violation of a civil norm. On this, see A Ashworth and L Zedner, ‘Defending the Criminal Law: Reflection on the Changing Character of Crime, Procedure, and Sanctions’ (2008) 2 Criminal Law and Philosophy 21, 29–30.

Why is Criminal Law Special? A Consequence-Based Approach  25 It is worth mentioning that the idea that imprisonment actually restricts the right to liberty was challenged on a conceptual level. The Israeli Supreme Court was asked whether the right to liberty itself includes the right ‘to ignore the criminal norms defined by law as offences which entail punishment of ­imprisonment’,24 which would thus be restricted by imprisonment sanctions; or if the right to liberty should be interpreted as excluding the liberty to disregard these norms, and therefore imprisonment should not be considered as a restriction on liberty. Neither the Court nor the literature actually gave a definitive answer on the matter. The book thus rests on the traditional assumption that imprisonment does restrict individual liberty. B.  The Impact on Individuals’ Dignity as a Defining Feature of Criminal Law Criminal law can also affect individual liberty intended more broadly – not simply as material liberty of movement, but also as one’s moral liberty, or human dignity, as the Italian Constitutional Court refers to it.25 The same concept is referred to in German legal culture as the ‘right to develop one’s personality’.26 Panu Mikkinnen describes imprisonment as a polymorphous sanction whose consequences include the restriction of physical liberty, but also the experience, or the fear of experiencing, prison violence, either as a victim or as a witness.27 The psychological or material condition to which inmates are subject, together with potentially very poor detention conditions,28 can amount to a restriction on individuals’ dignity, even if the threshold for inhuman and degrading treatment is not reached.29 Moreover, the negative impact on the offender’s dignity stems from the stigmatising effect of any criminal law sanctions, including imprisonment (though not limited to it). Stigma in this context can be defined as someone’s reluctance

24 See M Gur-Arye and T Weigend, ‘Constitutional Review of Criminal Prohibitions Affecting Human Dignity and Liberty: German and Israeli Perspective’ (2011) 44 Israel Law Review 63, 75. 25 See the Italian Constitutional Court case law interpreting art 13 of the Constitution on the right to liberty, as discussed in M Ruotolo, ‘Gli itinerari della giurisprudenza costituzionale in tema di libertà personale’ (2004) 2/3 Questione Giustizia 235. 26 Article 2(1) the German Constitution read in conjunction with art 1(1). For a comment on this, see du Bois Pedain (n 8) 310. 27 P Mikkinen, ‘“If Taken in Earnest”: Criminal Law Doctrine and the Last Resort’ (2006) 45(5) Howard Journal 521, 522. 28 For an assessment of the use of pre-trial detention in EU Member States, see the country reports in the most recent Fair Trial report and see also the key findings which denounce the over-reliance on this instrument also for punitive purposes: ‘A Measure of Last Resort? The Practice of Pre-trial Detention in the EU’, 2016, https://www.fairtrials.org/wp-content/uploads/A-Measure-of-LastResort-Full-Version.pdf. 29 P Mikkinen, ‘The “Last Resort”: A Moral and/or Legal Principle?’ (2013) 3(1) Onati Socio-legal Series 21, 25.

26  Why Criminal Law? to interact with someone else who has a criminal record. It can have either an economic (eg, employers granting a lower wage) or a social (eg, ­difficulties in establishing a network of friends or finding a spouse) nature.30 Ethics ­literature,31 social psychology analysis32 and law and economics analysis33 all attest to the presence of stigmatising consequences stemming from any criminal sanction, such as custodial sanctions, pecuniary sanctions and any other sort of community sanctions.34 The theoretical rationale for the stigma lies in the punitive function of criminal law. According to a large number of criminal legal scholars, criminal sanctions are special because, contrary to civil or administrative sanctions, they are meant to punish offenders. In this respect, Feinberg speaks of a symbolic function of punishment35 and Ashworth of a censuring function of criminal law, which he connects to the social significance of criminal law.36 Von Hirsch and Simester argue that such a function is directly connected to the acknowledgement of offenders as individuals capable of moral agency. Criminal law conveys a message that the behaviour is wrong. Such wrongfulness should deter individuals from holding the given form of behaviour.37 This punitive aspect of criminal law also finds an echo in the Strasbourg case law.38 The stigmatising consequences are then considered not simply as a side-effect of the actual

30 E Rasmusen, ‘Stigma and Self-Fulfilling Expectations of Criminality’ (1996) 39 Journal of Law and Economics 519, 519. 31 D Kahan and M Nussbaum, ‘Two Conceptions of Emotion in Criminal Law’ (1996) 96 Columbia Law Review 296, 352. 32 See the observation of Gabrio Forti on how it is difficult at present to think about criminal sanctions separately from the stigma that these entail: G Forti, ‘Principio del danno e legittimazione ‘personalistica’ della tutela penale’, in G Fiandaca and G Francolini (eds), Sulla Legittimazione del Diritto Penale. Culture Europee-continentali e Anglo-Americane a confronto (Turin, Giapichelli, 2008) 43, 49; KE Moore, JB Stuewig and JP Tangney, ‘The Effect of Stigma on Criminal Offenders’ Functioning: A Longitudinal Mediational Model’ (2016) 37(2) Deviant Behavior 196. 33 On income reduction as a result of criminal convictions, see JR Lott, ‘Do We Punish High Income Criminals Too Heavily?’ (1992) 30 Economic Inquiry 583; and J Grogger, ‘The Effect of Arrests on the Employment and Earnings of Young Men’ (1995) 110(1) Quarterly Journal of Economics 51. On the impact of criminal sanctions on unemployment figures, see J Grogger, ‘Arrests, Persistent Youth Joblessness, and Black/White Employment Differentials’ (1992) 74(1) Review of Economics and Statistics 100. 34 The term ‘community sanctions’ refers to all those sanctions that do not imply deprivation of liberty and are to be served within the community, such as community service. On this, see G Robinson and F McNeil (eds), Community Punishment, European Perspectives (Abingdon, Routledge, 2016). 35 J Feinberg, ‘The Expressive Function of Punishment’ in Doing and Deserving (Princeton, Princeton University Press, 1970) 96. 36 A Ashworth, Principles of Criminal Law (Oxford, Oxford University Press, 2009), 1. 37 AP Simester and A von Hirsch, Crimes, Harms, and Wrongs: On the Principles of ­Criminalisation (Oxford, Oxford University Press, 2011) 12. 38 For a detailed outline of all the relevant cases for the definition of criminal charges, see the Council of Europe report, ‘Guide on Article 6. Right to a Fair Trial (Criminal Limb)’, www.echr. coe.int/Documents/Guide_Art_6_criminal_ENG.pdf. For a comparison between the definition of a crime in the works of Glanville Williams and that underpinning the case law of the Strasbourg Court, see Ashworth (n 15) 230.

Why is Criminal Law Special? A Consequence-Based Approach  27 sanction, but as the essence of criminal punishment itself.39 Next to this normative explanation of stigma, economic analysis provides insights on stigma as a rational consequence of associating criminality with other directly undesirable consequences. Reluctance to interact with convicted criminals can also be in certain contexts an efficiency-maximising choice.40 It is widely acknowledged that all such important attempts on suspects’ and convicted criminals’ dignity are only associated with the criminal law and do not feature as consequences of civil or administrative proceedings and related sanctions.41 It is telling that law and economics models include stigma as one of the variables in favour of the higher effectiveness of criminal law sanctions.42 Lastly, impact on individuals’ dignity can also derive from their mere involvement in criminal proceedings, even if the trial ends in an acquittal. As a consequence of an increasing mediatisation of criminal justice, the moment in time when the stigmatisation effect materialises no longer coincides with the conviction, but rather with the start of the investigation. Stigma is extended not only to the convicted criminal but also to the suspect criminal from the moment the proceedings start.43 C.  Higher Procedural Safeguards as a Non-defining Feature of Criminal Law One last implication of a transgression of a criminal norm is that the ensuing criminal proceedings imply more enhanced procedural protections for the defendant when compared to civil or administrative proceedings. For instance, convictions can only be ordered if guilt is proven beyond any reasonable doubt. This rule does not necessarily exist in civil proceedings.44 Moreover,

39 See Feinberg (n 32) 118, who recalls that the essence of the punishment is social disapproval and, as a consequence, it is its appropriate expression that should fit the crime, and not hard treatment as such. 40 For a general model on the subject, see Rasmusen (n 30). On the effectiveness of stigma as a deterrent, see K Svatikova, ‘Economic Criteria for Criminalisation: Why Do We Need the Criminal Law?’ (2008) Rotterdam Institute of Law and Economics Working Paper Series 2008/12, 18 ff and the references therein. 41 For a critical view discussing stigma and administrative law, see Roberto Galbiati and Nuno Garoupa, ‘Keeping Stigma out of Administrative Law: An Explanation of Consistent Beliefs’ (2007) 15 Supreme Court Economic Review 273. 42 See Svatikova (n 40); and R Bowles, M Faure and N Garoupa, ‘The Scope of Criminal Law and Criminal Sanctions: An Economic View and Policy Implications’ (2008) 35(3) Journal of Law and Society 389, 406. 43 On the stigmatising impact of mediatised criminal justice trials, see the observations in Manes (n 20) 141; and S Moccia, La Perenne emergenza. Tendenze autoritarie nel sistema penale, 2nd edn (Rome, Edizioni Scientifiche Italiane, 2000) 159. On justice and media in general, see M Lemonde, ‘Justice and the Media’ in Delmas-Marty and Spencer (n 20) 688. 44 See art 533 para 1 of the Italian Criminal Code; for a similar formulation, see the decision of the Spanish Constitutional Tribunal STC 43/2003, de 3 de Marzo, STC 246/2004, de 20 de

28  Why Criminal Law? the defendant in criminal proceedings benefits from a more enhanced right to defence than he would in civil proceedings.45 While this is undoubtedly a defining characteristic of criminal proceedings, it should be noted that their rationale is arguably not an independent one; it can be traced back to the possible heavy outcome of criminal trials. Given that in the context of criminal trials there is a lot at stake, including potential deprivation of liberty and an impact on individuals’ dignity, the position of the defendant is particularly safeguarded.46 This connection between outcome and proceedings is evident in those arguments in national47 and European48 criminal law scholarship advocating the use of criminal law because of the attached special safeguards. It is argued that particularly severe administrative or civil sanctions represent a ‘mislabelling of reality’, to the extent that they are comparable to criminal sanctions, but are imposed using a procedure which does not respect the relevant procedural safeguards. II.  THE NEED TO LEGITIMATE CRIMINALISATION CHOICES

The previous sections have illustrated the heavy consequences flowing from the violation of a criminal norm in relation to individuals’ liberty and dignity. It follows that if a legal order keeps liberty and dignity in consideration, then opting for qualifying a behaviour as a crime is a choice which must be thoroughly justified. As Duff et al put it, if one agrees that criminal law is a special branch of law, then one has to specifically ask ‘why criminal law?’ and not just generally: ‘Why state regulation at all?’49 Baker and Husak even respectively argue that specific rights not to be criminalised50 and not to be punished exist.51 Diciembre, and STC 209/2002 de 11 de Noviembre. Compare also the formulation in art 338 of the Dutch Criminal Code, which states that guilt must be proven ‘wetting en overtuigend’ (lawfully and convincingly). For the rule in the English and Welsh criminal law system, see Woolmington v DPP [1935] UKHL 1, [1935] AC 462. 45 For a comparison between the right to the defence in the administrative proceedings and in criminal proceedings in the Italian system, see Manes (n 20) 145. For a similar comparison with the Spanish system, see S Calaza Lopez, ‘Principio Rectores del Proceso Judicial Espagnol’ (2011) 8 Revista de Derecho UNED 49, 53–56. For the presence of higher guarantees in the criminal law proceedings in English and Welsh law, see Ashworth (n 15) 238–40. 46 On the connection between the outcome of the proceedings and the protection awarded to the defendant, and how this also explains the Strasbourg insistence on policing the boundaries between criminal and civil law, see Ashworth (n 15) 238–40. 47 A Ashworth and L Zedner, ‘Preventative Orders: A Problem of Undercriminalisation?’ in RA Duff et al (eds), The Boundaries of the Criminal Law (Oxford, Oxford University Press, 2010) 59. 48 E Herlin-Karnell, ‘Is Administrative Law Still Relevant? How the Battle of Sanctions Has Shaped EU Criminal Law’ in M Bergstroom, V Mitsilegas and T Konstatinides (eds), Research Handbook on EU Criminal Law (Cheltenham, Edward Elgar, 2015) 233, especially s 5. 49 Duff et al (n 47) 5. 50 D Baker, The Right Not to Be Crimininalized: Demarcating Criminal Law’s Authority (Farnham, Ashgate, 2011). 51 D Husak, Overcriminalisation (Oxford, Oxford University Press, 2008) 92–103.

The Need to Legitimate Criminalisation Choices   29 According to these doctrinal constructions, it is thus the infringement of these rights which needs to be justified. Historically speaking, the question of safeguarding individuals’ liberty and dignity, and the connected debate on the justification for the use of criminal law first arose in the eighteenth century. During the Enlightenment, the cultural and philosophical debate put the individual and his liberty at the centre of the discussion.52 The political theory of liberalism, which developed on the basis of these premises, acknowledges the individual and his liberty, as opposed to religious morality, as the ‘primary unit of ethical concern’.53 Liberal t­hinkers accordingly envisage precise limits to state authority in order to preserve individuals’ autonomy. In terms of criminal law reform, the raise of liberalism meant, among other things,54 conceptualising secular basis for justifying criminalisation. The aim was to limit the abuse of power by the state that had characterised medieval, pre-modern criminal law where the concept of crime and sin basically coincided.55 A purely moralised and religious notion of crime as wrong or even evil (mala in se) was progressively substituted by a legalised notion of crime as infraction (mala prohibita). Criminal law saw its link with common moral sense attenuated, and it also increasingly became a legal or technical question.56 This paradigm shift necessarily triggered the question of the permissible content of criminal law;57 in other words, as soon as criminal law could not always benefit from spontaneous social legitimation, the need for a more sophisticated approach to its legitimacy materialised. In the nineteenth and twentieth centuries, a florid scholarship, traditionally referred to as the scholarship on the legitimacy of criminal law,58 developed on what the foundations of this new modern secular criminal law should be. Two broad trends can be identified in criminal legal scholarship, which reflect two general appproaches to the limits of state power. Roberts defines them as ‘deontological liberalism’ and ‘utilitarian liberalism’.59 As the next two sections will illustrate, in both cases a selective approach to criminal law is postulated, 52 Dorinda Outram, The Enlightenment (Cambridge, Cambridge University Press, 2013). 53 P Roberts, ‘Criminal Law Theory and the Limits of Liberalism’ in Simester et al (n 8) 326, 331. 54 For an overview of criminal law reform during the Enlightenment, see G Vassalli, La potestà punitiva statuale (Turin, UTET, 1942) 60–80. 55 For an account of ‘pre-modern’ criminal law and for the rationales behind penal enlightenment, see G Fiandaca and Ezio Musco, Diritto Penale – Parte Generale (Bologna, Zanichelli, 2014) ixv–xvii. See, however, on the authoritarian traits of modern criminal law, see M Vogliotti, Tra Fatto e Diritto. Oltre la modernità giuridica (Turin, Giappichelli, 2008). 56 Lacey observes that the link between legal and social norms was tight within Blackstone’s eighteenth-century definition of crime as a public wrong and that this connection got progressively lost: Nicola Lacey, ‘What Constitutes Criminal Law?’ in RA Duff et al, The Constitution of the Criminal Law (Oxford, Oxford University Press, 2013) 13, 21. 57 ibid 12, 14. 58 N Persak, The Harm Principle and its Continental Counterpart (New York, Springer, 2012) 4, 127. 59 Roberts (n 53) 326, 331.

30  Why Criminal Law? grounded only in secular rationales for criminal law and rejecting religious ones. Yet for deontological liberalism, which can be considered liberalism strictu sensu, what matters are the (secular) interests that the criminalised behaviour affects, whereas in utilitarian liberalism, the focus is on the (secular) objectives of criminalisation and, in particular, the utility and effectiveness of the state’s control and regulatory functions. These two approaches did not have the same success as paradigm for criminalisation. A deontological approach to c­ riminalisation has progressively become the ‘orthodox’ view among contemporary criminal law theorists,60 whereas the utilitarian approach enjoys less doctrinal support. However, as section V suggests, a truly conservative, and thus more legitimate, approach to criminal law would require jointly considering both utilitarian and normative rationales for criminalisation. III.  A (LIBERAL) DEONTOLOGICAL APPROACH TO THE LEGITIMACY OF CRIMINAL LAW

Mainly championed by Kant (but also Rawls), deontological liberalism as a political philosophy requires considering individuals as ends in themselves rather than as a means to someone else’s end. The liberty of individuals is thus the ultimate value of a political system.61 It follows that, as epitomised in Mill’s essay On Liberty, the state acts legitimately in restricting individuals’ autonomy if this is necessary to prevent harm to others, that is, to safeguard someone else’s liberty.62 In criminalisation theory, deontological liberalism accordingly implies only penalising those behaviours which directly and indirectly threaten the value of liberty63 or, according to a specific Italian doctrinal approach, other values that are as important as liberty.64 Each decision to criminalise requires an evaluation of the inherent value of the interests that the relevant behaviour affects. If such interests are necessary to safeguard everyone’s liberty or are as important as liberty, then qualifying that particular behaviour as a crime is legitimate. If they are not, then more lenient sanctions such as administrative or civil sanctions are sufficient. Briefly, what legitimises resorting to criminal law is thus the importance, and the connection with liberty, of the pre-existing values it seeks to protect. In the Anglo-Saxon tradition, the most influential expression of a deontological approach to criminalisation has been Joel Feinberg’s harm principle theory, which directly draws on Millian’s political theory. Feinberg’s core argument



60 ibid

325. 331. 62 JS Mill, On Liberty and Other Essays (Oxford, Oxford University Press, 1991 [1859]). 63 Roberts (n 53) 337. 64 See below in this section. 61 ibid

A (Liberal) Deontological Approach to the Legitimacy of Criminal Law  31 is that criminalisation is legitimate when it targets conducts that cause harm to others,65 where harm is defined as a ‘wrongful setback of interests’.66 This harm-based perspective, also referred to as instrumental perspective (criminal law is an instrument to prevention of harm),67 has often been opposed to legal moralism, one which Feinberg acknowledges only as an exception to the main harm criterion.68 Legal moralism, authoritiatively supported by Stephen,69 Lord Devlin at a later stage70 and more recently Duff,71 legitimises criminalising behaviours, although with some nuances, which do not cause harm to any interest, but offend morality. However, in his seminal article on ‘The Collapse of the Harm Principle’, Harcourt explains how moralistic arguments can also be rationalised under the umbrella of the harm principle.72 It all depends on how one defines the interests which must be free from harm and whether morality should be one of these. Harm-based arguments have been used, for instance, in the US to justify the criminalisation of homosexuality or prostiution.73 The opposition between liberalism and moralism is thus more apparent than real. For the purposes of our discussion here, legal moralism can be considered an expression of a deontological approach in the same way as the harm-based approach is, in that it focuses on the interests that the relevant behaviour affects. In Continental Europe, the main point of reference for a deontological approach to criminalisation has been the Rechtsgut Theory, which legitimises the use of criminal law only to protect Rechtsguts. This latter concept, which translates as legal goods or interests,74 was introduced in the nineteenth century

65 J Feinberg, Harm to Others: The Moral Limits of the Criminal Law (Oxford, Oxford University Press, 1984). 66 ibid 36. 67 See ‘Theories of Criminal Law’, Stanford Encyclopedia of Philosophy, https://plato.stanford. edu/entries/criminal-law/#InsMorConCriLaw. 68 Next to the harm principle, Feiberg admits as exceptional criteria: offence (criminalising behaviour that causes offence to others), legal moralism (criminalising behaviour that offends morality) and legal paternalism (criminalising behaviour that harms oneself). See J Feinberg, Offence to Others: The Moral Limits of the Criminal Law (Oxford, Oxford University Press, 1985); J Feinberg, Harm to Self: The Moral Limits of the Criminal Law (Oxford, Oxford University Press, 1986); J Feinberg, Harmless Wrongdoing: The Moral Limits of the Criminal Law (Oxford, Oxford ­University Press, 1988). 69 JF Stephen, Liberty, Equality, Fraternity, RJ White (ed) (Cambridge, Cambridge University Press, 1967 [1873]). 70 P Devlin, ‘Morals and the Criminal Law’ in The Legal Enforcement of Morality (Oxford, Oxford University Press, 1965). 71 RA Duff, ‘Towards a Modest Legal Moralism’ (2012) 8(1) Criminal Law and Philosophy 217. 72 BE Harcourt, ‘The Collapse of the Harm Principle’ (1999) 1 Journal of Criminal Law and Criminology 109. 73 ibid 145–72. 74 The German term Rechtsgut does not have a specific corresponding term in English. Lauterwein translates it as ‘legal goods/interest’; C Lauterwein, The Limits of Criminal Law: A Comparative Analysis of Approaches to Legal Theorizing (Farnham, Ashgate, 2010) 5. Other authors prefer ‘legal good’ rather than ‘legal interest’; see C-F Stuckenberg, ‘The Constitutional Deficiencies of the German Rechtsgutslehre’ (2013) 3(1) Onati Socio-legal Series 31, 35.

32  Why Criminal Law? by Birnbaum.75 A modern exhaustive and systematic discussion of the theory can then be found in Claus Roxin’s works of 1966.76 Roxin defines Rechtsguts as those interests/goods whose protection is necessary to ensure peaceful and liberal co-existence,77 where everyone’s liberty can be guaranteed. Rechtsguts cannot be based on ideology or morality and a Rechtsgut cannot consist only of a vague objective of regulation, such as the creation of a drug-free society.78 Rechtsgut Theory first developed in Germany and then spread across Europe, and was endorsed and re-elaborated by scholarship in several other countries.79 Italian scholars in particular (especically Franco Bricola)80 provided a precious contribution to the theor, by linking the criteria for the selection of legal goods to the norms in the Constitution. Bricola’s teoria del bene giuridico costituzionalmente orientato (theory of the constitutionally oriented legal interest) is grounded in the understanding that criminal law affects an interest protected by the Constitution, namely personal liberty (Article 13 of the Italian Constitution). As a consequence, he argues, the sacrificing of individuals’ liberty by means of the criminal law can only occur when interests of the same constitutional rank are at stake. Criminal law can only be used to protect those interests directly or indirectly safeguarded by the Constitution. On the basis of these authoritative premises, the liberal deontological debate on criminalisation has largely developed and continues to evolve. Giving a exhaustive account of all the trends within modern criminal legal theory would be beyond the scope of this chapter. However, it is worth recalling how scholars keep discussing and updating the concepts of harm (eg, is there only one or several harm principles?),81 of legal interest (eg, should freedom from humiliation be a relevant legal interest?)82 and the space for morality83 and paternalism84

75 JMF Birnbaum, ‘Uber das Erordernis einer Rechtsverletzung zum Begriff des Verbrechens’ (1834) (15) Archiv des Criminalrechts 149. 76 Roxin first presented this theory in C Roxin, ‘Sinn und Grenzen staatlicher Strafe’ (1966) Juristische Schulung 377, 381–82; a more up-to-date version can be found in C Roxin, Allgemeiner Teil: Grundlagen, Aufbau der Verbrechensehre, vol 1, 4th edn (Munich, CH Beck, 2006). 77 Roxin, Allgemeiner Teil (n 76) 16–17. 78 ibid. 79 German criminal law theory had widespread influence in the civil law world. Its impact arguably extends from European countries like Spain, Portugal, Croatia, Greece, Turkey, Austria, ­Switzerland, Finland and Sweden to Israel and the Spanish and Portuguese-speaking countries of Latin America, as well as to Asian countries such as Taiwan, South Korea and Japan. See H-J Hirsch (ed), Krise des Strafrechts und der Kriminalwissenschaften?Tagungsbeiträge eines Symposiums der Alexander von Humboldt-Stiftung, Bonn-Bad Godesberg, veranstaltet vom 1. bis 5. Oktober 2000 in Bamberg (Berlin, Duncker & Humblot, 2001), quoted in Markus D. Dubber, ‘Theories of Crime and Punishment in German Criminal Law’ (2005) Buffalo Legal Studies Research Paper 2005/02 1. 80 F Bricola, ‘Teoria Generale del Reato’ (1973) XIX Novissimo Digesto Italiano 3. 81 RA Duff and SE Marshall, ‘Remote Harms and the Two Harm Principles’ in Simester et al (n 8) 205. 82 T Hornle, ‘Criminalising Behaviour to Protect Human Dignity’ (2012) 6 Criminal Law and Philosophy 307. 83 D Scoccia, ‘In Defence of “Pure” Legal Moralism’ (2013) 7(3) Criminal Law and Philosophy 513. 84 V Tadors, ‘Harm Sovereignty and Prohibition’ (2011) 17 Legal Theory 35.

A (Liberal) Utilitarian Approach to the Legitimacy of Criminal Law  33 in liberal criminal theory. They consider how criminalisation theories should adapt to new challenges, including, for instance, the rise of cosmopolitan ­societies,85 social injustice86 and the advent of technologies,87 adding various layers of complexity to the debate. IV.  A (LIBERAL) UTILITARIAN APPROACH TO THE LEGITIMACY OF CRIMINAL LAW

Utilitarian liberalism, which was first introduced as a political philosophy by Jeremy Bentham in the eighteenth century, is grounded in the axiom that the measure of right or wrong lies in the greatest happiness of the greatest number.88 The legitimacy of institutions and therefore of law itself thus lies in their capacity of maximising collective utility.89 As Roberts clarifies it, despite different interpretations, Bentham and Kant are within the big liberal tent, just as much as Plato, Aquinas, Marx or the Ayatollah Khommeini are not, in that they embrace a secular political morality.90 Yet, contrary to liberals strictu sensu, they admit that in some cases, public authority can restrict the liberty of individuals to pursue the general goal of collective utility. The latter, and not individuals’ liberty, is thus the ultimate unit of concern of a political system. Individuals can, as a matter of principle, be treated instrumentally.91 In criminalisation theory, utilitarian liberalism implies justifying criminalisation on the basis of its contribution to collective utility. Interpreted more broadly, in this book, all those criminalisation agendas which justify resort to the criminal law on the basis of its suitability to any future objectives it can reach regardless of the pre-existing interests or values that the behaviour at stake affects, are considered to be utilitarian. Next to general welfare, or utility, as in Bentham’s original version, the relevant objective might be just effective law enforcement or any social engineering goals92 to which criminal law can contribute. A use of criminal law simply to spread a sense of security among citizens 85 O Hoffe, ‘Moral Reasons for an Intercultural Criminal Law: A Philosophical Attempt’ (1998) 11(3) Ratio Juris 206. 86 M Torny, ‘Can Deserts Be Just in an Unjust World?’ in Simester et al (n 8) 142. See also the debates on considering social harm as an alternative to crime in P Hillyard and S Tombs ‘From “Crime” to Social Harm?’ (2007) 48 Law and Social Change 9. 87 GS Moohr, ‘The Crime of Copyright Infringement: An Inquiry Based on Morality, Harm, and Criminal Theory’, www.law.uh.edu/faculty/gmoohr/Criminal.pdf. 88 J Bentham, A Fragment on Government, JH Burns and HLA Hart (eds) (Cambridge, Cambridge University Press, 1988 [1776]) Preface, para 2. 89 ibid. 90 See Roberts (n 53) 330. 91 ibid 331. 92 The expression ‘social engineering’ in this context is loosely understood to refer to those efforts to influence particular attitudes and social behaviours on a large scale, whether by governments, media or private groups, in order to produce desired characteristics in a target population. This aspect is further discussed in ch 3.

34  Why Criminal Law? would qualify as a utilitarian use of criminal law. Each decision to criminalise would require an evaluation of the suitability of criminal law with regard to the achievement of the desired objective. If, due to the specificities of the given policy area, the typologies of the offenders or any other reasons, it is deemed that criminal sanctions will not efficiently or effectively secure the relevant goal, then different regulatory tools should be considered, such as administrative or civil sanctions. Utilitarianism has influenced several aspects of criminal law theory. However, it has not received much support as a criminalisation theory. Bentham mainly discussed the question of ‘why punish?’ rather than ‘why criminalise?’ He has arrived in some cases to far-fetched conclusions, like requiring the punishment of individuals for the offences they have committed and for those they are likely to have committed, in order to maximise collective utility.93 In a similar vein, roughly a century later, von Listz proposed a utilitarian understanding of punishment in German scholarly circles.94 More modern utilitarian thinkers, such as Hart95 and Paker,96 have developed this orginal utilitarian approach introducing significant nuances to Benthamite absolute claims. The utilitarian debate has been particularly lively on the question of criminal law defences.97 A ‘updated’ variant of eighteenth-century utilitarianism that addressed the aspect of criminalisation more closely is the subject of law and economics.98 This stream of literature applies economic arguments to public policy questions on whether to regulate citizens’ behaviour and how. Authors like Becker,99 Posner100 and, more recently, Ayres and Braithwhaite,101 Bowels et al102 and Svatikova103 have developed economic models to identify when criminal law is the most efficient choice to secure law enforcement. This literature can be considered

93 J Bentham, Introduction to the Principles of Morals and Legislation, JH Burns and HLA Hart (eds) (York, Methuen, 1982 [1789]) 170. 94 See the observations in G Fletcher, ‘Criminal Theory in the Twentieth Century’ (2001) 2 ­Theoretical Inquiries in Law 267. 95 HLA Hart, Punishment and Responsibility: Essays in the Philosophy of Law (Oxford, ­Clarendon Press, 1968). 96 H Packer, The Limits of the Criminal Sanctions (Stanford, Stanford University Press, 1968). 97 See D Husak, ‘Justification and the Criminal Liability of Accessories’ (1988) 80 Journal of Criminal Law and Criminology 201. 98 The connection between utilitarianism and law and economics is acknowledged by e­ xponents of the latter movement themselves; see GS Becker, ‘Crime and Punishment: An Economic Approach’ in GS Becker and WM Landes (eds), Essays in the Economics of Crime and Punishment (New York, National Bureau of Economic Research, Columbia University Press, 1974) 1; see also N Lacey, ‘Historicising Criminalisation: Conceptual and Empirical Issues’ (2009) 72(6) Modern Law Review 936, 937–38. 99 Becker (n 98). 100 R Posner, ‘An Economic Theory of the Criminal Law’ (1985) 85 Columbia Law Review 1193. 101 I Ayres and J Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (Oxford, Oxford University Press, 1992). 102 R Bowels, M Faure and N Garoupa, ‘The Scope of Criminal Law and Criminal Sanctions: An Economic View and Policy Implications’ (2008) 35(3) Journal of Law and Society 389. 103 Svatikova (n 40).

A Deontological-Utilitarian Approach to Criminal Law  35 an expression of a utilitarian approach to criminalisation to the extent that it looks at the ‘suitability of criminal law to reach a certain future objective (i.e. efficient enforcement)’ as a factor legitimating resort to penal law. Welcomed as particularly innovative and insightful by some,104 the economic approach to criminal law nonetheless has several detractors. The focus on efficiency is said to neglect aspects such as coherence and fair treatment within the criminal justice system, as well as the expressive function of criminal law.105 However, it should be noted that law and economics authors do not necessarily take a stance on the function of criminal law. They provide efficiency-based tools to compare enforcement strategies, which legislators might rely upon should they wish to include effective enforcement as one of the criteria in public policy. They do not postulate efficiency as the only criteria which should play a role in the decision as to whether to resort to criminal law. In this sense, this literature cannot be considered to provide normative support to a purely utilitarian understanding of criminalisation. And actually, as the next section will explain, there is an interest in combining the two models. V.  A JOINT DEONTOLOGICAL-UTILITARIAN APPROACH TO THE LEGITIMACY OF CRIMINAL LAW?

Deontological liberalism and utilitarian liberalism, as described in the previous sections, are certainly conceptually distinct models for criminal law, respectively privileging values-based and utility/effective enforcement dimensions. The question is whether reconciling them in practice is possible and desirable. A.  Opposition and Reconciliation of Deontological and Utilitarian Approaches The combination of the deontological and utilitarian dimensions is frequent in debates on criminal punishment. Hart played a key role in mediating between the Benthamite focus on deterrence and Kantian non-consequentialist concerns in the fair imposition of punishment.106 Conversely, there is less consensus in terms of if and how to combine the two perspectives during the criminalisation

104 See, inter alia, praise of Becker’s model in D Teichmann, ‘The Economics of Crime Control’ in PH Robinson, Stephen Garvey and Kimberly Kessler Ferzan (eds), Criminal Law Conversations (Oxford, Oxford University Press, 2015) 87; and J Muraszkiewicz and P De Hert (2014) ‘Gary Becker and the Economics of Trafficking in Human Beings’ (2014) 5(2) New Journal of European Criminal Law 116. 105 See the criticism advanced by Yeung to Ayres and Braithwaite’s model in K Yeung, Securing Compliance (Oxford, Hart Publishing 2004); and Ashworth (n 15) 249–51. 106 See the observations in Fletcher (n 94) 265.

36  Why Criminal Law? phase. This is well illustrated by the debate on regulatory criminal law, namely criminal law adopted only for enforcement purposes, regardless of the involved values.107 Despite benefiting from weaker doctrinal support, a utilitarian, enforcement-based approach to criminalisation has significantly spread in practice. Especially in the last century, criminal law has been increasingly used to reinforce provisions of law adopted outside the penal sphere, which do not necessarily protect important interests and thus are not backed up by a corresponding social norm.108 These include prescriptions defined by administrative law or civil law, or even non legislative measures issued by administrative agencies or judicial authorities.109 In this context, the resort to criminal law is justified on the basis of its presumed effectiveness in securing law enforcement rather than on the interests at stake. The social and political context in which such bureaucratisation of criminal law takes place is that of a regulatory state.110 The modern liberal state aimed at preserving non-interference with citizens’ pre-existing liberties. Conversely, this post-modern regulatory state postulates a stronger interventionist state, relying on further interference with citizens’ liberties and using criminal law to safeguard its managing and control functions.111 Several leading criminal law scholars, who are champions of the deontological ‘orthodoxy’ as mentioned above, have been very critical of these developments both in Europe and in the US. Ashworth,112 Husak113 Delmas Marty114 Sanchez115 and Baratta,116 to name but a few, have denounced the unprecedented spread of regulatory criminal offences respectively in England, the US, France, Spain and Italy. Baratta speaks of a blunt betrayal of the original spirit of the Rechtsgut Theory.117 With different nuances, these authors

107 A Ashworth, ‘Conceptions of Overcriminalisation’ (2007–08) 5 Ohio State Journal of Criminal Law 407, 408. 108 Ashworth contrasts this use of criminal law with the traditional declaratory use, namely with the use of criminal law to sanction violations of pre-existing social norms: ibid. 109 Some representative, yet obviously not exhaustive, examples can be found in the Italian regulation of the managing of cultural heritage, where carrying out specific activities without the required authorisation might constitute a crime (Manes (n 20) 92), or the Italian data protection discipline, where in certain cases the violation of measures enacted by the Data Protection Authority can be sanctioned by the criminal law (Manes (n 20) 110) or finally in the English legislation on disposing of special waste outside the condition of licence, on making a misleading statement in relation to investment business and on the breach of a civil prevention order, which are punished with imprisonment sanctions (Ashworth (n 15) 243). 110 The term ‘regulatory state’ is taken from Ashworth and Zedner (n 23) 38. 111 Manes (n 20) 96. 112 Ashworth (n 15). 113 Husak (n 51). 114 M Delmas-Marty, Le Flou du droit, 2nd edn (Paris, PUF, 2004). 115 JM Silva Sanchez, La expansion del derecho pénal. Aspectos de la politica criminal en las ­sociedades post-industriales (Madrid, Civitas, 2001). 116 A Baratta, ‘Les fonctions instrumentales et les fonctions symboliques du droit pénal’ (1991) 15(1) Deviance et société 1. 117 ibid 8.

A Deontological-Utilitarian Approach to Criminal Law  37 advocated a return to a values-based deontological approach to criminal law.118 Horder eloquentely calls this literature a ‘back to basics’ campaign.119 As will be further elaborated upon in the next chapter, a similar call was made for a more deontological approach to criminal law at the EU level.120 The argument in favour of such an approach stems from a normative preference for valuesbased considerations in criminal law and from a supposed higher capacity of the deontological approach to limit the expansion of the criminal law. For instance, Ashworth begins his discussion on whether ‘the criminal law is a lost cause’121 by denouncing the fact that English criminal law has reached a peak of over 8,000 offences in the last 150 years due to utilitarian considerations creeping into criminalisation agendas. Interestingly, some authors start from the same premise – too much criminal law – but end up with an opposite conclusion: abandoning substantive criminal legal theory criteria tout court and adopting different procedural or institutional arrangements to contain the over-expansion of the criminal law. Brown suggests entrusting a technical body, like the American Law Institute, as opposed to a political body, like an elected parliament, with the drafting of criminal law.122 Dripps suggest instead introducing a super-majority requirement for the adoption of criminal norms.123 Conversely, other authors have been more open to a joint utilitarian and deontological approach to the legitimacy of criminalisation. Horder, for ­ instance, has a positive outlook on regulatory criminal law. Drawing examples from the regulation of animal welfare in the UK, he considers that a greater number of precisely defined regulatory offences, as opposed to fewer broadly defined values-based offences, better serves the objective of legal certainty. Moreover, it does not overburden the judges with the task of adequately developing the meaning of the broad offence in specific contexts.124 Horder also advocates recognising the legitimacy of both cultures of criminal law – the traditional values-based culture and the effectiveness-based culture – which underpins regulatory criminal law. However, he does not necessarily advocate an integrated approach as such; rather, he distinguishes two situations in which either approach should apply: values-based criminal law should apply when offenders are natural persons; and effectiveness-based criminal law should apply when offenders are legal persons.125 118 Ashworth, for instance, recognises that modern society, however reluctantly, should include some considerations of speed and efficiency in criminalisation decisions; A Ashworth, ‘Towards a Theory of Criminal Law’ (1989) 1 Criminal Law Forum 41, 59. 119 Horder (n 17) 7. 120 European Criminal Policy Initiative, ‘The Manifesto on European Criminal Policy’ (2011) 1(1) European Criminal Law Review 86. 121 Ashworth (n 15). 122 D Brown, ‘History’s Challenge to Criminal Legal Theory’ (2009) 3 Criminal Law and Philosophy 271. 123 D Dripps, ‘The Liberal Critique of the Harm Principle’ (1998) 17(3) Criminal Justice Ethics 3. 124 Horder (n 17) 11. 125 ibid 33.

38  Why Criminal Law? Furthermore, as mentioned earlier, law and economics authors have brought the effectiveness of criminal law into the modern debate on criminalisation. Their theorisation has arguably influenced the spread of regulatory criminal law. Yet they do not negate the normative importance of values-based considerations. Becker makes it explicit that he is not concerned with why the behaviour at stake is illegal. He assumes a consensus on the rationale for the prohibition and he focuses on working out rules for an optimal implementation of this consensus, and, in particular, under which conditions criminal law would be most efficient in relation to this aim.126 Bowels et al similarly make this distinction, clearly stating that decisions about the activities to be prevented or deterred are separate from the choice of legal instrument for control, that criminal law is a matter that depends on the historical and cultural context, and that their discussion does not contain normative statements as to the use of the criminal law, but rather a positive analysis of why, on the basis of economic arguments, criminal law has emerged as a preferred regulatory instrument.127 Finally, Svatikova declares that her first assumption is that efficiency is accepted as an end goal, but she makes no normative conclusions as to what should be the goal of criminal law per se.128 Admittedly, Richard Posner has concerned himself with the aims of the criminal law and has investigated when the efficiency rationales guide definitions of crimes.129 However, his attempt is more descriptive-analytical – he tries to explain existing legal constructions on efficiency grounds rather than normative grounds, and he does not postulate efficiency as the sole criterion for the ­criminalisation decisions. On a more principled level, von Hirsch and Simester provide an authoritative synthesis of the importance of both the utilitarian and deontological dimensions to criminal law. They distinguish between a neutral conception of criminal law, as simply an effective enforcement instrument, and a moral conception of criminal law, which is used to punish and censure wrongdoers: The truth is, we think, somewhere in between. The criminal law is a regulatory tool for influencing behaviour, and in some respects no more than that; but it is a special kind of tool …130 The most plausible direction of analysis is towards a principled account which accommodates deontological and consequential values … In discussing the appropriateness of criminal proscriptions, therefore, we need to be mindful of the interaction of these concerns. Institutions such as the criminal law call for moral justification; but those arguments need to address the law as a social institution,



126 Becker

(n 98) 45. et al (n 102) 392. 128 Svatikova (n 40). 129 Posner (n 100). 130 Simester and von Hirsch (n 37) 4. 127 Bowels

A Deontological-Utilitarian Approach to Criminal Law  39 involving the exercise of State power, that must have pragmatic goals concerning the welfare of citizens. The state is an instrumental actor, and not a moralising institution per se.131

Admittedly, Simester and von Hirsch exclusively speak of a moral approach to the criminal law. However, for the purposes of the discussion in this book, as noted above, legal moralism can be treated as just one expression of a deontological approach to criminalisation, like the harm principle or the legal interests principles. Von Hirsch and Simester’s conclusions can thus be generalised to broadly advocating a joint deontological and utilitarian approach to criminalisation. B.  The Empirical and Normative Advantages of a Joint Approach The more conciliatory approach, which suggests complementing a deontological approach with a utilitarian one, and especially with effectiveness consideration has at least two advantages. First, it acnowledges the empirical reality of the current regulatory state, which has an actual need to ensure the effective implementation of its managerial and administrative functions. Admittedly, a normative criminal legal theory is not meant to positively describe or incorporate trends stemming from legislative practice. This is, for instance, the case for the theories on the legitimacy of criminal law which are meant to propose ideal models of criminal law. If there is a gap between the two, as occurs between the theory purported by the supporters of the deontological ortodoxy and the practice of regulatory criminal law, one can always attribute such a gap to the radical imperfection of the practice.132 However, ‘normative theorists must be alert to the possibility that the gap might become so dramatic that it is no longer clear whether the theory can be seen as a theory of that practice’.133 Incorporating utilitarian and especially effectiveness consideration into criminalisation policy, which do not substitute but complement deontological considerations, can lead to a more realistic approach to criminal law, which is more in line with the current regulatory understanding of state functions. Second, from a normative point of view, a joint deontological and utilitarian approach to the legitimacy of criminal law can lead, at least as a matter of principle, to a more restrictive approach to criminal law. If it is agreed that the act of criminalisation implies a balancing exercise between the individual offender’s liberty and dignity on the one hand, and the rationale for the criminal norm on the other hand, then naturally the more restricted the number of such rationales, the greater the importance granted to the individual’s liberty



131 ibid

18.

132 Stanford 133 ibid.

Encyclopedia of Philosophy (n 67) para 9.

40  Why Criminal Law? and dignity. A deontological approach to criminal law represents a qualitative boundary to the expansion of criminal law, to the extent that it limits it only to worthy aims. The need to also incorporate utilitarian considerations to justify criminalisation choices as a necessary but not sufficient condition would put a further quantitative restriction on the expansion of criminal law. Economics and utilitarian arguments can also militate against criminalisation, even where the relevant behaviour would deserve criminalisation from a values-based perspective. For instance, as will be shown in Chapter 7, the effectiveness (or lack of thereof) of criminal law was relied upon as an argument against the use by the EU of penal sanctions, and in favour of administrative sanctions, in the context of market abuse.134 This integrated approach would thus also rule out symbolic criminal law, that is, criminal law which is used to sanction behaviour affecting important interests, but which is not necessarily the most effective tool to actually discourage the undesirable behaviours.135 In addition to these two arguments, the next chapter will show that a third constitutional reason in favour of a joint deontological and utilitarian approach to criminalisation exists specifically within the EU legal order, stemming from the principle of subsidiarity. Before that, the concluding section of this chapter will discuss one last important aspect, namely the predominantely scholarly character of the discussion on criminal law and its tenous link with constitutional law. VI.  THE DOCTRINAL CHARACTER OF THE DEBATE ON THE LEGITIMACY OF CRIMINAL LAW

The debate on the legitimacy of criminal law is overall a doctrinal one. National constitutions traditionally do not contain explicit criminalisation criteria, setting binding norms as to what interests criminal law is meant to protect or which objectives it is meant to pursue. Interesting exceptions are the German and the Spanish Constitutions. They respectively require the criminalisation of preparatory acts for a war or aggression136 or abuse of the procedure for the exceptional suspension of citizens’ rights.137 However, these are more the exception than the rule and they are probably connected to the two countries’ history. 134 See ch 7. 135 The term is borrowed from Elholm and Colson, who make reference to Durkheim’s idea of criminal law as an expression of collective values. See T Elholm and R Colson, ‘The Symbolic Purpose of EU Criminal Law’ in R Colson and S Field (eds), EU Criminal Justice and the Challenges of Diversity: Legal Cultures in the Area of Freedom, Security and Justice (Cambridge, Cambridge University Press, 2016) 48–64. 136 Article 26(1) of the German Constitution. 137 Section 55 of the Spanish Constitution.

The Doctrinal Character of the Debate  41 Admittedly, national constitutions and human rights instruments include the values-based premises for the debate on the legitimacy of criminal law: a provision safeguarding individual liberty and human dignity.138 As was mentioned earlier, Italian scholars have argued that a legal obligation to limit criminal law to the protection of Rechtsguts flows directly from Article 13 of the Italian Constitution on the right to liberty.139 This chapter has argued that a joint deontological and utilitarian approach to criminal law can better protect the values of liberty and dignity. It could accordingly be argued that an obligation to follow this approach in criminalisation practice flows from constitutional norms on liberty and dignity. However, this argument has almost never been recognised as a binding, judicially enforceable one. A number of constitutional courts, such as those in Lithuania,140 Belgium,141 Italy142 and Germany,143 have recognised a legal obligation to use criminal law only as a last resort. This principle, which normally goes by the label of ultima ratio,144 is considered a negative formal counterpart to the harm principle.145 While the harm principle or the Rechtsgut Theory provides substantive criteria on when it is legitimate to criminalise, the principle of ultima ratio establishes a negative presumption against criminalisation. Even when there are good reasons to criminalise – for example, an important interest is at stake – administrative or civil law should be preferable if they can equally protect the relevant interest.146 Yet, only the Portuguese Constitutional Court has explicitly recognised an autonomous and legally enforceable obligation to use the criminal law to protect ‘defined legal assets’.147

138 For the right to liberty, see art 5 of the European Convention for the Protection of Human Rights and Fundamental Freedoms; art 6 of the Charter of Fundamental Rights of the European Union; and by way of an example, art 2(1) of the German Constitution, art 13 of the Italian Constitution and s 15 of the Spanish Constitution. For the inviolability of human dignity, see art 1 of the Charter of Fundamental Rights of the European Union, and by way of an example, art 1(1) of the German Constitution and s 10 of the Spanish Constitution. See also art 2 of the Italian Constitution, which refers to the inviolability of fundamental rights, jointly with art 27, which states that criminal responsibility is personal, and the function of penalties is the resocialisation of the convicted criminals. These two articles read jointly are normally interpreted as requiring that the dignity of offenders should not be violated. 139 Bricola, the proponent of the teoria del bene giuridico costituzionalmente orientato, grounded his plea for specific limits to the content of the criminal law in, among others, art 27(1) of the I­talian Constitution, which by referring to the personal character of criminal responsibility i­ndirectly protected individual dignity, and in Article 13 protecting personal liberty; Bricola (n 80). 140 Lietuvos Respubliskos Konstitucinis Tesimas, 1997-11-13 nutarimas, bylos Nr 4/97, pt IV.1. 141 Corte Costituzionale Italiana, sentenza n 1085, del 13 Dicembre 1988. 142 Cour Constitutionnelle de Belgique, affaire 145/2012, pt B.18. 143 BverfGE 88, 203 (258). 144 N Jareborg, ‘Criminalisation as a Last Resort (Ultima Ratio)’ (2004) 2(2) Ohio State Journal of Criminal Law 526. 145 Persak (n 580 22. 146 ibid. 147 Tribunal Constitucional Portugal, Ruling 179/12, 4 April 2012, para 6 (English summary available at: www.tribunalconstitucional.pt/tc/en/acordaos/20120179s.html).

42  Why Criminal Law? The suitability of the Rechtsgut Theory as a legally binding model for criminalisation was raised before the German Constitutional Court in the famous incest case.148 The question concerned the legality of German criminal law provisions which criminalised incest, and the argument was raised that the provision was unlawful as it did not protect any worthy interest. However, the Court did not openly endorse the theory of legal interests and ultimately decided the case on the basis of the principle of proportionality. Still, it should be mentioned that Judge Hassemer disagreed with the majority and endorsed the theory as a constitutionally binding one.149 The Italian Constitutional Court acknowledged the criminalisation criterion of offensività (criminal law must only be used to protect worthy interests) in several cases, but always as a supporting argument and never as an autonomous criterion to strike down the definitions of crimes. The relevant cases were decided on grounds of the principle of equality.150 Some scholars see a clear advantage in leaving criminal legal theory outside constitutional codification. Crystallising ideal models for the content of criminal law in judicially enforceable constitutional norms would unduly restrict the freedom of democratically elected legislators to set goals and select the means necessary to achieve them.151 Moreover, it risks diluting the complexities and sophistication of the doctrinal constructions to have them fit into constitutional norms.152 Du Bois-Pedain also observes that even if criminal theory tenets were to be incorporated into national constitutions, this would hardly change the impact of their actual normative force. The system of checks and balances between judicial control and the legislation choice of liberal states would in any case reduce the policing of criminalisation choices to a mild adjudication of an outer boundary of arbitrary state action rather than a thorough scrutiny of the substance of legislators’ choices.153 Conversely, other scholars consider the fact that the models for the legitimacy of criminal law have not been incorporated into national constitutions as a weakness. They have accordingly tried to anchor their doctrinal constructions in positive constitutional standards, as was the case of the Italian version of the legal interests theory, or argued in favour of constitutional amendments to include indications on the content of the criminal law.154 In any case, regardless of the possible advantages of constitutionalising

148 German Federal Constitutional Court (BVerfGE 120, 224) 26 February 2008. For a comment, see CF Stuckenberg, ‘The Constitutional Deficiencies of the German Rechtsgutslehre’ (2013) 3(1) Oñati Socio-legal Series 31–34. 149 Abweichende Meinung des Richters Hassemer zum Beschluss des Zweiten Senats vom 26 Februar 2008–2 BvR 392/07. 150 Corte Cost. 26 1979, RIDPP 1980, Corte cost 370 1996, CG 1996; and Corte Cost. 354 2002, GC 2000. For an exhaustive comment on each decision, see Manes (n 20) 212–34. 151 Stuckenberg (n 148). 152 Du Bois-Pedain (n 8) 320. 153 ibid 319. 154 Bricola (n 80); however, see also RS Frase, ‘What’s ‘Different’ (Enough) in Eight Amendment Law?’ (2013) Ohio State Journal of Criminal Law 9 on an extensive reading of US Supreme Court case law on excessive imprisonment; and D Baker, ‘Constitutionalising the Harm Principle’ (2008)

Conclusions  43 criminal legal theory in a de lege ferenda perspective, there is not at present any constitutional consensus (or acknowledgement really) for a specific model for the legitimacy of criminal law. VII. CONCLUSIONS

This chapter has explained why the question of the legitimacy of criminal law arises and how it has been addressed in criminal legal scholarship. Among the different theoretical models for a legitimate criminal law, the chapter has advocated a joint deontological and utilitarian approach. This approach is said to be more coherent with empirical reality and more suited to preserving values of human dignity and liberty. The debate is not specific to a particular national legal order, but has a universal value. Criminal law should be considered legitimate when important values are at stake and it is the only effective way to deter individuals from displaying behaviours which would harm such values. Indeed, every sovereign legislator is free to resort to criminal law as a regulatory instrument, but all liberal democracies which recognise individual liberty and dignity as founding values should shape their criminalisation policy accordingly. By being doctrinal and not depending on positive law, the relevant criminal legal theory categories, such as the deontological and utilitarian approaches to criminalisation, can travel across borders.155 However, the translation of this debate to the transnational setting, and especially to the EU context, is not self-evident. The next chapter will turn to this issue.

27 Criminal Justice Ethics 3, which identifies the constitutional sources of the harm principle in the US and Canadian systems respectively; and also Husak (n 51) 92–103, who advocates the actual introduction of a specific right not to be punished. 155 Lauterwein (n 3) Introduction.

2 Why EU Criminal Law? The Question of, and the Models for, the Legitimacy of Supranational Criminal Law

W

hen examining whether and how the debate on the legitimacy of criminal law unfolds at the EU level, a key aspect to consider is the non-state nature of the EU and the consequent limits to its range of action. The fact that the EU is a ‘limited government’ polity,1 has a bearing both on whether the question of the legitimacy of criminal law arises at the EU level and on which model is suitable for its legitimacy. First, as per the principle of conferral, the EU can only act within the remit of its attributed competences. This principle is spelled out in Article 5(2) TEU2 and it is repeated in different variations throughout the Treaties.3 In matters relating to criminalisation, the EU has nothing more than indirect competence – that is, it only has the competence to set minimum standards by means of harmonising directives.4 Member States have to implement these standards in national law, but in principle they have a certain freedom in terms of how to reach the objectives set by the EU. Against this background, one can wonder to what extent EU law can have an impact on individuals’ liberty and dignity by means of criminal law, independently of national law. In other words, given the limited powers of the EU, to what extent does the theoretical discussion on the legitimacy of criminal law have any relevance for the EU legal order? Section I explores this question. Second, next to conferral, other principles restrict EU action and especially bind the exercise of EU competences in all policy areas. These include first

1 D Chalmers, G Davies and G Monti, European Union Law (Cambridge, Cambridge University Press, 2010) 213. 2 ‘The Union shall act only within the limits of the competences conferred upon it by the Member States in the Treaties to attain the objectives set therein. Competences not conferred upon the Union in the Treaties remain with the Member States.’ 3 See arts 1(1), 4(1) and 13(2) TEU. 4 Article 83 TFEU. See below, nn 9, 12 and 13 on regulations in EU criminal law.

EU Relevance of the Debate on the Legitimacy of Criminal Law  45 traditional constitutional principles which limit public interference with individuals’ liberty in any national legal order (eg, the principle of proportionality, and the need to respect liberal values and fundamental rights). Alongside this, the principle of subsidiarity is specific to the EU and constrains the range of centralised regulation, as opposed to national regulation.5 Assuming that the debate on the legitimacy of criminal law is relevant to the EU legal order, it is necessary to examine the normative implications (if any) that these further limitations of EU action have on the debate. This is discussed in section II. The overall aim of this chapter is to introduce the main theme of the book – the legitimacy of EU criminal law – and to answer the first of the three research questions listed in the Introduction, namely which criminal law legitimacy model the EU constitutional values and principles imply. I.  IS THE DEBATE ON THE LEGITIMACY OF CRIMINAL LAW RELEVANT TO THE EU LEGAL ORDER?

In 1992 the Treaty of Maastricht granted the EU the competence to adopt Conventions and Joint Actions in criminal matters.6 The Treaty of Amsterdam enlarged the material scope of the EU competence and also listed Framework Decisions as a legislative instrument.7 Finally, the Treaty of Lisbon systematised and further enlarged the material scope of action for the EU in criminal matters, listing directives8 and in some cases regulations as criminal law instruments.9 The relevant legal basis for EU action in substantive criminal law – namely, the definition of crimes and sanctions – is Article 83 TFEU: The European Parliament and the Council may, by means of directives adopted in accordance with the ordinary legislative procedure, establish minimum rules concerning the definition of criminal offences and sanctions in the areas of particularly serious crime with a cross-border dimension resulting from the nature or impact of such offences or from a special need to combat them on a common basis. These areas of crime are the following: terrorism, trafficking in human beings and sexual exploitation of women and children, illicit drug trafficking, illicit arms 5 Admittedly, all states which have several levels of government have to address the ­problem of allocation of competences. Each federal state would thus have a functional equivalent to EU subsidiarity. However, reference here is made to the principle of subsidiarity as specifically interpreted within the EU legal order. On different criteria and the specificity of EU subsidiarity, see M Jachtenfuchs and N Krisch, ‘Subsidiarity in Global Governance’ (2016) 79 Law and Contemporary Problems 1, 10. 6 Article K.3 TEU, consolidated in Maastricht [1992] OJ C191/01. 7 Article 34(2)(b) TEU, consolidated in Amsterdam [1997] OJ C340/2. 8 Article 82(2) TFEU on harmonisation of criminal procedure; and art 83 TFEU on harmonisation of substantive criminal law, ie, the definition of crimes and sanctions. 9 Article 82(1) TFEU on measures on mutual recognition; art 84 TFEU on crime prevention speaks of measures which might include both directives and regulations. Harmonisation is excluded in these cases, though art 85 TFEU on the functioning of Eurojust refers to regulation, and so does art 86 TFEU on the establishment of the EPPO.

46  Why EU Criminal Law? trafficking, money laundering, corruption, counterfeiting of means of payment, computer crime and organised crime. … If the approximation of criminal laws and regulations of the Member States effective implementation of a Union Policy in an area which has been subject to harmonisation measures, the area concerned. Such directives shall be adopted by the same ordinary or special legislative procedure as was followed for the adoption of the harmonisation measures in question, without prejudice to Article 76.3.

While the material scope of Article 83 TFEU will be extensively discussed in Chapter 4, it is of interest here that the article only permits the harmonisation of the definition of crimes and sanctions through directives.10 Admittedly, in the past there has been some discussion on whether criminal law regulations, including the definition of fraud, could be adopted on the basis of Article 325(4) TFEU11 or whether the EU Regulation on the European Public Prosecutor, based on Article 86 TFEU, could include the definition of crimes.12 However, both possibilities have been ruled out in actual legislative developments.13 This is not to say that other Treaty norms or EU secondary law adopted on other legal bases cannot have an impact on the national definition of crimes.14 Free movement

10 On this provision, see extensively chs 4 and 5. 11 The article states that measures (and thus both directives and regulations) for the prevention of fraud should be taken, but it does not require that these measures be criminal. Member States remain free to decide whether they want to attach criminal sanctions or not. Satzger argued in favour of having criminal law regulations based on this provision: H Satzger, International and European Criminal Law (Oxford, CH Beck/Hart Publishing, 2012) 56. Sicurella argued in favour of having criminal law measures adopted on this basis, but opted for directives: R Sicurella, ‘“Prove tecniche” per una metodologia dell’esercizio delle nuove competenze concorrenti dell’Unione Europea in materia penale’ in G Grasso, R Sicurella and L Piccotti (eds), L’evoluzione del diritto penale nei settori d’interesse europeo alla luce del Trattato di Lisbona (Milan, Giuffrè, 2011) 896. 12 Article 86(2) TFEU states that: ‘The European Public Prosecutor’s Office shall be responsible for investigating, prosecuting and bringing to judgment, where appropriate in liaison with Europol, the perpetrators of, and accomplices in, offences against the Union’s financial interests, as determined by the regulation provided for in paragraph 1. It shall exercise the functions of prosecutor in the competent courts of the Member States in relation to such offences.’ The mention of offences against the EU’s financial interests as determined by the regulation raised the question of whether the regulation itself should contain a directly applicable definition of crimes; see K Ligeti, ‘Approximation of Substantive Criminal Law and the Establishment of the European Public Prosecutor’s Office’ in A Weyembergh (ed.), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Editions de l’Université de Bruxelles, 2013). However, see contra G Grasso, ‘Il Trattato di Lisbona e le nuove competenze penali dell’Unione europea’ in M Bertolino, M Eusebi and L Forti (eds), Studi in onore di Mario Romano (Naples, Jovene, 2011) 2344. 13 The Commission had initially proposed the Directive on the protection of the financial interests of the EU on the basis of art 325(4) TFEU (COM/2012/0363 final), but the text was eventually adopted on the basis of art 83(2) TFEU (Directive 2017/1371/EU of the European Parliament and of the Council on the fight against fraud to the Union’s financial interests by means of criminal law [2017] OJ L198/29). Similarly, the adopted EPPO Regulation does not include any definition of crime: Council Regulation 2017/1939/EU implementing enhanced cooperation on the establishment of the European Public Prosecutor’s Office [2017] OJ L283/1. On this, see further ch 7, s V. 14 Klip also includes EU competition law as form of EU criminal law, due to the harsh and punitive character of some EU competition law sanctions. In this he follows the approach of the Strasbourg

EU Relevance of the Debate on the Legitimacy of Criminal Law  47 Treaty provisions have been interpreted as requiring the elimination of criminal provisions that are an obstacle to free movement.15 Moreover, EU instruments, both directives and regulations, envisaging administrative sanctions in a certain policy area can arguably impose de-criminalisation obligations. Mitsilegas states that a choice in favour of administrative sanctions by the EU legislator, based on the grounds of effectiveness (administrative law is considered to be more effective than criminal law) imposes a duty on Member States to stick to the administrative regulatory tools in the given policy area and to refrain from turning to criminal law sanctions.16 However, in both these cases, the impact on national criminal law is more of a side-effect of EU law than the result of a proper supranational criminalisation policy.17 Moreover, when EU law implies some de-criminalisation obligations for Member States, this does not raise a question of when to use EU criminal law. As such, this chapter mainly focuses on the influence of EU criminal law directives on national criminalisation policies. Despite the EU only being allowed to set minimum rules in criminal law matters, directives often present national legislators with detailed ready-made definitions of crimes and sanction thresholds, from which Member States have little room to deviate (see section I.A). Member States are naturally legally bound to incorporate such definitions into national law, and failure to do so could trigger infringement proceedings (see section I.B). Admittedly, other means for the private enforcement of EU law, such as direct effect, were traditionally prevented in criminal matters. Nevertheless, the bar on the direct effect of directives in criminal law has arguably been nuanced by the Taricco and MAS and MB case law (see section I.C). In light of this, it is submitted that the choices of the EU in matters of criminalisation require justification in a similar manner to national criminalisation choices (see section I.D), if not even more justification, considering the greater repressiveness of EU criminal law (see section I.E).

Court in its definition of criminal law. See A Klip, European Criminal Law, 3rd edn (Antwerp, Intersentia, 2016) 5. However, the analysis here only considers the influences of EU law on what is explicitly considered as criminal law at the national level. 15 This phenomenon is referred to as ‘negative harmonisation’ in criminal law. On this and for a detailed account of internal market case law which influenced national criminal law, see S Miettinen, Criminal Law and Policy in the European Union, 2nd edn (London, Routledge, 2014) 123–32. 16 V Mitsilegas, ‘From Overcriminalisation to Decriminalisation: The Many Faces of Effectiveness in European Criminal Law’ (2014) 5(3) New Journal of European Criminal Law 416, 422. The double track between administrative and criminal law sanctions has been used in the field of the fight against terrorism in the context of money laundering and market abuse. On this, see extensively E Herlin-Karnell, ‘Is Administrative Law Still Relevant? How the Battle of Sanctions Has Shaped EU Criminal Law’ in M Bergstroom, V Mitsilegas and T Konstatinides (eds), Research Handbook on EU Criminal Law (Cheltenham, Edward Elgar, 2015) 233. 17 It should also be noted that even more indirect influence on the shape of national criminal law can be derived from the adoption of judicial cooperation instruments, which in order to work smoothly begs the question of whether the harmonisation of substantive and criminal law is needed. On this indirect influence, see the various contributions in G Giudicielli-Delage and S Manacorda (eds), L’intégration Pénale Indirecte (Paris, Société de législation comparée, 2005). On the duty to resort to harmonisation in order to support cooperation, see extensively ch 3, s V.

48  Why EU Criminal Law? A.  EU Definitions of Crimes and Member States’ Limited Discretion EU criminal law directives normally include the following formulation – ‘Member States shall take the necessary measures to ensure that the following acts are punishable’ – and then detail the relevant acts. Despite only being intended to create a general framework, in practice, EU criminal law directives quite often contain very detailed descriptions of offences and sanctions. Examples can be found, among others, in texts on fraud,18 counterfeiting of means of payment,19 money laundering,20 terrorism,21 drug trafficking22 and trafficking in human beings.23 Admittedly, not all the offences listed in these texts can be considered model definitions. Indeed, some of the aforementioned instruments have been criticised for their vagueness.24 However, such criticism can also be levelled at national definitions of offences where they are often poorly drafted and require further judicial interpretation. It is only exceptionally that EU texts explicitly leave a margin of discretion to Member States. For instance, in the field of market abuse25 and attacks against information systems,26 the EU texts provide definitions of crimes, but leave Member States with the discretion to use criminal law only for the most serious cases. Other examples also exist in the field of organised crime, child pornography, and racism and xenophobia, where Member States are given a choice on what to criminalise. Yet these exceptions

18 Articles 2, 3 and 4 of Directive 2017/1371 (n 13). 19 Article 3 of 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, and replacing Council Framework Decision 2000/383/JHA [2000] OJ L151/1; arts 2, 3 and 4 of Council Framework Decision 2001/413/JHA on combating fraud and counterfeiting of non-cash means of payment [2001] OJ L149/1. 20 Article 3 of Directive 2018/1673/EU of the European and of the Council of 23 October 2018 on combating money laundering by criminal law [2018] OJ L284/22. 21 Articles 3–12 of Directive 2017/541/EU of the European Parliament and of the Council of 15 March 2017 on combating terrorism and replacing Council Framework Decision 2002/475/JHA and amending Council Decision 2005/671/JHA [2017] OJ L88/6. 22 Article 2 of Council Framework Decision 2004/757/JHA of 25 October 2004 laying down minimum provisions on the constituent elements of criminal acts and penalties in the field of illicit drug trafficking [2004] OJ L335/8. 23 Article 2 of Directive 2011/36/EU of the European Parliament and of the Council of 5 April 2011 on preventing and combating trafficking in human beings and protecting its victims and replacing Council Framework Decision 2002/629/JHA [2011] OJ L101/1. 24 See the criticism of the definition of ‘participation in a criminal organisation’ in the r­ elevant Framework Decision, the concept of public order in the Framework Decision on racism and ­xenophobia, and the concept of ‘adult looking like a child’ in the Directive on child pornography in European Criminal Policy Initiative, ‘The Manifesto on European Criminal Policy’ 2011 (2011) 1(1) European Criminal Law Review 86, 92 ff. 25 See arts 3, 4 and 5 of the Market Abuse Directive, which impose the criminalisation of certain conduct when they are serious, but only provides some guidelines as to what serious crime is (Directive 2014/57/EU of the European Parliament and of the Council of 16 April 2014 on criminal sanctions for market abuse [2014] OJ L173/17). 26 Article 3, 4, 5 and 6 of Directive 2013/40/EU of the European Parliament and of the C ­ ouncil of 12 August 2013 on attacks against information systems and replacing Council Framework Decision 2005/222/JHA [2013] OJ 218/8.

EU Relevance of the Debate on the Legitimacy of Criminal Law  49 can be easily explained in light of the particular sensitivity of the matter, which has made a higher degree of harmonisation either politically unfeasible or unwise from a normative perspective, rather than as a result of the EU legislature’s self-restraint and commitment to keeping harmonisation to a minimum. The Framework Decision on the fight against organised crime leaves Member States with the option to criminalise actual participation in a criminal organisation or even just the agreement to jointly commit a criminal offence.27 This is because common law and civil law countries traditionally have a diverging approach to the definition of participation in a criminal organisation, the former sanctioning the agreement and the latter only the actual participation.28 Given that the unanimity rule applied at the time to criminal matters, the relevant Joint Action and Framework Decisions probably needed to include the option to adopt one or other definition of the crime. Similarly, the Directive on child pornography gives Member States the option of whether to criminalise virtual child pornography.29 Likewise, the Framework Decision on racism and xenophobia leaves Member States with the choice of either criminalising the public denial or gross trivialisation of international crimes as such or only if these have been established by the final decision of an international court.30 The EU legislature probably did not want to take a strong normative stance on these two points. Criminalisation of virtual child pornography and of Holocaust denial are both very controversial from a doctrinal point of view, since many authors fail to see the harmful character of either conduct.31

27 Article 2 of Council Framework Decision 2008/841/JHA of 24 October 2008 on the fight against organised crime [2008] OJ L35/8. 28 F Calderoni, Organized Crime Legislation in the European Union, Harmonization and Approximation of Criminal Law, National Legislations and the EU Framework Decision on the Fight against Organized Crime (Dordrecht, Springer, 2010). 29 Article 5(7) and 5(8) of Directive 2011/92/EU of the European Parliament and of the Council of 13 December 2011 on combating the sexual abuse and sexual exploitation of children and child pornography, and replacing Council Framework Decision 2004/68/JHA [2011] OJ L335/1. 30 Article 1(2) and 1(4) of Council Framework Decision 2008/913/JHA of 28 November 2008 on combating certain forms and expressions of racism and xenophobia by means of criminal law [2008] OJ L328/55. 31 On virtual child pornography, see among scholars from the US, where the debate on virtual child pornography is quite developed, see D Johnson, ‘Why the Possession of Computer-Generated Child Pornography Can Be Constitutionally Prohibited’ (1994) 4 Albany Law Journal of Science & Technology 311, 327; N Malamuth and M Huppin, ‘Drawing the Line on Virtual Child Pornography: Bringing the Law in Line with the Research Evidence’ (2007) NYU Review of Law & Social Change 773, 794 ff. Before the adoption of the Directive, only a few Member States included provisions explicitly targeting the conduct of virtual child pornography (eg, Poland, Cyprus, Italy, Ireland and the UK). Yet, even in these countries, the criminalisation of this conduct has been the subject of considerable criticism; by way of example on the Italian system, see C Sotis, ‘Diritto penale della rete e prospettiva europea: il caso della pedopornografia virtuale’ (2011) 4 Diritto dell’informazione e dell’informatica 561. For an overview of the criminalisation of virtual child pornography throughout the EU, see the European Commission, ‘Study on Criminal Sanction Legislation and Practice in Representative Member States’, 22 ff, https://publications.europa.eu/en/publication-detail/-/publica tion/35472a32-42d0-43ae-b86b-f75d0f1a1832. On the Framework Decision on racism and xenophobia, and Holocaust denial, see extensively ch 7.

50  Why EU Criminal Law? Outside of these cases, it is generally acknowledged that Member States cannot deviate from the minimum rules set by the EU by adding further details to the criminal law definitions: ‘Member States … remain free to criminalise also additional forms of behaviour/acts. They are, however, not allowed to criminalise less than that, which is required by the minimum rules, by introducing additional prerequisites for punishment.’32 Klip in particular warns that the understanding of ‘minimum rules’ as allowing additional actions can lead to unwanted results, such as defeating the actual purpose of the legal instrument. In the Spector Photo case, the court held that the purpose of Directive 2003/6 on Insider Dealing and Market Manipulation is to ensure the integrity of Community financial markets and to enhance investor confidence in those markets.33 In Klip’s view, if Member States were to add an additional element to the definition of the criminal conduct, such as the presence of a specific mental element, the overall purpose of the Directive would be weakened.34 Hans G Nilson goes even further by advocating an understanding of minimum rules as also setting a threshold for maximum criminalisation. He argues that the standards contained in the EU instruments set not only a minimum but also a maximum standard. Member States cannot criminalise less than what is included in the Directive, but equally they cannot criminalise more.35 They need to include in national legislation the definition included in the EU text as it stands. Nielson considers that directives should be ‘unifying instruments’36 in this context. Acting otherwise would endanger legal certainty and might lead to excessive criminalisation hindering full enjoyment of the four freedoms.37 However, there is no unanimity on this position. Petter Asp opposes it, arguing that Member States remain free to criminalise further by dropping some of the constituent elements contained in the definition of crimes.38 He argues that when the EU texts want to set a threshold for maximum criminalisation, it does so explicitly.39 Moreover, it would be very difficult in practice to define the scope

32 DB Hecker, Europaisches Stratfrech, 3rd edn (Dordrecht, Springer, 2010) 371, cited in P Asp, The Substantive Criminal Law Competence of the EU (Stockholm, Stiftelsen Skrifter utgivna av Juridiska fakulteten vid Stockholms universitet, 2012) 111. 33 Case C-45/08 Spector Photo Group NV, Chris Van Raemdonck v Commissie voor Bank-, Financie- en Assurantiewezen (CBFA) [2009] ECR I-12073, para 37. 34 See Klip (n 14) 182. 35 H Nilsson, ‘How to Combine Minimum Rules with Maximum Legal Certainty?’ (2011) 4 Europarättslig tidskrift 665, 668. 36 ibid. 37 ibid. 38 Asp (n 32) 116–27. 39 Asp (n 32) finds examples of provisions which explicitly limit criminalisation in arts 12, 13 and 14 of the Directive on electronic commerce (Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the Internal Market [2000] OJ L178/1), but also in art 5 of the Directive on ship-source pollution (Directive 2005/35/EC of the European Parliament and of the Council of 30 September 2005 on ship-source pollution and on the introduction of penalties for infringements [2005] OJ L255/1) and art 4(3) of the Directive on sanctions for employers of illegally staying

EU Relevance of the Debate on the Legitimacy of Criminal Law  51 of the application of maximum criminalisation and thus what would fall within the prohibited ‘further criminalisation’.40 In any case, regardless of the debate on minimum rules as minimum or maximum criminalisation, it remains the case that directives harmonising definition of crimes often include very precise definitions of crimes. Member States are bound to implement such policy choices in several instances and almost verbatim in national legal systems. Failure to do so can trigger the activation of any of the relevant mechanisms to enforce EU law, as will be illustrated below. B.  Enforcement of EU Criminal Law Obligations: Infringement Proceedings and the Bar on Direct Effect There exist at least four means of enforcing Member States’ EU law obligations: infringement proceedings, namely the possibility for the court to sanction a state that is in breach of EU law; state liability, namely the possibility for an individual of claiming compensation for a damage suffered as a result of the state not complying with EU norms granting individual rights; direct effect, namely the possibility of directly relying on EU norms in individual cases; and indirect effect, namely the duty for national authorities to interpret national law in light of EU law, even if it is unimplemented. When EU criminal law was regulated under the third pillar, the Treaty explicitly ruled out the possibility of infringement proceedings41 and direct effect.42 After the Treaty of Lisbon, all these means of enforcement are in principle available in the field of EU criminal law, even though some specifics exist with respect to enforcing EU criminalisation obligations through state liability and direct and indirect effect. State liability is not a useful tool to enforce criminalisation provisions in unimplemented directives. Provisions setting criminalisation obligations do not envisage any individuals’ rights, whereas this is one of the requirements for claiming compensation under state liability.43 Conversely, individuals could claim compensation under state liability if they suffer damage as a result of the state failing to implement one of the procedural rights directives, which explicitly grants rights to individuals in criminal proceedings.44 third-country nationals (Directive 2009/52/EC of the European Parliament and of the Council of 30 September 2009 providing for minimum standards on sanctions and measures against employers of illegally staying third-country nationals [2009] OJ L168). 40 Asp (n 32) 117. See also S Peers, EU Justice and Home Affairs Law (Oxford, Oxford University Press, 2011) 763. 41 Article 35 TEU, consolidated in Amsterdam [1997] OJ C340/02. 42 ibid art 34. 43 The doctrine was established in Joint Cases C-6/90 and 9/90 Francovich and Bonifaci [1991] ECR I-05357. The relevant criteria are spelled out in Joint Cases 49/93 and 48/93 Brasserie du Pêcheur/Factortame [1996] ECR I-01029, para 51. 44 See, for instance, Directive 2012/13/EU of the European Parliament and of the Council of 22 May 2012 on the right to information in criminal proceedings [2012] OJ L142/1.

52  Why EU Criminal Law? Furthermore, the case law of the Court of Justice has imposed some limitations on the direct and indirect effect of EU law in criminal matters. In particular, for a long time, the Court has consistently argued that individuals’ criminal liability cannot be aggravated in individual cases as a result of the direct application of unimplemented directives45 or of the interpretation of national norms in light of EU norms.46 The Court relied on two arguments. In Kolpinghuis Nijmegen, it resorted to the traditional EU constitutional law estoppel argument that a state cannot rely on its own failure against individuals by depriving them of rights contained in directives it had not implemented.47 The flipside of this argument is that obligations cannot be imposed on individuals as a result of unimplemented directives. Thus, in litigation between private parties, one individual cannot enforce an unimplemented directive against another individual. In a criminal law case, this means that a state cannot rely on a directive it has not implemented to impose additional duties on individuals that would stem from the EU criminalisation provisions.48 Second the Court relied on the principle of legality of criminal law. This argument can be first appreciated in Criminal Proceedings against X, which concerns the interpretation of national criminal law in light of EU law.49 According to the Court, as per the principle of legality, criminal liability can only be grounded in an act of a national parliament, and criminal norms cannot be interpreted extensively, not even in light of EU law. Acting otherwise would aggravate the position of the individual as a result of EU law, going against the original intention of the national legislator, and thus breaching the principle of legality.50 This argument can also be appreciated in the later Berlusconi case, which concern the ‘exclusionary’ direct effect of directives.51 In this scenario, the direct effect of EU norms implies setting aside (and thus excluding) national norms which are in conflict with EU law and applying other existing national rules. This is a variation of the more traditional ‘substitution’ direct effect, which implies setting aside national norms that are in breach of EU norms and applying the relevant EU provisions instead. EU law thus substitutes national law.52 In the Berlusconi case, the referring judge asked

45 Case C-80/86 Criminal Proceedings against Kolpinghuis Nijmegen BV [1987] ECR I-03969; Joined Cases C-74/95 and C-129 Criminal Proceedings against X [1996] ECR I-06609, para 24; and Joined Cases C-387/02, C-391/02 and C-403/02 Criminal Proceedings v Silvio Berlusconi and Others [2005] [2005] ECR I-03565, para 74. 46 Case C-105/03 Criminal Proceedings against Maria Pupino [2005] ECR I-05285, paras 45–46. 47 For early cases in which the argument is raised, see Case C-148/78 Criminal Proceedings against Ratti [1979] ECR I-01629; Case C-8/81 Becker v Finanamt Munster-Innenstadt [1982] ECR I-00053. 48 Kolpinghuis (n 45) para 9. 49 Criminal Proceedings against X (n 45) para 25. 50 ibid. 51 Berlusconi (n 45). See s I.E below for a more detailed description of the case. 52 For a detailed explanation of the differences between substitution and exclusion direct effect, see S Prechal, EC Directives in EC Law, 2nd edn (Oxford, Oxford University Press, 2005), especially ch 9.

EU Relevance of the Debate on the Legitimacy of Criminal Law  53 whether the more lenient national criminal law, which in principle should apply but was in breach of EU law, should be set aside and whether the pre-existing stricter national criminal provisions should apply instead. Briefly, it asked whether EU law can be given exclusionary direct effect in criminal matters. An answer in the positive sense would have clashed with the Italian constitutional interpretation of the principle of legality, which includes lex mitior as a corollary, namely the principle that the more lenient legislation should apply to a case even if it is adopted after the facts of the case. The Court ruled out this possibility as incompatible with the principle of legality, and notably the aspect of the principle, which requires the more lenient (even if subsequent) criminal law to apply.53 It thus ruled against the possibility of EU exclusionary direct effect if this aggravates the position of the offender. Having said that, this consistent position on a ban on criminalisation directive’s direct effect was partially nuanced in the recent case law of Taricco54 and MAS & MB.55 C.  Taricco and MAS and MB as a Challenge to the Bar on the Direct Effect? The Taricco case concerned the interpretation of Italian norms on statutes of limitation. Italian law set particularly short deadlines, which made prosecution in very complex cases like financial crime cases particularly difficult. The Court of Justice was then asked whether Italian legislation on this point was compatible with EU obligations to combat fraud against the financial interests of the EU enshrined in Article 325 TFEU. The Court first considered that Article 325 TFEU, read in conjunction with Article 4(3) TEU on loyal cooperation, imposes an obligation to Member States both to collect VAT taxes and to effectively combat VAT fraud.56 This is because a part of Member States’ contribution to the EU budget comes from VAT taxes. As a consequence, failure to duly collect these revenues negatively impacts on the Member States’ contribution to the EU budget. The Court then admitted that, in principle, the Treaty provision left the choice between criminal or administrative sanctions for fraud to the Member States. However, it stated that for serious VAT fraud, such as that in the facts at stake, criminal sanctions might be essential.57 It then added that strict rules on statutes of limitation such as the Italian ones might neutralise the deterrent effect of criminal sanctions.58 It nonetheless left it to the referring judges 53 Berlusconi (n 45) 78. 54 Case C-105/14 Criminal Proceedings against Ivo Taricco and Others [2015] ECR Court General Reports. 55 Case C-105/14 Criminal Proceedings against MAS and MB [2017] ECR Court General Reports. 56 Taricco (n 54) paras 36–38. 57 ibid paras 39, 42–43. 58 ibid para 46.

54  Why EU Criminal Law? to consider whether the Italian rules would actually prevent the imposition of effective and dissuasive penalties in ‘a significant number of cases’. If that were to be the case, then these norms had to be disapplied. Acting otherwise would have been in breach of EU primary law obligations.59 Unconvinced by the Taricco judgment, the Italian Constitutional Court issued a second reference to the European Court basically reformulating the question. In this second case, MAS and MB, the Court adopted a more conciliatory approach.60 It admitted that at the time of the first reference, the Italian Republic could lawfully abstain from disapplying national norms despite their tension with the EU obligation of prosecuting fraud against the financial interests of the EU. This was because EU law had not yet harmonised the norms on the prescription of VAT fraud.61 The EU has subsequently harmonised such norms in the 2017 Directive on the Protection of the Financial Interests of the EU (hereinafter the PIF Directive).62 From the MAS and MB case, one can thus argue that, a contrario, had the PIF Directive been in force at the time of the facts, Italy would have had an obligation to disregard the national norms on statutes of limitations. If similar criminal law cases arise in the future, the Italian authorities must disapply national norms on statutes of limitation and apply the harmonised EU norms. The Court does not specify whether the fact that the Directive has been implemented or not should play a role. However, it is telling that the judgments were handed down while the implementation period was still running. The Court thus appears to suggest (or at least does not explicitly rule out the idea) that Member States must give effect to the PIF Directive’s norms on statutes of limitation, even if they are unimplemented. The facts in Taricco and MAS and MB have some analogies with the cases on direct effect in criminal law discussed in the previous paragraph, which makes them interesting in relation to the question of the enforcement of EU criminal law. In both scenarios, the question is what norms (either EU or national) should regulate prosecution and conviction in national cases? In Kolpinghuis, Criminal Proceedings v X and Berlusconi, the question posed to the Court referred to whether or not national definition of crimes that are in conflict with EU norms should be disapplied or interpreted in a manner consistent with EU law. In Criminal Proceedings v X and Berlusconi, the Court held that the principle of legality prevents grounding criminalisation solely in EU norms. In Taricco, the main question is rather whether national norms on statutes of limitation which are in breach of EU law were to be disapplied and the EU norms on statutes of limitation of VAT fraud should be applied instead.



59 ibid

para 47. and MB (n 55). 61 ibid paras 43, 44. 62 Article 12 of Directive 2017/1371 (n 13). 60 MAS

EU Relevance of the Debate on the Legitimacy of Criminal Law  55 The interested parties nonetheless queried whether or not the principle of legality would not similarly bar the application of EU norms on statutes of limitation in place of national norms. The Court gave a negative answer in response to this. It recalled that according to Strasbourg case law, Article 7 ECHR does not extend to rules on statutes of limitations. As a consequence, allowing for the disapplication of criminal law in order to comply with EU obligations would not amount to a breach of the principle of legality.63 However, such an understanding of the relationship between legality and statutes of limitations is not universally accepted. Some legal systems, including the Italian system, consider statutes of limitation to be covered by the principle of legality, which is intended as a safeguard for the individual.64 This is because, on a factual level, norms on statutes of limitations have a determining impact on the criminal liability of the offender. The deadline they set has a strong influence on whether a prosecution can be brought to a successful end in particularly complex cases. Incidentally, this was exactly the issue in Taricco – that too strict statutes of limitations might lead to impunity. As a consequence, when in Taricco the Court permitted the disapplication of national norms on statutes of limitation, it de facto permitted negative consequences for the position of the suspect in criminal proceedings to flow from EU norms. If then one agrees with the a contrario interpretation of MAS and MB outlined above, such negative consequences for the position of the suspect would flow from an unimplemented directive. Such an interpretation does not technically overrule the Court’s stance on the ban on criminalisation directives having substitution direct effect. However, it is at least in tension with its rationale, namely preventing an aggravation of the position of the offender as a direct consequence of an unimplemented EU directive. Admittedly, the PIF Directive, which was at stake in these cases, implements an EU primary law obligation to prosecute fraud, which makes it a special case. It remains to be seen whether similar reasoning could be applied to unimplemented directives which do not specify an obligation already present in EU primary law. To conclude, EU criminal law did not always benefit from all means of enforcement. Indeed, the possibility of infringement proceedings and direct effect, with the limitations established by the Court, were only introduced in this policy with the Treaty of Lisbon. Yet, we must not rule out further evolutions in the enforcement of EU criminal law obligations. Taricco and MAS and MB might have inaugurated a trend where, in future case law, the Court could abandon the principle of lack of direct effect of a directive in criminalisation matters.

63 Taricco (n 54) para 57. 64 See Italian Constitutional Court, decision no 324 of 30 July 2008, or decision no 393 of 23 October 2006.

56  Why EU Criminal Law? D.  The Significant Impact of EU Criminalisation Choices and Their Ensuing Need for Legitimation Sections I.B and I.C illustrated that EU autonomous policy choices on the definition of crimes and sanctions can have a direct impact on the liberty and dignity of individuals. EU criminal law framework decisions and directives were, and often are, drafted in a very precise way. It is generally acknowledged that Member States cannot add any specification in relation to the definition of crimes or the level of sanctions which would result in a restriction of the number of punishable behaviours or the severity of punishment. Member States are bound to implement these definitions of crimes, in some cases almost verbatim, in their national legal systems. Failure to do so might result (and has resulted) in infringement proceedings. The Court did not have the possibility of bringing sanctions against Member States for a failure to implement EU criminal law in the past.65 Yet, as soon as this possibility was made available by the Treaty of Lisbon, the trend has been radically changed, with the Commission being very active in securing the enforcement of EU criminal law. So far, the EU has adopted eight directives containing criminalisation obligations. For two of these – namely the Directive on the protection of EU financial interests66 and the Money Laundering Directive67 – the implementation deadline is still pending at the time of writing. Meanwhile, the deadline for the implementation of the Directive on terrorism expired in October 2018, but implementation data is not yet available.68 The Commission has already taken steps to launch 41 i­nfringement proceedings to secure the enforcement of the remaining five Directives for which the implementation deadline has already passed, namely the Directive on trafficking in human beings,69 the Directive on child pornography and sexual exploitation of children,70 the Directive on attacks against information systems,71

65 See, for instance, the Council Joint Action 98/742/JHA Concerning Corruption in the Private Sector of 22 December 1998 [1998] OJ L358/2, which three Member States failed to implement. For a list of Member States that have implemented the Framework Decisions and those who that have not, see https://eur-lex.europa.eu/legal-content/EN/NIM/?uri=CELEX:32003F0568. 66 Article 17 of Directive 2017/1371 (n 13). 67 Article 4 of Directive 2018/843/EU of the European Parliament and of the Council of 30 May 2018 amending Directive 2015/849/EU on the prevention of the use of the financial system for the purposes of money laundering or terrorist financing and amending Directives 2009/138/EC and 2013/36/EU. 68 Article 28 of Directive 2017/541 (n 21). 69 Proceedings were launched against Austria (Reference Number 20130204), Germany (Reference Number 20130214), Greece (Reference Number 20130217), France (Reference Number 20130221), Portugal (Reference Number 20130244) and Slovakia (Reference Number 20130253). 70 Proceedings were launched against the UK (Reference Number 20140219), Hungary (Reference Number 20140115), Cyprus (Reference Number 20140036) and the Netherlands (Reference Number 20140168). 71 Proceedings were launched against Belgium (Reference Number 20150512), Bulgaria (Reference Number 20150514), Ireland (Reference Number 20150524), Greece (Reference Number 20150521) and Slovenia (Reference Number 20150529).

EU Relevance of the Debate on the Legitimacy of Criminal Law  57 the Directive on market abuse,72 and the Directive on the protection of euros and other currencies against counterfeiting.73 Moreover, failure to implement EU law might be also corrected with other means. Directives including criminalisation obligations cannot yet be enforced by means of direct effect. However, given the Taricco and MAS and MB cases, the possibility of change in the future cannot be ruled out. In light of the above, it is argued that the EU has de facto criminalisation powers. The law applied within domestic criminal proceedings is national criminal law because national implementation is formally necessary. The individual offender’s dignity and liberty are thus formally endangered on the basis of national law; however, they are substantially affected on the basis of policy decisions taken at the supranational level.74 This conclusion is naturally not meant to oversimplify the complex relationship between the supranational and national policy-making levels. In the pre-Lisbon framework, the Council was the only body, involved in the adoption of Framework Decisions, and the relevant decision-making procedure was unanimity. National governments were thus the most important players. It is not unknown for them to have proposed and have approved at the supranational level initiatives that were high on the national agenda.75 However, even within the Lisbon framework, which envisages a qualified majority in the Council and involves the European Parliament as co-legislator, national governments can play an important role. They can lobby for a proposal to be adopted and decisions may still be taken by consensus within the Council.76 Briefly, the national and supranational policy-making levels are naturally intertwined. The point here is that policy choices taken at the supranational

72 Proceedings were launched against Austria (Reference Number 20160583), Belgium (Reference Number 20160590), Bulgaria (Reference Number 20160596), Cyprus (Reference Number 20160604), the Czech Republic (Reference Number 20160609), Greece (Reference Number 20160624), Spain (Reference Number 20160627), Croatia (Reference Number 20160638), Hungary (Reference Number 20160644), Ireland (Reference Number 20160646), Luxembourg (Reference Number 20160655), Malta (Reference Number 20160658), the Netherlands (Reference Number 20160660), Poland (Reference Number 20160664), Portugal (Reference Number 20160671), Romania (Reference Number 20160677), Sweden (Reference Number 20160680) and Slovakia (Reference Number 20160689). 73 Proceedings were launched against Belgium (Reference Number 20160488) Greece (Reference Number 20160517), Ireland (Reference Number 20160538), Malta (Reference Number 20160549), Slovakia (Reference Number 20160575), Bulgaria (Reference Number 20170247), Croatia (Reference Number 20170262), Luxembourg (Reference Number 20170266) and Slovenia (Reference Number 20170275). 74 On the subject of criminalisation as a result of policy decisions taken at the EU level, see Sicurella (n 11) 56. 75 See more on this phenomenon in ch 7, s I. 76 On institutional changes within the Area of Freedom, Security and Justice from an intergovernmental to a ‘communitarized procedure’, especially the role for the Council, and the impact these have on the content of policies, see F Trauner and A Ripoll Serventi, Policy Change in the Area of Freedom, Security and Justice: How EU Institutions Matter (Abingdon, Routledge, 2015), especially the contributions of de Hert and Papaconstantinou, and Mitsilegas and Vavoula.

58  Why EU Criminal Law? level (whether following an EU or a national agenda) and afterwards incorporated into EU law text substantially, even if not formally, affect the position of individual citizens. It thus seems fair to conclude that the doctrinal question of what legitimates the restriction of the liberty and dignity of individuals by means of criminal law is also relevant to the EU legal order. E.  The Repressive Potential of EU Criminal Law and its Accentuated Need for Legitimacy Further, it could be argued that EU criminalisation policy choices affect individuals even more than purely national choices. The question of the legitimacy of EU criminal law is thus possibly even more pressing at the supranational level. This is for at least two reasons. First, the previous chapter highlighted the fact that criminalisation choices require justification because criminal convictions have a stigmatising impact on the dignity of individuals and because they can imply deprivation of liberty if imprisonment is imposed as a sanction. National criminal law normally envisages a wide range of criminal sanctions, including imprisonment, but also fines and community sanctions.77 Although primary law does not give any specific indication of the type of sanctions, most of the EU harmonising texts oblige Member States to attach imprisonment sanctions. Older instruments, namely joint actions or conventions, simply included a reference to ‘effective proportionate and dissuasive criminal penalties involving deprivation of liberty which can allow for extradition’.78 Some Framework Decisions included a reference to ‘Effective Proportionate and Dissuasive Criminal Sanctions’ (emphasis added),79 while most Framework Decisions and directives contain specific references to maximum imprisonment sanctions.80 Conversely, there is a striking absence in

77 These are all sorts of sanctions which do not imply detention, but which must be served within the community, eg, wearing an electronic tag or doing unpaid work, and which require, most of the time, regular meetings with a probation officer in charge of supervising the execution of the sentence. For a list, see art 1(2) of the Council Framework Decision 2008/947/JHA of 27 November 2008 on the application of the principle of mutual recognition to judgments and probation decisions with a view to the supervision of probation measures and alternative sanctions [2008] OJ L237/27. 78 See Joint Action 98/742/JHA (n 65). 79 See art 6(1) of Council Framework Decision 2005/222/JHA of 24 February 2005 on attacks against information systems [2005] OJ L69/67. 80 See art 4 of Directive 2011/36/EU (n 23). Only on two occasions did the Commission also propose the introduction of a minimum threshold for the sanctions in its ‘Proposal for a Directive of the European Parliament and the Council on the protection of the euro and other currencies against counterfeiting by criminal law, and replacing Council Framework Decision 2000/383/JHA’ (COM (2013)42 final) and its ‘Proposal for a Directive of the European Parliament and the Council on the protection of the Union’s financial interests by means of criminal law COM(2012)363 final’. However, in both cases, such provisions did not survive negotiations, and the final texts only contain indications as to the maximum of the sanctions.

EU Relevance of the Debate on the Legitimacy of Criminal Law  59 the EU texts of alternative sanctions to detention. Fines and other penalties are mentioned for legal persons.81 However, no community sanctions or probation orders are mentioned in any instrument for natural persons.82 Supranational criminal law that implies deprivation of liberty is thus the rule and not the exception. Second, EU-derived definitions of crimes benefit from a special resistance in national legislation since they are harder to de-criminalise. National legislatures normally have the freedom to backpedal and de-criminalise certain behaviours when they judge that the conditions for having criminal law are no longer fulfilled. Yet, this is not possible if the criminalisation obligation derives from EU law. De-criminalising would mean running counter to EU obligations. The Berlusconi case mentioned above83 is a good illustration of the tensions that de-criminalisation in a field regulated by EU law can trigger. The case originated in Italian national criminal proceedings, where Silvio Berlsuconi and others were being prosecuted for the offence of false accounting. The Italian legislature had de-criminalised the facts for which the accused were prosecuted after the facts had occurred. The Italian judge was thus faced with a dilemma: either applying the old, more severe Italian legislation, which was in compliance with EU law, or the subsequent, more lenient regime, as the general principles of Italian criminal law would require. It decided to refer the case to the Court of Justice. As mentioned above, the Court admitted that in light of the principle of legality, which it said included the principle of lex mitior, this mismatch between EU criminal law and the more lenient Italian criminal law could not go to the detriment of the offender in a concrete case. As such, he would have the right to benefit from this more lenient regime.84 However, the Court did not in any way imply that Italy was relieved of its implementation obligations. Advocate General Bobek addressed this point in a more explicit manner in Scialdone, a case that concerned a similar situation, namely more lenient national criminal law in non-compliance with EU law obligations. The Advocate General quite neatly clarified that: [A]s a matter of EU law, lex mitior and the requirement of legal certainty in criminal matters prevent the disapplication of the more lenient rules of national law … [but this] just carve[s] out a narrow exception for the individual ongoing criminal cases, while leaving the general normative consequences of the incompatibility untouched.

81 Article 11 of Directive 2013/40/EU (n 26). 82 The Framework Decision on ship-source pollution had included some non-custodial sentences, such as disqualification, which were however additional rather than an alternative to detention: Council Framework Decision 2003/80/JHA of 27 January 2003 on the protection of the environment through criminal law [2003] OJ L29/55. 83 Berlusconi (n 45). 84 ibid paras 76–78.

60  Why EU Criminal Law? The findings of the Court could naturally give rise to infringement proceedings for failure to comply with the obligations arising from it … and would, in any event, lead to the obligation to modify accordingly the national legal order for the future.85

Given that it is established that national legislatures cannot repeal EU law-based definitions of crimes, it can then be wondered whether the EU itself can decriminalise. As was mentioned earlier, incidental obligations to set aside national criminal law can arise as a result of EU free movement provisions.86 Moreover, the EU choice for administrative sanctions in a certain field arguably prevent Member States from introducing criminal sanctions.87 However, the Treaties do not grant the EU with explicit de-criminalisation powers. Article 83 TFEU, the relevant legal basis for the harmonisation of substantive criminal law, simply speaks of establishing minimum standards for definition of crimes and sanctions, but does not say anything about setting explicit boundaries to the range of criminal law. To date, in practice, it has never been the case that an EU text adopted on the basis of Article 83 TFEU included an explicit de-criminalisation obligation88 or de-criminalised a behaviour which had previously attracted criminal sanctions. Conversely, directives repealing previous Framework Decisions have systematically increased the repressiveness of criminal law by enlarging the scope of punishable acts, by raising the penalty thresholds or by introducing criminal sanctions where there were previously only administrative sanctions, as soon as the competence was granted in the relevant policy area. The Directives on trafficking in human beings,89 child pornography,90 attacks

85 Opinion of the Advocate General of 13 July 2017 in Case C-574/15 Criminal Proceedings against Mauro Scialdone. 86 See n 15 above. 87 Mitsilegas (n 16). 88 The closest to a decriminalisation provision is art 8 of Directive 2011/36/EU of the European Parliament and of the Council of 5 April 2011 on preventing and combating trafficking in human beings and protecting its victims and replacing Council Framework Decision 2002/629/JHA [2011] OJ L101/1). The text binds Member States to allow competent national authorities not to prosecute crimes committed by victims of trafficking in human beings. It does not per se refer to decriminalisation, but rather to non-prosecution. 89 See the broader definition of exploitation in art 2(3) of Directive 2011/36/EU (n 23) when compared with art 1(d) of the 2002/629/JHA Framework Decision on Trafficking in Human Beings [2002] OJ L203/1. The penalties have been increased in the Directive with respect to the Framework Decision. The Framework Decision included ‘effective dissuasive and proportionate sanctions’ for the criminalised trafficking offences and eight years’ imprisonment for the more serious cases (see art 3(1) of the Framework Decision). The Directive conversely lists five years’ for the criminalised offences and 10 years’ imprisonment in the most serious cases (art 4). 90 Directive 2011/92/EU (n 28) introduces the crimes of sexual abuse of children (art 3) and solicitation of children for sexual purposes (art 6), which were not present in the pre-existing Council Framework Decision 2004/68/JHA of 22 December 2003 on combating the sexual exploitation of children and child pornography [2004] OJ L13/44. Moreover, the levels of sanctions were increased. The Framework Decision requires Member States to punish the conduct of sexual exploitation of children and child pornography with a maximum penalty of at least one to three years (art 5 of the Framework Decision). The Directive raises that threshold to five years for sexual exploitation, and three years with no discretion for Member States to go below that (arts 4 and 5 of the Directive).

EU Relevance of the Debate on the Legitimacy of Criminal Law  61 against information systems,91 terrorism92 and protection of EU financial ­interests93 include both new crimes and in some cases higher penalties for pre-existing offences. The instruments in the field of corruption in the private sector94 and of protection of euros against counterfeiting95 contain more severe sanctions than their predecessors. Lastly, the field of money laundering and market abuse, which had traditionally been regulated by administrative law, saw the introduction of criminal sanctions immediately after Lisbon.96 Briefly, EU criminal law de facto irreversibly ‘crystallises’ a certain degree of repressiveness in national criminal justice systems.97 The established level can roughly correspond to the one already existing in Member States, but can also trigger an increase in criminal repressiveness, as in the Nordic countries, where EU law systematically increased penalty thresholds.98 91 Directive 2013/40/EU (n 29) includes the crime of ‘illegal interception’ which was not present in the previous text: Framework Decision 2005/222/JHA (n 79). The level of sanctions had been increased from ‘effective dissuasive and proportionate sanction’ for illegal access to information systems (art 2 of the Framework Decision) to at least two years’ imprisonment (art 3 of the Directive), and from one to three years’ imprisonment for illegal system interference and illegal data interference (arts 3 and 4 of the Framework Decision) to two to five years’ imprisonment for more serious cases (arts 4 and 5 of the Directive). 92 Directive 2017/541 (n 21) introduces the crimes of receiving training for terrorism (art 8) and travelling for terrorism (art 9), which were not present in the previous texts: Council Framework Decision 2002/475/JHA of 13 June 2002 on combating terrorism [2002] OJ L164/3; and Council Framework Decision 2008/919/JHA of 28 November 2008 amending Framework Decision 2002/475/ JHA on combating terrorism [2008] OJ L330/1. 93 Directive 2017/1371 (n 13) introduces the crime of misappropriation which was not present on the previous instruments: Convention drawn up on the basis of Article K.3 of the Treaty on European Union on the protection of the European Communities’ financial interests [1995] OJ C316/49; Council Act of 27 September 1996 drawing up a Protocol to the Convention on the protection of the European Communities’ financial interests [1996] OJ C313/31; Council Act of 19 June 1997 drawing up the Second Protocol of the Convention on the protection of the European Communities’ financial interests [1997] OJ C221/11. Moreover, the Directive raises the severity of punishment by setting a terms of four years’ imprisonment when the conduct causes serious damage (art 7(3) of the Directive). The previous instruments only required ‘Effective proportionate and dissuasive criminal penalties involving deprivation of liberty’ which could give rise to extradition (art 2 of the Convention; art 5 of the First Protocol; art 4 of the Second Protocol). 94 Article 5(4) of Directive 2014/62/EU (n 19) introduces a penalty of five years’ imprisonment for ‘fraudulent uttering of counterfeit currency’ and for the ‘import, export, transport, receiving or obtaining or counterfeit currency with a view to uttering of counterfeit currency’. These forms of behaviour were previously only sanctioned with ‘effective dissuasive and proportionate sanctions which might include imprisonment and can give rise to extradition’ as per art 6(1) of the C ­ ouncil Framework Decision of 29 May 2000 on increasing protection by criminal penalties and other sanctions against counterfeiting in connection with the introduction of the euro [2000] OJ L140/1. 95 Article 4 of Joint Action 98/742/JHA of 22 December 1998 concerning corruption in the private sector [1998] OJ L388/98 only envisaged ‘effective proportionate and dissuasive sanction’ and only in more serious cases a term of imprisonment which could give rise to extradition. Article 4 of Council Framework Decision 2003/568/JHA of 22 July 2003 on combating corruption in the private sector [2003] OJ L192/54 set the penalty threshold at one to three years’ imprisonment. 96 On this subject, see extensively ch 7, s V. 97 C Sotis, ‘Il Trattato di Lisbona e le competenze penali dell’Unione Europea’ (2010) 3 Cassazione Penale 326, 328. 98 See the comparison between the EU penalty threshold and the level of sanctions in Nordic countries in T Elholm, ‘Does EU Criminal Law Cooperation Always Mean Increased Repression?’ (2009) 17(3) European Journal of Crime Criminal Law and Criminal Justice 191.

62  Why EU Criminal Law? In conclusion, considering the direct impact of EU criminal law on the liberty and dignity of individuals, as well as the insistence on imprisonment, and their ‘longer-lasting’ character in national legal systems, EU criminalisation choices in this area require justification even more pressingly than national criminalisation choices. II.  WHAT DO EU CONSTITUTIONAL VALUES AND PRINCIPLES TELL US ABOUT THE LEGITIMACY OF EU CRIMINAL LAW?

Established that the EU has de facto criminalisation powers, whose exercise begs normative justification, this section considers the extent to which the EU constitutional framework provides any indication on when their use is legitimate. It looks at whether anything in the Treaties suggests whether deontological or utilitarian rationales, as were defined in Chapter 1, should guide EU criminalisation. Sections II.A and II.B look at those constitutional principles that traditionally bind public authority in the national and the EU constitutional legal order alike, namely respect of fundamental rights and liberal values and proportionality, while section II.C looks at the EU principle of subsidiarity. A.  EU Values as a Foundation for a Deontological Approach to Criminalisation Article 2 states: The Union is founded on the values of respect for human dignity, freedom, democracy, equality, the rule of law and respect for human rights, including the rights of persons belonging to minorities. These values are common to the Member States in a society in which pluralism, non-discrimination, tolerance, justice, solidarity and equality between women and men prevail.

Of relevance for the present discussion is that Article 2 includes liberal values that would also be found in modern national constitutions, notably human dignity, freedom and the right to liberty. The latter is also protected by Article 6 of the EU Charter of Fundamental Rights. As was extensively discussed in Chapter 1, authors advocating a deontological, values-based approach to criminal law have grounded their approach in the need to preserve the liberty and dignity of individuals to the highest degree possible. They consider the restriction of offenders’ dignity and possibly physical liberty via criminal law legitimate only to punish those forms of behaviour that harm liberty, or interests that are are necessary to safeguard everyone’s liberty, or interests which are as important as liberty.99



99 See

ch 1, s III.

EU Values and the Legitimacy of EU Criminal Law  63 In particular, the Italian scholarship on criminalisation grounded its version of the theory of legal interests in the Italian constitutional provisions safeguarding personal liberty in Article 13, a provision akin to Article 5 ECHR and Article 6 of the EU Charter.100 A similar argument can be (and has been) reproduced with respect to EU constitutional law and EU criminalisation. Immediately after the adoption of the Treaty of Lisbon, which gave higher visibility to fundamental rights,101 the Members of the European Criminal Policy Initiative (ECPI) published their Manifesto on a European Criminal Policy advocating a valuesbased EU criminalisation agenda,102 grounding themselves precisely in Article 2 TEU. Kaiafa-Gbandis, one of the founders of the ECPI, argued that ‘fundamental European values (article 2 TEU) constitute cornerstones of the EU and should thus be key features of EU initiatives designed to bind its Member States to adopt criminal law rules … Even within international entities, then, no legitimate approach to criminal law can ignore its dual nature as a means of protecting fundamental interests as well as acting as a yardstick for civil liberties’.103 The argument which one finds in the ECPI works can be further developed as follows. The EU Treaties did not initially include normative preferences in terms of fundamental rights protection; however, they have progressively incorporated respect of human rights among EU general principles and have elevated liberal values to EU founding principles.104 This was not a mandated choice, since not all international organisations opt for such a strong values-based core, but a political one. With such policy choices come both justiciable obligations to respect fundamental rights and also ‘policy-shaping’ obligations.105 In criminal matters, this meant embracing a deontological approach to criminalisation. This is all the more the case considering that Article 2 TEU refers to the protection of fundamental rights, which naturally already includes human dignity106 and individual liberty.107 These latter values are the traditional normative foundations of deontological criminalisation, but they also re-state dignity and freedom as EU values. Briefly, a deontological approach to criminal law is not inherent to the EU as an international organisation. Yet, the moment that the EU decided to acquire some of the constitutional features of a liberal constitutional state, a deontological approach to criminalisation came with it. Moreover, it was mentioned in 100 See the references in ch 1, s III. 101 See also s IV.B in the Introduction. 102 European Criminal Policy Initiative (n 24). 103 M Kaiafa-Gbandis, ‘Approximation of Substantive Criminal Law Provisions in the EU and Fundamental Principles of Criminal Law’ in A Weyembergh and F Galli (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Edition de l’Université de Bruxelles, 2013) 27. 104 See also s IV.B in the Introduction. 105 See also the discussion on ‘policy shaping’ obligations stemming from fundamental rights in s II.E in the Introduction. 106 Article 1 of the EU Charter on Fundamental Rights. 107 ibid art 6.

64  Why EU Criminal Law? the Introduction that the EU has attributed an identity-defining character to the respect of fundamental rights and liberal values.108 Thus, if one agrees with this argument, then a values-based liberal criminalisation policy is for the EU also a matter of coherence with its own, self-appointed identity. However, it should be noted that in making this argument, the ECPI adopted what was defined in Chapter 1 as a ‘deontological orthodoxy’. This relies on the assumption that only a purely values-based approach to criminalisation can ensure the safeguarding of the liberty and dignity of individuals. The authors of the Manifesto strongly criticised the use of harmonisation to create a level playing field for judicial cooperation, which, as will be explained in the next chapter, can be considered an expression of a utilitarian approach to criminal law. Yet, naturally, the observations made in Chapter 1 on the advantages of a joint deontological and utilitarian approach for a selective resort to criminal law apply here. Grounding the choice in favour of EU criminal law both in values and in utilitarian effectiveness considerations would lead to the more protective approach towards the liberty and dignity of individuals.109 It thus seems fair to conclude, in agreement with the authors of the manifesto, that the EU constitutional acknowledgement of liberal values represents a normative premise for the adoption of a deontological approach to criminalisation, and, a fortiori, of a joint deontological and utilitarian approach. This does not mean that there is a legally justiciable obligation to adopt an integrated approach to criminalisation that stems from Article 2 TEU. Yet, EU criminal law still needs to be coherent with the EU normative core, so it must rest on both deontological and utilitarian grounds. This is to be contrasted with the obligation, of a legal nature in this case, stemming from the principle of subsidiarity to include include utilitarian considerations when assessing the necessity to adopt EU definitions of crimes. Before turning to this, section II.B discusses one of the traditional constitutional axioms of liberal constitutional orders and of the European legal order: proportionality. B.  EU Proportionality as a Neutral Framework to Incorporate Deontological and Utilitarian Considerations The principle of proportionality has a long history deriving from German administrative law and is a key principle in liberal constitutional orders, aiming to limit public authorities’ intrusiveness in relation to the liberty of individuals.110 It is a multi-form concept with various meanings and various

108 See s IV.B in the Introduction. 109 See ch 5, s II. 110 For a historical overview, see E Engle, ‘The History of the General Principle of Proportionality: An Overview’ (2012) 10 Dartmouth Law Journal 1.

EU Values and the Legitimacy of EU Criminal Law  65 applications in different contexts. Broadly speaking, carrying out a proportionality testing implies balancing two elements against each another. Depending on which elements are included in the balancing exercise, we can distinguish between prospective and retrospective proportionality.111 The EU legal order acknowledges both kinds of proportionality112 and they are both relevant to EU criminal law. However, none of these contains a precise indication as to when the use of the EU definitions of crimes is legitimate. Prospective proportionality is conceived within the EU legal order as a neutral criterion in relation to the approach to EU criminalisation. Retrospective proportionality, as spelled out by the EU Charter of Fundamental Rights, postulates a deontological approach, but only with respect of criminal sanctions, not of definition of crimes. i.  Prospective Proportionality as a Neutral Criterion for EU Criminal Law Prospective proportionality applies to any public policy decision and requires a balance between the means, namely the specific regulatory instrument selected (eg, hard law v soft law) and the end pursued, the desired policy objective. The means need to be proportionate to the desired aim. More concretely, prospective proportionality requires the balancing of those interests, values or objectives that the selected mean jeopardises in order to achieve the desired goal against the values that the desired measure protects or the objective it is trying to achieve. The test is prospective because it is forward-looking, since the benchmark for proportionality is the achievement of a policy goal that is yet to happen. Scholarship normally outlines four steps for the proportionality test: ‘legitimate purpose’ (the measure must pursue a legitimate aim which justifies the restriction of a specific interest); ‘suitability’ (the measure would actually lead to the desired aim); ‘necessity’ (there are no other, less intrusive measures available which would reach the same end, but with a smaller sacrifice of the interests involved); and ‘proportionality stricto sensu’ (there must be a balance between the aims pursued, the legitimate purpose and the interests which would be sacrificed by the selected means – if there is not, even if there are not any less intrusive means to achieve the given objective, then arguably the relevant measure should not be taken at all).113 This quadripartite structure is well acknowledged in the literature. However, Tridimas observes how courts in general tend to focus

111 For an introduction to the distinction between prospective and retrospective proportionality, see Asp (n 32) 188–90. See also E Herlin-Karnell, The Constitutional Dimension of EU Criminal Law (Oxford, Hart Publishing, 2012) 127 and 128 and the references given therein. 112 See, ceteris paribus, G de Búrca, ‘The Principle of Proportionality and its Application in EC Law’ (1993) 13(1) Yearbook of European Law 105; E Ellis (ed), The Principle of Proportionality in the Laws of Europe (Oxford, Hart Publishing, 1999). 113 See N Emiliou, The Principle of Proportionality in European Law: A Comparative Study (Dordrecht, Kluwer, 1996) 191, K Lenaerts and P van Nuffel, Constitutional Law of the European Union, 2nd edn (London, Sweet & Maxwell, 2005) 5-3 ff.

66  Why EU Criminal Law? merely on the suitability and necessity test, and in particular do not investigate the proportionality stricto sensu test.114 Since the prospective proportionality applies to all public policy decisions, it also applies to criminalisation decisions. Furthermore, as was mentioned in the Chapter 1, criminal law theorists have developed a more elaborate, criminal law-specific form of proportionality, namely the ultima ratio principle. This principle establishes that criminal law should always be the last resort.115 If one agrees that proportionality is a balance between means and ends, ultima ratio provides insights on the first element of the balance test, namely the end. It reiterates that criminal law sacrifices very important interests and as a consequence it is by definition the most intrusive means.116 It thus invites a very careful selection of the relevant ‘end’ for this ‘means’. Neither prospective proportionality nor ultima ratio provides any indications as to the justification for criminal law; they only require a balance between the legitimate aim of criminal law (protecting values or pursuing forwardlooking objectives) and the values or interests affected by criminal law, namely the liberty and dignity of individuals. Persak describes ultima ratio, but the same applies to proportionality, as a negative formal principle cautioning against criminalisation.117 Conversely, she qualifies the principle of harm, offence, legal moralism or paternalism, but we could say the same for the legal interest theory, as positive substantial principles providing indications on when to use criminal law.118 Persak underlines how the two tests are interconnected. They are two sides of a joint formal and substantive judgement on the legitimate use of criminal law. Yet, when faced with questions on the legality of substantive criminal law norms containing the definition of crimes, national courts have often relied on proportionality and ultima ratio119 as formal criteria. They have not engaged

114 T Tridimas, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press, 2006) 139. For the doctrine of proportionality in the case law of the Court of Justice, see Case C-331/88 R v Minister of Agriculture, Fisheries and Food ex parte Fedesa [1990] ECR I-4023, para 13. 115 On ultima ratio, see, among others, N Jarborg ‘Criminalisation as a Last Resort (Ultima Ratio)’ (2005) 2 Ohio State Journal of Criminal Law 53; D Husak, ‘The Criminal Law as a Last Resort’ (2004) 24(2) Oxford Journal of Legal Studies 207; P Mikkinen, ‘“If Taken in Earnest”: Criminal Law Doctrine and the Last Resort’ (2006) 45(5) Howard Journal 521; J Bengoetxea, H Jung and K Nuotio, ‘Ultima Ratio, a Principle at Risk: European Perspectives’ (2013) 3(1) Onati Socio-legal Series, http://opo.iisj.net/index.php/osls/issue/archive. 116 On a comparison between ultima ratio and proportionality, see Herlin-Karnell (n 111) 127; R Wendt, ‘The Principle of “Ultima Ratio” and/or the Principle of Proportionality’ (2013) 3(1) Onati Socio-legal Series 81, http://opo.iisj.net/index.php/osls/issue/archive. 117 N Persak, The Harm Principle and its Continental Counterpart (Dordrecht, Springer, 2012) 22. 118 ibid. 119 For a survey of the German Constitutional Court case law on the matter, see Wendt (n 116). For an example of proportionality assessment in a criminalisation case within the Strasbourg Court, see Reichman v France App No 50147/11 (ECtHR, 12 July 2016) 57 ff. See also Lietuvos Respubliskos Konstitucinis Tesimas (Constitutional Court of the Repulblic of Lithuania), 1997-11-13 nutarimas, bylos Nr 4/97, pt IV.1; Cour Constitutionnelle de Belgique, affaire 145/2012, pt B.18, BverfGE 88, 203 (258); Corte Costituzionale Italiana (Italian Constitutional Court) sentenza n 1085, del 13 Dicembre 1988.

EU Values and the Legitimacy of EU Criminal Law  67 with any discussion on positive criminalisation criteria indicating the functions for criminal law. This approach allowed them to avoid principled rulings on the general function of criminal law, which they considered politically sensitive. The EU legal order includes two provisions requiring EU action in order to respect prospective proportionality.120 First, Article 5(4) TFEU states: ‘Under the principle of proportionality, the content and form of Union action shall not exceed what is necessary to achieve the objectives of the Treaties.’ This sort of proportionality naturally applies to all EU policies, including the EU criminalisation policy. However, the formulation is very general. There is doctrinal debate as to whether Article 5(4) TEU is intended to be a safeguard against excessive encroachment in private autonomy, in its traditional sense as a liberal ­principle,121 or if it is a safeguard against excessive national regulatory autonomy in a more federal sense.122 The Court has mainly used Article 5(4) TFEU proportionality to assess the acceptable impact of EU measures harmonising internal market regulations on economic operators and thus in a liberal sense.123 So far, there has not been any case involving EU criminalisation decisions decided on the basis of Article 5(4) TFEU proportionality. Second, Article 52(3) of the EU Charter of Fundamental Rights states that: Any limitation on the exercise of the rights and freedoms recognised by this Charter must be provided for by law and respect the essence of those rights and freedoms. Subject to the principle of proportionality, limitations may be made only if they are necessary and genuinely meet objectives of general interest recognised by the Union or the need to protect the rights and freedoms of others. (Emphasis added)

This kind of fundamental rights (or liberal) proportionality protecting private autonomy is bound to be more relevant to the criminal law field. As was discussed in Chapter 1, criminal law measures are liable to restrict the fundamental right to liberty and dignity, and any other right relevant to a specific definition of crime.124 They are thus liable to fall within the measures covered in Article 53(2) of the EU Charter. 120 EU law also requires a prospective proportionality test on national measures respectively restricting free movement and fundamental rights. For a distinction between the different kinds of proportionality within the EU, see M Kumm, ‘Constitutionalising Subsidiarity in Integrated Markets: The Case of Tobacco Regulation in the European Union’ (2006) 4(12) European Law Journal 503, 523. 121 R Schütze, From Dual to Cooperative Federalism (Oxford, Oxford University Press, 2009) 263–64. 122 Davies criticises the suitability of subsidiarity as a principle to protect national regulatory autonomy and maintains that proportionality should be entrusted with the task: G Davies, ‘Subsidiarity: The Wrong Idea, in the Wrong Place, at the Wrong Time’ (2006) 43(1) Common Market Law Review 63; similarly, Kumm (n 120) admits a federal dimension for proportionality. 123 See by way of example Case C-491/01 The Queen v Secretary of State for Health ex parte British American Tobacco (Investments) Ltd and Imperial Tobacco Ltd [2002] ECR I-11453; Case C-58/08 The Queen on the Application of Vodafone Ltd and Others v Secretary of State for Business, Enterprise and Regulatory Reform [2010] ECR I-04999; Case C-176/09 Grand Duchy of Luxembourg v European Parliament and Council of the European Union [2011] ECR I-03727. 124 See ch 1, ss I and II.

68  Why EU Criminal Law? Article 53(1) of the EU Charter is more detailed than Article 5(4) TFEU, providing a more elaborated version of proportionality that is closer to the Persak joint test mentioned above. The provision arguably admits both utilitarian and deontological approaches to criminalisation. Indeed, it states that the limitation of Charter rights can both pursue ‘objectives of general interest’ and ‘protect the rights and freedom of others’. Thus, the application of this limitation by means of criminal law may see the former including the utilitarian goal of law enforcement and the latter resonating with a deontological use of criminal law to protect important values. Yet, the provision presents them as two equally valid alternatives. As such, it does not provide any clear indication on the most desirable approach for criminalisation. So far, the ‘fundamental rights proportionality’ of the EU criminalisation decisions has not been questioned before the Court.125 Yet, in Digital Rights Ireland, the Court had the chance to rule on the proportionality of data retention rules designed with the aim of crime prevention. When carrying out the proportionality test, the Court acknowledged security as a legitimate objective for restricting fundamental rights, without giving any indication on the further objectives of such criminal justice measures.126 ii.  Retrospective Proportionality: A Deontological Approach to EU Criminal Sanctions Retrospective proportionality applies to the implementation of sanctions and thus has a more restricted scope of application than prospective proportionality. It concerns the balance between the harshness of the sanction and the seriousness of the behaviour which is being sanctioned. The ‘seriousness’ of the behaviour depends on the harm or damage caused and the degree of guilt of the transgressor.127 The test is retrospective because the benchmark for proportionality is based on facts that have already occurred. As a matter of principle, retrospective proportionality can apply both at a legislative level, when prescriptive norms are designed and sanctions for their breach must be selected, and at an adjudication level, when judges and sanctioning bodies, must choose which penalties to impose in concrete cases. With reference to the legislative level, the level relevant to the discussion in this chapter, retrospective proportionality can first guide the decision on whether to criminalise a certain form of behaviour. In this case, it requires the imposition 125 This is not to say that the Court has not itself pronounced on the point of the function of ­criminal law; it has, but in other contexts, namely in legal basis disputes. On this see extensively ch 5. 126 Cases 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 [2012] ECR General Court Reports. 127 Herlin-Karnell (n 111) 127. The idea of the seriousness of the crime as depending on harm and culpability can be found in A Ashworth and L Zedner, ‘Prevention and Criminalisation: Justification and Limits’ (2012) 15(4) New Criminal Law Review 542, 544.

EU Values and the Legitimacy of EU Criminal Law  69 of criminal sanctions for the most serious forms of behaviour. Second, it can guide the decision on the kind of criminal sanction, requiring a scaled approach depending on the seriousness of each criminal offence. What is interesting for the purposes of the current discussion is that in both cases, the approach is deontological. Retrospective proportionality requires looking at the harm caused by the behaviour at stake. As Feinberg theorised, the harm caused by a certain conduct depends on the importance of the interests it affects.128 The EU legal order acknowledges retrospective proportionality, but only with respect to the choice among several criminal sanctions. Article 49(3) of the EU Charter of Fundamental Rights states that: ‘The severity of penalties must not be disproportionate to the criminal offence.’ The charter is thus concerned with the choice among possible sanctions for behaviours that the legislator has already decided should be criminalised. Article 49(3) of the EU Charter thus postulates a deontological approach to criminal punishment. Conversely, it does not provide indications on the normative criteria for criminalisation. C.  Subsidiarity: A Utilitarian Approach to EU Criminalisation The last principle that is relevant to the analysis of the EU constitutional framework binding EU action in criminal matters is subsidiarity, which is established in Article 5(2) TEU. This principle indicates that EU action, as opposed to national regulation, is justified when it is comparatively more efficient. The principle is thus concerned with the federal dimension of power, that is, with the protection of national regulatory autonomy from centralised intervention. Admittedly, the early discussions on subsidiarity in the EU legal order also identified a liberal dimension for the principle as a ‘dividing line between the private sphere and that of the State, in the broad meaning of the term’.129 However, no liberal dimension has been incorporated into the structure and the discourse on Article 5(2) TEU subsidiarity.130 At first sight, it would seem that the principle does not have much relevance to the debate on the legitimacy of criminalisation, which concerns the liberal dimension of power, namely the protection of private autonomy and individuals’ liberty in particular, from state intervention.131 Nonetheless, such a focus on efficiency has indirect implications for the debate on the legitimacy of EU criminal law, as it favours a utilitarian approach to

128 J Feinberg, Harm to Others: The Moral Limits of the Criminal Law (Oxford, Oxford University Press, 1984) 36. 129 J Delors, ‘The Principle of Subsidiarity: Contribution to the Debate’ in European Institute of Public Administration, Subsidiarity: The Challenge of Change. Proceedings of the Jacques Delors Colloquium (Maastricht, EIPA, 1991) 7, cited in Schütze (n 121) 246. 130 This dimension can still be found in the Preamble to the TEU and in art 1(2) TEU, which states that a ‘decision should be taken as closely as possible to the citizens in accordance with the principle of subsidiarity’. 131 For a distinction between the two liberal and federal dimensions, see Schütze (n 121) 246.

70  Why EU Criminal Law? criminalisation. Sections II.C.i and II.C.ii respectively focus on the efficiencybased nature of the principle and its connection with utilitarian criminalisation. i.  The Choice in Favour of an Efficiency-Based Criterion to Justify EU Action Article 5(3) TEU states that: In areas which do not fall within its exclusive competence, the Community shall take action, in accordance with the principle of subsidiarity, only if and in so far as the objectives of the proposed action cannot be sufficiently achieved by the Member States and can therefore, by reason of the scale or effects of the proposed action, be better achieved by the Community.

The principle was codified in the Treaty of Maastricht, even though evidence of a ‘subsidiarity-inspired’ approach can be found before Maastricht.132 Historically, it was introduced as a safeguard for Member States that had seen their role within the EU decision-making process decline due to the expansion of the qualified majority voting rule.133 The Treaty formulation of the principle is neutral. The provision does not specify when Member States cannot sufficiently achieve a certain policy objective and when it is considered that the EU can do ‘better’. In its 1992 Communication immediately after the introduction of the Treaty, the Commission clarified that the principle had to be understood as a requiring a ‘comparative efficiency’ test.134 More specifically, the EU is meant to act when Member States are not capable of sufficiently tackling a certain phenomenon (the national insufficiency test) and the EU is comparatively more efficient (the actual comparative efficiency test). EU action can be defined as more efficient than national law when it is effective in achieving a certain policy objective and it is more cost-effective than national action. Kumm speaks in this context of a ‘collective action problem’.135 132 For instance, the Spinelli draft Treaty on the EU of 1984 mentioned subsidiarity; see the last recital of the draft Treaty on the EU, stating that the intention of the Treaty was to ‘entrust common institutions, in accordance with the principle of subsidiarity, only with the powers required to complete successfully the tasks they may carry out more satisfactorily than the States acting independently’. The text is available at: http://www.cvce.eu/en/obj/ draft_treaty_establishing_the_european_union_14_february_1984-en-0c1f92e8-db44-4408-b569c464cc1e73c9.html. A subsidiarity ‘logic’ was also incorporated into the Single European Act; see art 130r(4) of the Treaty on European Community as amended by art 25 of the European Single Act [1987] OJ L169/11. The article relates to the field of environmental law and states that: ‘The Community shall take action relating to the environment to the extent to which the objectives referred in paragraph 1 can be attained better at Community level than at the level of the individual Member State.’ See also the speech in Delors (n 129). 133 GA Bermann, ‘Taking Subsidiarity Seriously’ (1994) 94 Columbia Law Review 332, 354–65; P Craig, ‘Subsidiarity: A Political and Legal Analysis’ (2012) 50(1) Journal of Common Market Studies 72, 73. 134 The Principle of Subsidiarity. Communication of the Commission to the Council and European Parliament, SEC (92) 1990 final, 2. 135 Kumm (n 120) 520.

EU Values and the Legitimacy of EU Criminal Law  71 A collective action problem arises when a certain policy objective would benefit each Member State, but the costs of achieving such an objective for each Member State acting alone would outweigh the benefits. If acting collectively and sharing the cost would strike a better balance, then the EU is justified in acting.136 According to this interpretation, substantive judgments on the wisdom of each Member States’ policies, which the EU and the other Member States consider inadequate, should not be a ground for further EU action. Nor should it be the goal of fostering integration as such.137 The Commission itself also clarifies that the principle of subsidiarity ‘has nothing to do with a “democratic deficit” that has to be made good: it should not be confused with democratic control of Community Action’.138 Thus, it appears that an efficiency-based criterion for allocating powers across levels of government was a deliberate choice rather than a mandated one. For instance, Kumm also observes that the reasons for favouring local action, as opposed to centralised action, can be constructed using the language of democracy, identity and efficiency.139 Similarly, de Búrca lists ‘process’, ‘willingness’ and ‘outcome’ as possible criteria for the allocation of powers.140 In other words, when choosing which level to act upon, one might look at the decision-making procedures of each government level and opt for the most democratic. Alternatively, one might consider the interests that a certain policy should protect. Depending on whether they are closer to the identity of the local or the centralised level of government, this might make either level more willing to address the problem. Finally, one might consider the level at which a certain policy objective is more performant or more efficient. Out of these three criteria, the Commission communication clearly opted for a performance-based approach. For the sake of completeness, it should be mentioned that the Council also provided some further guidelines on the interpretation of subsidiarity which were later incorporated into a Protocol attached to the Treaty of Amsterdam.141 Yet, but for a presumption in favour of EU action when the relevant policy area contains transnational aspects,142 these guidelines are, for the most part, unhelpful.143 Authors have contested subsidiarity’s conceptual lack of suitability to effectively safeguard Member States from excessive centralisation, suggesting that 136 ibid. 137 ibid. 138 Better Lawmaking 1998: A Shared Responsibility. Commission Report to the Council, COM (98) 715 final, 3. 139 Kumm (n 120) 120, 518. 140 G de Búrca, ‘Reapprasing Subsidiarity’s significance after Amsterdam’ (1999) 7/99 Harvard Jean Monnet Working Paper 4, www.jeanmonnetprogram.org/archive/papers/99/990701.rtf. 141 Protocol annexed to the Treaty of the European Community (consolidated in Amsterdam) – Protocol on the application of the principles of subsidiarity and proportionality [1997] OJ C340/105. (hereinafter ‘Amsterdam Protocol’). 142 ibid art 5. 143 See the observations in de Búrca (n 140) 24–25.

72  Why EU Criminal Law? proportionality would do a better job.144 Moreover, the case law of the Court has not always been very consistent and articulated in its application of the principle.145 Yet, the substance and especially the efficiency-based nature of the principle have not been contested at an institutional and policy level. Only procedural amendments were introduced to ensure a better enforcement of the principle by involving national parliaments.146 This book therefore relies on the understanding of subsidiarity as an efficiency-based principle. ii.  More Efficient EU Criminal Law as Utilitarian Criminal Law When applied to the criminal law field, subsidiarity first requires measuring the comparative efficiency of the national and supranational levels in punishing and preventing human behaviours that are judged to be undesirable. For instance, under Article 83 TFEU, the EU has the competence to harmonise the definition of crimes in the field of drug trafficking. Let us assume that the Commission wanted to propose legislation on this legal basis requiring Member States to criminalise and set high imprisonment sanctions for the possession and consumption of soft drugs (eg, cannabis) with the aim of addressing the demand side of drug trafficking. When assessing the compliance of the proposed text with subsidiarity,147 one would have to rely on the bipartite test illustrated above. Starting with the national insufficiency limb of the test, it would be necessary to verify whether single Member States acting alone are able to tackle the problem of curbing the phenomenon of the consumption of soft drugs, bearing in mind that some might resort to criminal law sanctions while others might prefer administrative law sanctions. Then, it would be necessary to carry out the ‘comparative efficiency’ limb of the test. This part of the test first requires an assessment of the effectiveness of EU action in addressing the problem. Second, one would have to compare a scenario in which all Member States have similar definitions of crimes with a scenario in which every Member State regulates the

144 Among the most authoritative critics of the principle, see A Estella, The Principle of Subsidiarity and its Critique (Oxford, Oxford University Press, 2002). See also Davies, for instance, who argues that the principle is structurally unfit to do so: Davies (n 122); Kumm (n 120) 80. 145 Schütze argues that the Court has in several instances ‘short-circuited’ the national insufficiency and the comparative efficiency test, considering the second automatically to be met when the first is met: Schütze (n 121). On the case law of the Court more broadly, see T Horsley, ‘Subsidiarity and the European Court of Justice: Missing Pieces in the Subsidiarity Jigsaw’ (2012) 50(2) Journal of Common Market Studies 267; Craig (n 133); and G de Búrca, ‘The Principle of Subsidiarity and the Court of Justice as an Institutional Actor’ (1998) 36(2) Journal of Common Market Studies 217. 146 See, among others, I Cooper, ‘The Watchdogs of Subsidiarity: National Parliaments and the Logic of Arguing in the EU’ (2006) 44(2) Journal of Common Market Studies 281; J Louis-Victor, ‘National Parliaments and the Principle of Subsidiarity: Legal Options and Practical Limits (2008) 4(3) European Constitutional Law Review 429. 147 On the need to look at subsidiarity in concreto, looking at the specific instrument, and not in abstracto in a certain policy area, see Schütze (n 121) 263, 265 and the case law cited therein.

EU Values and the Legitimacy of EU Criminal Law  73 matter autonomously either through criminal law or through other means, such as administrative law. In order for EU action to be justified, the first scenario must be more efficient in addressing the problem. It flows from this that the EU definition of crimes only meets the subsidiarity test if criminal law is an effective tool to prevent behaviours that society deems to be undesirable. If this is not the case, then harmonised criminal law would certainly not be efficient, let alone comparatively more efficient than action at a national level. In the example of an EU directive requiring harmonisation of the definition of crimes in relation to soft drugs, it would be necessary to assess whether criminal law and harsh imprisonment sentences are effective in reducing the consumption of drugs. In fact, drug policies are quite diverse, where a public health approach is often in opposition to a more criminal justice-focused model.148 The effectiveness of harsh criminal punishments in this policy area is thus in some instances contested.149 In order to meet the subsidiarity test, the Commission would first have to justify its decision on this point. Second, it would of course have to demonstrate why all Member States having similar definitions of the crime of possession and consumption of soft drugs would help tackle the problem more efficiently. One argument in favour of EU h ­ armonised definitions of crime being more efficient than national law in achieving the EU objective of fighting crime is that it would create a level playing field enabling easier mutual recognition and judicial cooperation. This ‘cooperation rationale’ for EU criminal law is extensively explained in the next chapter.150 The broader point here is that by grounding the legality of EU action in its capacity to be more efficient than national action, the principle of subsidiarity requires the inclusion of effectiveness considerations when deciding whether or not to resort to a harmonised definition of crimes in a given policy area. To use the criminal legal terminology referred to in Chapter 1, subsidiarity imposes a constitutional law obligation to include a utilitarian justification for the use of criminal law. A purely symbolic EU criminal law targeting behaviours affecting values that the EU considers to be particularly important,151 but that is not effective in discouraging them would not be compliant with subsidiarity. The argument that EU law is more efficient in solving a certain problem would indeed be hard to make. This argument could easily be rebutted by stating that Member States were not inefficient in addressing a certain problem; they simply had different

148 On various approaches to drug policy, see H Bergeron and R Colson (eds), European Drug ­Policies: The Ways of Reform (Abingdon, Routledge, 2016). 149 Elhom and Colson have in fact argued that the EU should adopt legislation imposing a more lenient approach in the field of drug consumption on Member States; see T Elholm and R Colson, ‘The Symbolic Purpose of EU Criminal Law’ in R Colson and S Field (eds), EU Criminal Justice and the Challenges of Diversity: Legal Cultures in the Area of Freedom, Security and Justice (Cambridge, Cambridge University Press, 2016) 48. 150 See ch 3, s V.A. 151 See ch 1, s V.B for a broader discussion on the term ‘symbolic criminal law’.

74  Why EU Criminal Law? normative preferences on what values are so important as to deserve symbolic criminalisation.152 Conversely, subsidiarity as such does not require any deontological judgement on the interests that the policy in question protects. Effectively sanctioning behaviours affecting low-ranked interests or values in EU criminal law would still be legitimate, and lawful for that matter, under the principle of subsidiarity if Member States acting alone were comparatively less efficient. In other terms, regulatory EU criminal law153 would meet the subsidiarity test. III. CONCLUSIONS

This chapter has addressed the core theme of this book, namely the legitimacy of EU criminal law. Whether such a fundamental criminal legal theory debate is also relevant to the EU legal order is not self-evident. The EU is not a sovereign state, but has a limited range of action, and the exercise of its competence is further constrained by EU constitutional principles. In particular, it only has indirect criminalisation competence, eg, it can only set minimum standards in terms of definitions of crimes. Yet, the EU definitions of crimes are often very detailed, and Member States have very little discretion in terms of implementing them in their national legal systems. Moreover, once the EU has defined a certain form of behaviour as a crime, Member States cannot decriminalise such behaviour, as that would run counter to their EU law obligations. It is not clear whether the EU itself has the competence to repeal existing definitions of crimes with more lenient ones in subsequent legislative instruments. In practice, the EU has systematically increased the level of repression of its criminal law when repealing previous instruments, almost always requiring imprisonment sanctions. Briefly, EU criminalisation choices can have a direct impact on the liberty and dignity of individuals and it is therefore not only relevant but also important to reflect on the theoretical justifications for such policy choices. The constitutional law framework binding the exercise of such EU criminalisation powers provides substantial indications on when resorting to EU criminal law is legitimate. With the Treaty of Maastricht in 1992, the treaty-drafters made two policy choices for the EU constitutional framework. First, respect for fundamental rights was acknowledged as a founding principle for the EU, which would later be considered an EU value. With this policy choice, namely giving a specific normative core to the EU as an international organisation, came the obligation to have a values-inspired, deontological criminalisation policy. Second, in the Treaty of Maastricht, the principle of subsidiarity was 152 This point is more widely addressed in the discussion on the Framework Decision on Racism and Xenophobia in ch 7. 153 See ch 1, s V.B for a broader discussion on the terms ‘symbolic criminal law’ and ‘regulatory criminal law’.

Conclusions  75 also introduced. The choice of the EU to rely on an efficiency-based criterion to justify EU action, as opposed to national action, also necessitates the inclusion of utilitarian considerations in criminalisation decisions. Chapter 1 illustrated various approaches to criminalisation, distinguishing between a deontological, values-based approach, and utilitarian, effectivenessbased approach. It illustrated how there is a strong emphasis in traditional criminal legal theory literature in favour of justifying resorting to criminal law on the basis of the values it protects – what some authors call a ‘deontological’ orthodoxy. Conversely, effectiveness as a legitimating factor for criminal law has less doctrinal support. Nevertheless, this chapter has advocated an integrated deontological and utilitarian approach, which was considered both theoretically and empirically desirable. It has demonstrated that a ‘deontological orthodoxy’ is not even ‘constitutionally affordable’ at the EU level. Symbolic EU criminal law could be struck down for failing to meet the subsidiarity test, whereas regulatory EU criminal law (that is, effective criminal law used to address forms of behaviour harming low-ranked interests) might meet the test and thus would not be unlawful as such. However, it would be inconsistent with the EU acknowledgement of its founding liberal values. An integrated approach to criminal law is thus also desirable for the EU level and has possibly a higher constitutional stronghold than it does at the national level.

3 Rationales for the Harmonisation and Legitimacy of EU Criminal Law

C

hapter 2 clarified that EU harmonisation of national criminal law has an impact on the liberty and dignity of individuals and therefore is in need of justification. Still, the harmonisation of national criminal norms and criminalisation remain two different processes, whose rationales, and thus theoretical justifications, might or might not overlap. This chapter addresses the relationship between these two processes. In particular, it will look into what the traditional rationales for the harmonisation of definitions of crimes are and which of them implies a deontological or a utilitarian understanding of EU criminal law. The aim is to develop a theoretical framework for the analysis of EU primary and secondary law and EU policy documents in the subsequent chapters. Each of these sources speaks of the functions of harmonisation and only occasionally makes reference to the function of criminal law. Such references to harmonisation rationales will then be interpreted in light of what is discussed here. Section I first defines the concept of harmonisation in general and as a legal process. Section II clarifies that harmonisation is per se a neutral legal process without an inherent normative dimension. It can be used to promote a valuesbased agenda. Yet, it can also only pursue the attenuation of legal differences as a means to a further end goal. Section III discusses the harmonisation of national criminal law as a strategy of legal integration. Section IV distinguishes between values-based and instrumental rationales for harmonisation of criminal law. Sections IV–VII discuss if each of these imply a deontological or a utilitarian legitimacy for EU criminal law. Finally, section VIII provides some concluding remarks. When discussing the various rationales for harmonisation, some examples are included that are taken either from existing EU legislation or simply hypothetical scenarios. More detailed case studies on EU legislation are discussed in Chapter 7. I.  A DEFINITION OF HARMONISATION AS A LEGAL PROCESS

The concept of harmony and of harmonisation can be found in Greek ­mythology, which acknowledged a deity named Harmonia, daughter of Ares, the god of

A Definition of Harmonisation as a Legal Process  77 war, and Aphrodite, the goddess of beauty.1 In modern usage, when used in natural language, harmonisation can be defined as a ‘process in which diverse elements are combined or adapted to each other so as to form a coherent whole, while retaining their individuality’.2 Harmony, as a result of harmonisation, thus postulates the presence of different elements that nonetheless peacefully co-exist in a whole. In music, for instance, harmony is intended as the aesthetically pleasant simultaneous sounding of different tones.3 In this context, harmony is opposed to uniformity. If there were only identical tones, there would only be a monotone, or silence if there were no tones at all.4 However, in the legal context, harmonisation has a somewhat different meaning from the one used in natural language.5 It is the process that aims at reducing differences between existing diverging legislation by proposing a new standard to which the various pre-existing norms should align themselves.6 In this context, differences are seen as problematic. Harmonisation attempts not only to juxtapose the various different components so as to create a harmonious ensemble, but also to reduce the difference between them. Harmonisation as a legal process is thus not opposed to unification, but is complementary to it, or even a pre-condition to it.7 The relationship between the two is better understood if one conceptualises harmonisation as a stage within a dynamic process.8 At the two ends of the process, there are complete inhomogeneity (all differences remains) and uniformity (all differences are eliminated). Harmonisation 1 F Tadic, ‘How Harmonious Can Harmonization Be? A Theoretical Approach towards Harmonization of (Criminal) Law’ in A Klip and H van der Wilt (eds), Harmonisation and Harmonising Measures in Criminal Law (Amsterdam, Koninklijke Nederlandse Akad van Wetenschappen, 2002) 1; and U Nelles, ‘Definition of Harmonization’ in A Klip and H van der Wilt (eds), ­Harmonisation and Harmonising Measures in Criminal Law (Amsterdam, Koninklijke Nederlandse Akad van Wetenschappen, 2002) 31, 32–33. 2 M Boodman, ‘The Myth of Harmonisation of Laws’ (1991) 39 American Journal of Comparative Law 699, 702. 3 In the Oxford Dictionary, ‘Harmony’ is defined as: ‘The combination of simultaneously sounded musical notes to produce a pleasing effect.’ See www.oxforddictionaries.com/definition/ english/harmony?q=Harmony. 4 Boodman (n 2) 700. 5 Tadic (n 1) 8–9; and Nelles (n 1) 34–38. 6 This definition is from Delmas-Marty, who sketches a distinction between unification (absence of differences), cooperation (full preservation of differences) and harmonisation (elimination of some differences): M Delmas-Marty, ‘Le processus d’interaction’ in M Delmas-Marty, M Pieth and U Sieber (eds), Les Chemins de l’harmonisation pénale (Paris, Société de Legislation Comparé, 2008) 419, 420. It should be noted that this definition of harmonisation of legislations only refers to the process of substituting different norms with a single norm; it does not take into account the complementary process of harmonisation of institutions aimed at securing the implementation and the enforcement of such uniform rules. On this point, see M Delmas-Marty, ‘Introduction’ in M Delmas-Marty, M Pieth and U Sieber (eds), Les Chemins de l’harmonisation pénale (Paris, Société de Legislation Comparé, 2008) 27 ff; and A Klip, ‘Definitions of Harmonisation’ in Klip and van der Wilt (n 1) 23. 7 M Delmas-Marty, ‘The European Union and Penal Law (1998) 4(1) European Law Journal 87, 106. 8 The idea of harmonisation as a process is well explained in M Delmas-Marty, ‘Introduction’ in Delmas-Marty et al (n 6) 20.

78  Rationales for the Harmonisation and Legitimacy of EU Criminal Law situates itself in between these two ends, reducing the differences without eliminating them. Depending on the degree to which differences can be tolerated, harmonisation can lead to minimum, partial or full unification.9 If pre-existing legislation is only partially substituted with the new harmonised standard, and if some discretion is left to the various actors that have to implement the new rule, we have minimum or partial harmonisation. In this case, we would be closer to the end of the process where there is inhomogeneity – that is, where all differences remain. If the newly introduced rule covers most of the areas regulated by pre-existing legislation and there is very little discretion in implementing the standard rule, then we have maximum harmonisation. In this case, we are closer to the uniformity part of the process, namely where all differences are eliminated. II.  DOES HARMONISATION HAVE AN INHERENT, VALUES-BASED DIMENSION?

Harmonisation of legislation is a neutral process which does not have an inherent normative dimension. It can promote a normative agenda or set of values, but it can also be pursued as an end in itself that is instrumental to further policy objectives (see section II.A). This distinction between normative and instrumental harmonisation is key to understanding when harmonisation of criminal law implies a deontological or a utilitarian understanding of criminal law (see section II.B). A.  Normative v Instrumental Types of Harmonisation Harmonisation can be used as a tool to promote a certain set of values enshrined in the new harmonised standard. In this case, the reduction of differences is justified in light of the ‘higher quality’ of the imposed harmonised standard. Throughout this chapter, we will speak in this respect of ‘values-based harmonisation’. However, the reduction of differences can also be justified in light of an end goal to which creating a legal level playing field is instrumental, regardless of the substance of the imposed harmonised standard. In this context, harmonisation does not have an inherent normative dimension. The values protected by the relevant norm are not particularly important, as any harmonised rule would result in the achievement of the goal. We will speak in this case of ‘instrumental harmonisation’.

9 Tadic observes that once it is established what objectives one wants to pursue through ­harmonisation, this determines the degree of harmonisation requested, following a proportionality logic: Tadic (n 1) 11.

Does Harmonisation have an Inherent, Values-Based Dimension?  79 To give but a few examples, a values-based justification is for instance behind colonial powers imposing their own form of legislation on colonised countries based on the assumption that the new model was a better one.10 A second example of values-based harmonisation is the adoption of human rights international treaties. Instruments such as the European Convention on Human Rights have a harmonisation impact on national law to the extent that each Member State must amend its legislation to bring it into line with these instruments.11 Such amendments to national legislation are legitimate on the basis of the higher standard in terms of fundamental rights set at the international level. Sieber speaks of universal human rights as ‘common legal positions’ which shape the face of criminal law both by limiting the resort to criminal law and also legitimating it.12 Conversely, end goals which instrumentally justify harmonisation can include ensuring coherence and fairness of treatment of individuals, especially citizens of the same supranational or federal entity, who would otherwise be subject to different legal regimes. This rationale is relied upon, admittedly jointly with a values-based rationale, when harmonising the rights of individuals – for instance, procedural rights in criminal proceedings.13 A second end goal, instrumentally legitimating the harmonisation of criminal law, can be ensuring legal certainty for individuals. Thanks to harmonisation, citizens do not need to deal with different norms applying to the same conduct. Nells cites the example of confessions, which can be considered as evidence in civil proceedings but not in criminal proceedings. The same individual – the defendant in civil proceedings and the accused in criminal proceedings – is faced with two conflicting norms. This author identifies this as a problem triggered by 10 JR Spencer, ‘Why is the Harmonisation of Penal Law Necessary?’ in Klip and van der Wilt (n 1) 43, 44. 11 ibid 45–46. For a detailed description of the differences between harmonisation by means of criminal law instruments and by means of human rights instruments, see J Vogel, ‘Why is Harmonisation of Penal Law Necessary? A Comment’ in Klip and van der Wilt (n 1) 55, 56–58. On the impact of human rights law on national legislation, see A Ashworth, Principles of Criminal Law (Oxford, Oxford University Press, 2009) 49. For a more general discussion on the impact of fundamental rights on substantive criminal law ranging from criminalisation provisions to the kinds and features of penalties, or the extension of aggravating circumstances, see P de Hert, S Gutwirth, S Snacken and E Dumortier, ‘La montée de l’Etat pénal: que peuvent les droits de l’homme?’ in Y Cartuy­ vels, H Dumont, F Ost, M van de Kerchove and Sebastien van Drooghenbroeck (eds), Les droits de l’homme: bouclier ou épée du droit pénal? (Brussels, Publications des FUSL/Bruylant, 2007) 235, 247–65 and the references cited therein. 12 U Sieber, ‘The Forces behind the Harmonisation of Criminal Law’ in M Delmas-Marty et al (n 6) 385, 388–90. 13 Directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012 establishing minimum standards on the rights, support and protection of victims of crime, and replacing Council Framework Decision 2001/220/JHA [2012] OJ L315/57. The text grants a number of rights to cross-border victims (individuals victimised in cases involving transnational aspects) as well as to national victims (individuals victimised in purely internal cases); see art 1(1). It was argued that one of the reasons for the extension of the relevant rights to national victims was precisely to spread a sense of equality and of justice among EU citizens. See J Öberg, ‘Subsidiarity and EU Procedural Criminal Law’ (2015) 5 European Criminal Law Review 19, 27 ff.

80  Rationales for the Harmonisation and Legitimacy of EU Criminal Law different regulations of the same fact, and harmonisation as a possible solution to it,14 although of course it is not the only one.15 In ‘composite’ legal orders, namely federal or supranational legal orders where different legal sub-systems (federate or national) co-exist, harmonisation can ensure legal certainty for those individuals who have to interact with several systems. A third connected, although different, aim for harmonisation can be that of creating a level playing field to ease transactions. In this context, the presence of different sets of legislation does not necessarily create legal uncertainty. Yet a fragmented legal system can slow down cross-border economic transactions, for instance. This rationale is often relied upon to justify the harmonisation of laws regulating trade at both the global and regional levels, including the EU level,16 and, as explained below, is a key rationale in criminal law. Lastly, harmonisation can also have the aim of signalling the formation of a new community and/or to generate a sense of belonging to this community. For instance, the adoption of national codifications in the nineteenth century was intended, among other things, to highlight the formation of a new community or even of a new state.17 Admittedly, in this last case, while harmonisation is pursued with a final goal in mind, the normative dimension to the process – that is, the ‘values’ that the new norm pursues – can be relevant. Harmonised standards protecting common interests to which people can relate, and which reflect the normative foundations of the newly created community, would naturally have a stronger socialising potential. For instance, crimes against morality such as adultery would have a greater socialising impact in a theocratic state than they would in a secular state. Similarly, criminalising hate and racist speech to protect tolerance and co-existence would have more appeal in a multi-ethnic state or community than in a monoethnic national state or community. B.  Rationales for Harmonisation and the Legitimacy of EU Criminal Law Values-based and instrumental understanding of harmonisation are naturally not mutually exclusive and can and do co-exist in practice, since the legitimacy of each harmonising instrument normally rests on a number of factors. Yet, depending on which rationale is the pre-eminent one, political compromise on the selected harmonised standard might be easier or more difficult to achieve.

14 Nelles (n 1) 31. 15 For instance, the problem of conflicting norms on the same individual is a crucial one in ­multi-level systems, including the EU, where an individual might be subject to conflicting national and supranational norms. In this context, the conflict is solved through the principle of primacy of EU law. On vertical conflicts and the principle of primacy, see D Chalmers, G Davies and G Monti, European Union Law (2010 Cambridge, Cambridge University Press) 203. 16 See art 114 TFEU, which gives the competence to the EU to harmonise national legislation to establish the internal market, where transnational economic transactions take place. 17 Spencer (n 10) 43.

Does Harmonisation have an Inherent, Values-Based Dimension?  81 For instance, if harmonisation is mainly pursued for values-based reasons, for instance, in the context of human rights, compromising on the substance of the harmonised standards might be more difficult than if the ultimate policy objective is to create a level playing field for economic operators. More importantly, the conceptual distinction between values-based harmonisation and instrumental harmonisation is key to understanding what kind of legitimacy of criminal law each harmonisation instrument implies. Chapter 1 illustrated how criminal law has an impact on the liberty and dignity individuals and that a theoretical justification is needed to legitimate such impact. It further distinguished between two broad categories of legitimacy for criminal law: deontological legitimacy and utilitarian legitimacy. The former is based on the the importance of the pre-existing values that the criminal law seeks to protect by punishing behaviours which threaten them, while the latter rests on the contribution that criminalising certain forms of behaviour can bring to any future policy objectives, regardless of the pre-existing interests or values that the behaviour in question affects. Next to general welfare, or utility, as in the original Bentham utilitarianism, the relevant objectives might be simply effective law enforcement or any social engineering goals18 to which criminal law can contribute. When trying to identify which understanding of criminal law lies behind each rationale for harmonisation, it is necessary to take the perspective of the single individual. Harmonisation imposes the introduction of a specific crime within each national legal order. Such crime might or might not exist within each Member State, or most likely might exist, but be defined somehow differently. Whenever a new crime is introduced into each legal system or the legislation is amended making the existing definitions of crimes more severe as a result of harmonisation, a legitimacy question is raised from the perspective of the individuals: why is their dignity and possibly liberty being further restricted? Is it because there has been normative judgement on the inherent value of the conduct or are there some more general objectives the achievement of which can be reached by the enactment of further extensions of the criminal law? Bearing in mind what has been discussed so far on the rationales for harmonisation, it can be concluded that, as a general rule, values-based harmonisation of criminal law implies adopting a deontological approach to criminal law. Conversely, instrumental harmonisation implies a utilitarian understanding of criminal law. However, this is not a perfect equation and there are at least two exceptions to this rule. The following section provides a general introduction to the harmonisation of national law as a strategy of EU integration and lists criminal law rationales. The following sections then focus more specifically on

18 The expression ‘social engineering’ is loosely understood in this context to refer to those efforts to influence particular attitudes and social behaviours on a large scale, whether by governments, media or private groups, in order to produce the desired characteristics in a target population.

82  Rationales for the Harmonisation and Legitimacy of EU Criminal Law the rationales that scholars have identified for the EU harmonisation of national definitions of crimes, distinguishing between utilitarian and deontological rationales. III.  HARMONISATION OF NATIONAL CRIMINAL LAW WITHIN THE EU LEGAL ORDER

Next to unification and coordination, harmonisation of national legislation is one of the strategies of integration within the EU legal order in order to achieve EU policy objectives.19 A first important legal basis that provides the EU with the competence to harmonise national legislation is Article 114 TFEU, which concerns the internal market.20 Other legal bases exist in the Area of Freedom, Security and Justice, especially judicial cooperation in civil,21 and criminal matters, which was covered in Chapter 2 and will be discussed further in Chapters 4 and 5.22 To be precise, both in the field of internal market and the Area of Freedom, Security and Justice, the Treaty speaks of the approximation of national legislation rather than of harmonisation. However, the two terms have been used interchangeably in the literature.23 Furthermore, there are a number of provisions that grant the EU with the competence to develop an ad hoc policy without explicitly mentioning approximation as an integration strategy, but they do not exclude it either. This is the case, for instance, for non-discrimination on the grounds of nationality24 or certain aspects of social security.25 These are also considered to be legal bases for harmonisation.26

19 L Azoulai, ‘The Complex Weave of Harmonisation’ in A Arnull and D Chalmers (eds), Oxford Handbook of EU Law (Oxford, Oxford University Press, 2015) 589, 589. Harmonisation is also seen as the ‘royal way of integration’ in the international process of integration; see M Delmas-Marty, ‘Les processus d’interaction’ (n 6) 423. 20 Articles 114, 115, 116 and 118 TFEU. 21 ibid art 81. 22 ibid arts 82 and 83. 23 On the confusion generated by the interchangeable use of terms such as ‘harmonisation’ and ‘approximation’, as well as ‘cooperation’, ‘unification’, ‘assimilation’ and ‘convergence’, see Tadic (n 1) 2 and 9. Nonetheless, it seems that in the literature, harmonisation is the term more widely used to refer to the process of attenuating differences among national rules and bringing them closer to a supranational standard. See, for example, Azoulai (n 19); S Weatherill, ‘Why Harmonise?’ in T Tridimas and P Nebbia (eds), European Union Law for the Twenty-First Century: Rethinking the New Legal Order (Oxford, Hart Publishing, 2004) 11; M Dougan, ‘Minimum Harmonisation and the Internal Market’ (2000) 37(4) Common Market Law Review 853; J Pelkmans, ‘The New Approach to Technical Harmonisation and Standardization’ (1987) 25(3) Journal of Common Market Studies 249; S Weatherill, ‘The Limits of Legislative Harmonisation Ten Years after Tobacco Advertising: How the Court’s Case Law Has Become a “Drafting Guide”’ (2011) 12(3) German Law Journal 827. 24 Article 18 TFEU. 25 ibid art 48. 26 For an exhaustive list of the policy areas in which harmonisation is admitted, see Azoulai (n 19).

Harmonisation of National Criminal Law within the EU Legal Order  83 Lastly, there are a number of policy areas such as culture or tourism where the Treaty excludes any harmonisation competence for the EU.27 The biggest part of scholarship theorising on harmonisation has focused on legislation adopted on the internal market legal basis of the Treaty,28 discussing among other aspects the values-based dimension of the process.29 The debate on the justifications for the harmonisation of criminal law came at a later stage. This is first because the relevant competence for the EU was only introduced by the Treaty of Maastricht in 1992,30 whereas the competence in relation to the internal market was already present in the 1957 Treaty of Rome.31 Second, until the Treaty of Lisbon, unanimity was required to adopt EU criminal law instruments,32 which thus benefited from strong political legitimacy33 and made the question on their theoretical justifications less pressing.34 Although more recent, the debate on the harmonisation of criminal law has nonetheless rapidly developed.35 Scholars have identified at least six possible justifications for the EU enacting a harmonised definition of crimes: the criminalisation rationale; the cooperation rationale; the free movement rationale; the justice rationale; the socialising rationale; and the regulatory rationale. Among these, the criminalisation rationale is an example of values-based harmonisation implying a deontological approach to criminalisation and is discussed in the next section. The cooperation, free movement and justice rationales are examples of instrumental harmonisation implying a utilitarian approach to

27 See art 195 TFEU on tourism, and art 167 TFEU on culture, art 19 TFEU on non-discrimination on grounds other than nationality and art 84 TFEU on crime prevention. 28 See by mere way of an example Azoulai (n 19); Weatherill (n 23); Dougan (n 23); R Schütze, ‘Limits to the Union’s “Internal Market” Competences: Constitutional Comparisons’ in L Azoulai (ed), The Question of Competence in the European Union (Oxford, Oxford University Press, 2014); A Saydé, ‘One Law, Two Competitions: An Inquiry into the Contradiction of Free Movement Law’ (2010–11) 13 Cambridge Yearbook of Legal Studies 365. 29 On the presence of a normative dimension within internal market harmonisation, see Azoulai, who speaks of the ‘re-evaluation’ of harmonisation: Azoulai (n 19) 592. On normative values and especially lifestyle regulations within EU Law, including internal market harmonising legislation, see also A Alemanno and A Garde, ‘The Emergence of an EU Lifestyle Policy: The Case of Alcohol, Tobacco and Unhealthy Diets’ (2013) 50(6) Common Market Law Review 1745; and F de Witte, ‘Sex, Drugs & EU Law: The Recognition of Ethical and Moral Diversity in Europe’ (2013) 50(6) Common Market Law Review 1545. 30 See extensively ch 4. 31 Article 100 of the Treaty on the Establishment of the European Economic Community (1957). Article 100a of the Single European Act [1986] OJ L169/8 introduced a requirement for qualified majority. 32 Article 34(2) TEU, consolidated in Amsterdam [1997] OJ C340/145. 33 See the distinction between the various kinds of political, social and normative legitimacy discussed in the Introduction. 34 A Klip and H van der Wilt, ‘Introduction’ in Klip and van der Wilt (n 1) vii. 35 Azoulai observes how very similar trends in the role and the nature of harmonisation can be appreciated in the internal market and in the Area of Freedom, Security and Justice, but he then does not elaborate in any great depth on this comparison: Azoulai (n 19) 591. At least to the knowledge of the author, there are not at present any detailed comparisons on the function of harmonisations in the two contexts.

84  Rationales for the Harmonisation and Legitimacy of EU Criminal Law criminalisation and are discussed in section V. Both the socialising and the regulatory rationales can be evidence of either a values-based or an instrumental harmonisation, but they both imply a utilitarian approach to criminalisation. They are discussed separately in sections VI and VII. It should be specified that all these rationales are referred by scholars as ‘virtuous’ rationales. They justify harmonisation on the basis of positive consequences stemming from the process and generally benefiting all the players involved. They must be distinguished from ‘hidden’ or ‘dirty’ rationales for harmonisation, which are not considered in the following sections. A ‘dirty’ rationale is, for instance, the EU using the supranational arena to have texts adopted which they did not manage to have approved at the national level; this will be discussed more extensively in Chapter 7.36 IV.  THE ‘VALUES-BASED’ CRIMINALISATION RATIONALE FOR HARMONISATION: DEONTOLOGICAL EU CRIMINAL LAW

A first values-based rationale which plays a role in the context of EU integration in criminal matters is the criminalisation rationale. The harmonisation of the definition of crimes is pursued to ensure that the behaviour affecting important values are repressed in all EU Member States. This rationale played an important role in justifying the adoption of the first international conventions harmonising the definition of crimes from the 1960s onwards. After the Second World War, the international community developed a global criminal policy agreeing that a number of areas of human behaviour, including terrorism, drug t­rafficking, trafficking in human beings and organised crime, deserved more stringent and more effective – and thus better – criminal law regulation.37 Such a global ­criminal policy was admittedly often driven by the US criminal justice agenda, at least in relation to drug policy, and corporate crime.38 This criminalisation rationale is also discussed by EU criminal law scholars, who formulate it in a twofold manner: first, they point to the need to repress those forms of behaviour that are undesirable because they harm inherent EU

36 See ch 7, s I. 37 On the development of an international criminal policy, see D Flore, ‘Fondements et objectifs de la coopération internationale en matière pénale’ in D Bernard, Y Cartuyvels, C Guillain, D Scalia and M van de Kerkhove (eds), Les fondements et objectifs des incriminations et des peines en droit pénal européen et international (Brussels, Anthémis, 2013) 432’ and Sieber (n 12) 391. 38 On the US exporting its criminal justice agenda on drugs, see Vogel, who speaks of ‘legal colonisation’ with respect to the US lobbying for the incorporation of its standards into terms of drugs policy in the Vienna Convention against Illicit Traffic in Narcotic Drugs and Psychotropic Substances of 1998: Vogel (n 11) 58. For a detailed description of the social and political rationales behind the adoption of the international law on drugs and the role of a US prohibitionist agenda, see R Colson, ‘Prendre le droit international et européen de la drogue au sérieux?’ in Bernard et al (n 37) 207. On corporate crime, see A Niento Martin, ‘Americanisation or Europeanisation of Corporate Crime?’ in Delmas-Marty et al (n 6) 327.

The ‘Justice’, ‘Free Movement’ and ‘Cooperation’ Rationales  85 interests, for instance, EU financial interests;39 and, second, they also frame this argument as the ‘safe haven’ argument. The claim is that failing to have harmonised criminal legislation throughout Europe, criminals would find safe havens where they could seek refuge or in which they could base their criminal activities.40 The fear of ‘safe havens’ is intuitively predicated on the assumption that the activities at stake are generally considered to be undesirable and deserving criminal punishment. If that was not the case, then the problem of safe havens would not exist; the legislative differences would simply be respected as equally legitimate policy choices on the same subject. In this case, harmonisation of legislation is not pursued as a neutral end goal, but in order to impose a qualitatively better rule. The values-based dimension – that is, the values protected by the specific norm – is structural to the rationale for harmonisation. This arguably means that when criminalisation occurs at the supranational level, EU definitions of crimes are justified on a deontological basis. A normative judgement takes place at a supranational level of what is or is not a desirable behaviour for the society in question. The outcome of such a judgement is imposed as law in all Member States. The individual sees his liberty and dignity potentially restricted in order to protect some specific interests. An example of actual legislation where the values-based rationale has a pre-eminent role41 is where the new proposed crimes did not exist in a large number of Member States. In this case, it could be argued that the problem did not really lie in the presence of differences between existing 28 versions of criminal law legislation on specific forms of behaviour; rather, it is more the absence of legislation criminalising specific behaviours that the supranational level conversely deemed worthy of reproach.42 V.  THE ‘JUSTICE’, ‘FREE MOVEMENT’ AND ‘COOPERATION’ RATIONALES FOR HARMONISATION: UTILITARIAN EU CRIMINAL LAW

The instrumental rationales for harmonisation previously defined as the ‘­coherence rationale’, the ‘legal certainty rationale’ and the ‘easing transactions 39 This rationale is explained in detail in A Weyembergh, L’harmonisation des législations: ­condition de l’espace pénal européen et révélateur de ses tensions (Brussels, Editions de l’Université de Bruxelles, 2004) 179–81. 40 ibid 181–82; see also G Vermeulen, ‘Where Do We Currently Stand with Harmonisation in Europe?’ in Klip and van der Wilt (n 1) 73. 41 As was explained in ch 2, each legislative instrument is normally based on multiple rationales; there is rarely only one justification as to why each instrument has been adopted. 42 An example is Directive 2017/1371/EU of the European Parliament and of the Council of 5 July 2017 on the fight against fraud to the Union’s financial interests by means of criminal law [2017] OJ L198/29. The impact assessment clarifies that some of the introduced crimes were absent in most Member States. See Commission Staff Working Paper, Impact Assessment (Part I) – Accompanying the document Proposal for a Directive of the European Parliament and of the Council on the protection of the financial interests of the European Union by criminal law, SWD (2012) 195 final, 18 ff. See infra Ch 7, Section IV.

86  Rationales for the Harmonisation and Legitimacy of EU Criminal Law rationale’ are also relevant to the EU harmonisation of criminal law. In this context, they are conceptualised as the ‘justice’, ‘free movement’ and ‘­cooperation’ rationales (see section V.A). In each of these cases, harmonisation of national criminal law implies a utilitarian understanding for the legitimacy of criminal law (see section V.B). A.  Instrumental Rationales for Harmonisation … The general coherence rationale for harmonisation is connected in the field of criminal law to the objective of spreading a sense of justice among EU c­ itizens. Equality of treatment in the context of criminal justice is considered to be connected to a perception of fairness and thus justice.43 In the same way as guaranteeing the same rights for everyone,44 ensuring that the same forms of behaviour are condemned throughout the whole Area of Freedom, Security and Justice would arguably contribute to spreading a common sense of justice among EU citizens.45 This is referred to here as the justice rationale. In the context of the harmonisation of criminal law, the legal certainty rationale is connected to the key EU objective of free movement. Lack of knowledge of foreign criminal justice systems, and thus lack of legal certainty, is generally assumed to be an element that discourages citizens from exercising their free movement rights. The harmonisation of criminal law would reduce these differences and would make EU citizens more familiar with foreign criminal justice systems and thus willing to move.46 This is referred to here as free movement rationale. Finally, the argument for harmonising to ‘ease transactions’, which is more commonly used with reference to economic actors, is translated into the criminal justice system as ‘easing transactions’ between judicial authorities’. In other words, the harmonisation of the definitions of crimes is said to be necessary in order to ease judicial cooperation. This is referred to here as the cooperation rationale.47 National judicial authorities might need the cooperation of foreign authorities in order to collect evidence abroad or to arrest fugitives

43 See T Tyler, ‘Procedural Justice, Legitimacy, and the Rule of Law’ (2003) 30 Crime and Justice 283. 44 Nelles (n 1) 31. 45 See Weyembergh (n 39) 185; and Vogel (n 11) 60–61. See also T vander Beken, ‘Freedom, Security and Justice in the European Union: A Plea for Alternative Views on Harmonisation’ in Klip and van der Wilt (n 1) 95, 96. 46 Weyembergh (n 39) 184. This rationale is also recalled in official documents of the Commission. This is further discussed in ch 6. 47 Weyembergh (n 39) 139–76. Spencer includes this rationale among his ‘pragmatic’ reasons to harmonise. He broadly speaks of pragmatic reasons when referring to the need to support other Member States in the enforcement of their own criminal law and to the need to combat crime jointly: Spencer (n 10) 47. See also Sieber, who speaks in this context of protecting ‘international security interests in prosecuting transnational crime’: Sieber (n 12) 393 ff.

The ‘Justice’, ‘Free Movement’ and ‘Cooperation’ Rationales  87 beyond national borders. It is argued that a harmonised framework of criminal law legislation across Member States would help achieve the goal of effective judicial cooperation and would thus enable Member States to counteract crime more effectively for at least three reasons. First, the harmonisation of the definitions of crimes would enable the respect of the double criminality requirement. Some legislative texts regulating judicial cooperation require that the offences, whose investigation or prosecution requires the assistance of another state, be punishable in both the requesting state and the requested state. If this is not the case, then in some cases cooperation might be refused.48 If the law is harmonised across Europe, and Member States all have the same or similar crime definitions, then double criminality is automatically met. Second, a harmonised definition of crimes would allow a more uniform definition across Member States of the material scope of action of EU judicial and police cooperation agencies such as Eurojust and Europol, the European Judicial Network and the Joint Investigation Teams which support Member States in judicial cooperation.49 In the case of the EPPO, the question was raised as to whether full harmonisation (and thus unification) would not be an even better solution. This is because the EPPO has actually direct investigative and prosecution powers, and thus a maximum level of legal certainty is necessary. This can only be ensured if all Member States provide the EPPO with exactly the same scope of action.50 In the end, the EPPO regulation opted for referring to a directive rather than a regulation for the definition of its scope of application.51 Third, it is argued that similar or even identical definitions of crimes across Member States would ease judicial cooperation by creating a climate of mutual trust between judicial authorities. The argument is that investigators, prosecutors and judges would be more inclined to provide assistance to one another, to surrender individuals to other jurisdictions or transfer proceedings to

48 Weyembergh (n 39) 139–44. Weyembergh nonetheless acknowledges that this argument was more compelling when older mutual legal assistance instruments were used, which all included a double criminality requirement (see art 5(1) of the European Convention on Mutual Legal Assistance in Criminal Matters, Strasbourg, 20.IV.1959 ETS n 30) and that it lost some importance with the introduction of mutual recognition instruments, which partially eliminated this requirement such as the European Arrest Warrant (Framework Decision 2002/584/JHA of 13 June 2002 on the European Arrest Warrant and the surrender procedures between Member States [2002] OJ L190/1) or the European Investigation Order (Directive 2014/41/EU of the European Parliament and of the Council of 3 April 2014 regarding the European Investigation Order in criminal matters [2014] OJ L130/1). 49 Vermeulen (n 40) 74; see also Weyembergh (n 39) 167–77. 50 Weyembergh (n 39) 174. A similar argument is made in K Ligeti, ‘Approximation of Substantive Criminal Law and the Establishment of the European Public Prosecutor’s Office’ in A Weyembergh and F Galli (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Editions de l’Université de Bruxelles, 2013) 73, 82. 51 Article 22 of Council Regulation 2017/1939/EU of 12 October 2017 implementing enhanced cooperation on the establishment of the European Public Prosecutor’s Office. See also the discussion in ch 7.V.B.

88  Rationales for the Harmonisation and Legitimacy of EU Criminal Law jurisdictions that have a similar legal system. A practical example can help clarify the importance of the harmonisation of criminal law in this context. It was stated above that similar forms of legislation can help meet the double criminality requirement in the case of judicial cooperation instruments. The European Arrest Warrant, which regulates extradition proceedings among EU Member States, is one instrument which sensitively reduced the scope of application of this rule. The text requires Member States to surrender suspects accused of or convicted of an offence that does not exist under national law for a large number of crimes.52 This can raise tensions with the principle of legality, and the question of the compatibility of this rule with this principle was raised before the Court of Justice.53 In the Advocaten voor de Wereld case, the Court nonetheless held that it suffices that legality is checked in the state issuing the request for surrender, which is prosecuting or has convicted the requested person. Nonetheless, surrendering a person whose action does not represent a crime in the executing state can still create some anxieties for the executing authorities. Having similar legislation would reduce the threat to legality and thus Member State nervousness in this respect.54 Finally, having more similar legislation, and thus increasing the level of trust between judicial authorities of different Member States, could also arguably help to solve conflicts of jurisdiction. If legislation and criminal justice systems were more alike, compromises between Member States would possibly be easier to reach. Judges would not systematically feel that the legal principles of their legal system might not be respected if the case were to be tried in another Member State’s court and would thus be less inclined to fight to retain jurisdiction over a certain case.55 B.  … Implying a Utilitarian Legitimacy for EU Criminal Law In each of the three cases discussed above, the process of bringing norms together is pursued as a means to an end goal, be this spreading a sense of justice, fostering free movement or supporting judicial cooperation, but without an inherent

52 Article 2(2) and 2(4) of Framework Decision 2002/584/JHA of 13 June 2002 on the European Arrest Warrant and the surrender procedures between Member States [2002] OJ L190/2. 53 Case C-303/05 Advocaten voor de Wereld VZW v Leden van de Ministerraad [2007] ECR I-03633. 54 Weyembergh (n 39) 144–66; Vogel (n 11) 64. This author also mentions the example of the 1998 EU Convention on Driving Disqualification, which required mutual recognition of driving disqualifications, which required mutual recognition, but could not work smoothly because the substantive rules in Member States were not really compatible (Council Act of 17 June 1998 drawing up the Convention on Driving Disqualifications (98/C 216/01) [1998] OJ C216/1). 55 N Long, ‘Harmonisation of Criminal Law in the EU, Study for the LIBE Committee’, 8, www.europarl. europa.eu/RegData/etudes/etudes/join/2010/419631/IPOL-LIBE_ET%282010%29419631_EN.pdf.

The ‘Justice’, ‘Free Movement’ and ‘Cooperation’ Rationales  89 values-based dimension. A policy choice in favour of one or another harmonised standard must obviously be made.56 However, technically any harmonised rule would serve this purpose. The content of the minimum standard is not decisive for the achievement of the goal, nor is the rationale which justifies harmonisation. For instance, EU citizens undertaking a business could be motivated to operate transnationally and feel that they were treated equally both if false accounting is criminalised or de-criminalised across Europe, as long as everyone is subject to the same constraints in terms of reporting their annual business accounts. Similarly, double criminality is met, and thus cooperation works smoothly, both if the possession or consumption of soft drugs is criminalised or de-criminalised, as long as everyone has the same rule.57 In each case, the instrumental character of harmonisation also coincides with a utilitarian approach to criminalisation. As suggested above, it is necessary to take the perspective of the individuals who see their liberty and dignity restricted by the introduction of new crimes. In the case of the free movement and justice rationales, such restrictions are justified in order to achieve the relevant ‘social engineering’ goals, such as encouraging free movement or spreading a sense of justice. EU citizens see their liberty and dignity limited by the EU through harmonising action in order to reach some extra-criminal law objectives. In the case of the cooperation rationale, it is not social engineering that is pursued through harmonisation, but arguably the enforcement of criminal law in transnational cases. When the EU relies only on this instrumental rationale for harmonisation, it requires Member States to introduce new crimes into their legislation for the mere sake of having a uniform crimes definition scheme so that it would be easier to fight these crimes through inter-state cooperation. The final goal is that of ensuring that a number of Member States that have a more repressive criminal law than others are able to enforce such criminal law even in transnational cases. Citizens of a less repressive Member State would then see their liberty and dignity further restricted due to the enforcement needs of some other Member States which have stricter criminal law legislation. The following example will help to clarify this point. When protests against the Expo Exhibition in Milan in 2015 occurred, alongside peaceful demonstrations, violent confrontations also took place. A prosecution in Italy was

56 Tadic (n 1) 14; with respect to the internal market, see Weatherill (n 23) 16. Weatherill explains that market-building through harmonisation implies only the need for common rules and makes no assumptions about the quality of the common rule. However, he continues, in practice choices have to be made about the quality of the standard of regulatory protection set by the EU. 57 Naturally, from the perspective of Member States with stricter criminal laws, the ideal solution would be to have the same level of repressiveness across Europe, so that they would be able to enforce their criminal law in each transnational case. The quality of the new uniform rule obviously matters to them. However, the content of such uniform norm does not matter in absolute terms for harmonisation to be justified. See the example of Italy and Greece given below.

90  Rationales for the Harmonisation and Legitimacy of EU Criminal Law launched against several demonstrators who were accused of causing damage and aggravated resistance.58 A number of them were Greek and had already returned to Greece when the trial began. The Italian authorities issued five European Arrest Warrants to have the fugitives arrested in Greece and tried in Italy. However, the Greek authorities refused to surrender the suspects on the basis that, among other grounds, the offence of ‘destruction and looting’, which some of the protesters were accused of, did not exist in the Greek legal system. The offence did not fall within the list of crimes for which the European Arrest Warrant Framework Decision rules out the requirement of double criminality,59 and therefore the Greek authorities had the right to refuse cooperation. This case thus illustrates how differences in the form of national criminal legislation can represent an obstacle to cooperation. In practice, differences in legislation meant that Italy could not have its criminal law properly enforced in this crossborder case. Let us imagine that Italy argued in the Council that legislative differences on looting and destruction be eliminated for the sake of making cooperation more effective and that it managed to have harmonisation adopted on this point.60 For those Member States that already have these crimes defined in their legislation (Italy in our case), such EU criminalisation choices would simply coincide with their own autonomous, arguably values-based, criminalisation choices. Those Member States that have to introduce the relevant offences ex novo (Greece in our example) would have their values-based policy choices not to criminalise the specific conduct neutralised in order to ensure Italy’s enforcement needs, which would be supported and promoted by the EU through harmonising legislation. Admittedly, the Italian criminal policy choice of criminalising looting and destruction might not per se be enforcement-based, but values-based. Italy might think that these are dangerous forms of behaviours and thus deserve to be criminalised. However, if the EU endorses such an agenda solely in order to ease cooperation, with no prior EU reasoning on whether the relevant conduct harms interests sufficiently to deserve criminalisation in all Member States, the EU is purely seeking an enforcement goal. De facto, in our example Greek citizens would see their freedom limited for the sake of Italian enforcement needs.

58 A description of the case as well as full references can be found in AE Kouroutakis, ‘The Italian European Arrest Warrants for the Five Greeks Taking Part in Riots and Their Rejection by the Greek Authorities’ (2016) 7(3) New Journal of European Criminal Law 295. 59 Article 2(2) and 2(4) of Framework Decision 2002/584/JHA (n 52). 60 This example is hypothetical, as no legislative proposal was advanced at the EU level on harmonising looting and destruction. However, the question on how to align Dutch legislation with the legislation in other Member States has been raised in the past with respect to drugs policy. Indeed, Berkman argued in 1996 that ‘the most obvious route towards harmonisation of drug laws within the EU would be for the Netherlands to adopt soft drug policies that other Member States presently observe’ and the other option would be for the Netherlands to promote Dutch policy throughout the EU. See ET Berkman, ‘Sacrificed Sovereignty?: Dutch Soft Drug Policy in the Spectre of Europe without Borders’ (1996) 11(1) Boston College International and Comparative Law Review 173, 180.

The Socialising Rationale   91 Similar hypothetical examples might concern crimes such as escape from prison, which is a crime in the UK but not in the Netherlands,61 or the rules on the freedom of the press, which are particularly liberal in Sweden and might allow forms of behaviour that are considered criminal in other countries.62 Scholars have been critical of this utilitarian understanding of EU c­ riminal law. De Hert speaks of ‘moral philosophy’ or a principled objection to use harmonisation to support cooperation. If Member States made different criminal policy choices in certain specific fields, they should not be forced to reconsider them for the sake of enabling mutual recognition; they should simply be entitled to deny procedural cooperation in light of these substantive reasons.63 Kaiafa-Gbandis observes how in this case, harmonisation of definitions of crimes (substantive criminal law) is undesirably subordinated to judicial cooperation (procedural law). Normally the substance should come before the procedure.64 VI.  THE SOCIALISING RATIONALE FOR HARMONISATION: UTILITARIAN EU CRIMINAL LAW

The adoption of a new form of legislation to signal the formation of a new community is also particularly relevant to the EU harmonisation of national definitions of crimes. The EU, as a non-state entity, is in need of communitybuilding factors. Criminal law is particularly appealing in this respect, since the monopoly on the use of force and the administration of criminal justice is one 61 Escape is a common law offence in the UK as can be read in the Crown Prosecution Office list of public justice offences: www.cps.gov.uk/legal/p_to_r/public_justice_offences_incorporating_the_ charging_standard/#a33. The example includes the UK, which at the time of writing is still part of the EU and thus to which the Framework Decision on the European Arrest Warrant still apply. Admittedly, the example is a fairly hypothetical one. In practice, it is quite unlikely that a prisoner would manage to escape without committing any other crimes, such as ‘corruption of prison staff’ or ‘damaging public property’ (for instance, by breaking a window). A European Arrest Warrant request would probably then be issued from the UK to the country where the person has escaped in order to ensure the prosecution of this person for these additional offences rather than for the escape from prison. Moreover a European Arrest Warrant request could be sent to ensure that the person is sent back to prison to serve the prison term for which he or she was initially sentenced (see art 1(1) of Framework Decision 2002/584/JHA, which makes it possible to issue warrants both for prosecution and execution purposes). 62 Interestingly enough, Julian Assange relied on the particularly wide freedom of the press in Sweden when deciding to manage WikiLeaks by basing himself there. See the observations in P de Hert and D Kloza, ‘WikiLeaks, Privacy, Territoriality and Pluralism: What Does Matter and What Does Not?’, TILT Weblog Law and Technology, https://cris.vub.be/files/37521046/­WikiLeaks_ privacy_territoriality_and_pluralism._What_does_matter_and_what_does_not.pdf. 63 P de Hert, ‘Cybercrime and Jurisdiction in Belgium and the Netherlands: Lotus in Cyberspace – Whose Sovereignty is at Stake?’ in B Koops and SW Brenner (eds), Cybercrime and Jurisdiction: A Global Survey (The Hague, TCM Asser Press, 2006) 88; vander Beken (n 45) 97, M Fletcher, R Lööf and B Gilmore, EU Criminal Law and Justice (Cheltenham, Edward Elgar, 2008) 192. 64 M Kaiafa-Gbandi, ‘The Importance of Core Principles of Substantive Criminal Law for a ­European Criminal Policy Respecting Fundamental Rights and the Rule of Law’ (2011) 1(1) ­European Criminal Law Review 11.

92  Rationales for the Harmonisation and Legitimacy of EU Criminal Law of the sovereign functions of the state (jura regalia). As a consequence, acquiring competence in this policy area and adopting autonomous policy decisions therein has a specific symbolic significance for a supranational organisation. Indeed, it might work as a legitimising factor for the EU project and move it towards being a state-like entity.65 Harmonising criminal law with this aim in mind is referred to here as following a socialising rationale.66 As mentioned above, when using harmonisation as a community-building tool, the ‘quality’ of the selected harmonised standard matters. While any set of common norms would have at least some community-building effect, some norms will have a greater socialising potential. For instance, given the multicultural nature of the EU polity and the consequent centrality of the fight against racism, the harmonised criminalisation of hate speech arguably has a greater community-building impact than harmonised definitions of, for instance, armed robbery. Thus, the socialising rationale for harmonisation might imply a valuesbased understanding of harmonisation. Yet this does not necessarily imply a deontological understanding of criminal law. Indeed, the relevant interests that all Member States would have to protect via criminal law are selected on the basis of their socialising potential and not necessarily on the basis of their absolute importance. An EU-wide prohibition of burning EU flags might, arguably, foster a sense of respect towards the EU and have some socialising potential, but this does not mean that using criminal law is justified in this context. This is thus the first exception to the rule that values-based justified harmonisation necessarily implies a deontological approach to EU criminal law. More pertinently, a socialising rationale was arguably behind the adoption of the Framework Decision on racism and xenophobia.67 In this case, clarifying the values on which the EU grounded and establishing the identity of the EU as a community of values played a role in the decision to harmonise these crimes. This is further elaborated upon in Chapter 7. VII.  THE ‘REGULATORY’ RATIONALE FOR HARMONISATION: UTILITARIAN EU CRIMINAL LAW

A sixth justification for the harmonisation of definitions of crimes is that of ensuring the enforcement of EU norms.68 This rationale is not part of those 65 The appeal of criminal law as a legitimating factor for the EU is discussed by C Sotis, ‘“Criminaliser sans punir”: Reflexions sur le pouvoir d’incrimination (directe et indirecte) de l’Union européenne prévu par le traité de Lisbonne’ (2011) 4 Revue de Science criminelle et de droit pénal comparé 773. 66 See Weyembergh (n 39) 186. Spencer highlights that some supporters of the Corpus Juris (a broad project which included among other aspects the establishment of a European Public ­Prosecutor) also might have had in mind some community-building objectives: Spencer (n 10) 43. 67 Council Framework Decision 2008/913/JHA of 28 November 2008 on combating certain forms and expressions of racism and xenophobia by means of criminal law [2008] OJ L328/55. 68 Vogel (n 11) 61. See also the literature mentioned in ch 5.

Conclusions  93 listed above for harmonisation in general, but is specific to the harmonisation of criminal law. It will be referred here as the regulatory rationale. This is relevant for the EU as the Treaty of Lisbon introduced a legal basis in Article 83(2) TFEU justifying harmonisation on this basis. While the history and text of the article will be discussed more in detail in Chapter 5, here it suffices to clarify the following. The scenario is that of the EU having already regulated a specific policy area. In order to ensure that EU citizens and legal entities comply with the relevant provisions of EU law, the EU can also take a decision to attach criminal law provisions to their violation. The two passages – regulating a policy area and attaching criminal sanctions to the violation of such norms – are chronologically distinct from a ‘conceptual point of view’, even if they can naturally be pursued simultaneously. The EU can either first adopt a text on a non-criminal law basis with the relevant rules of conduct and require through a separate criminal law directive that all Member States impose criminal sanctions on the transgression of such rules, or it can include behavioural rules and criminal sanctions in the same text. Regardless of whether harmonisation of the behavioural rules follows a values-based or an instrumental agenda, the justification for the resort to criminal law in this case is purely based on enforcement grounds. This rationale replicates at a supranational level the utilitarian rationale for criminalisation that was defined in Chapter 2 as ‘regulatory’ criminal law. Assuming that the harmonisation of internal market legislation pursues a values-based agenda, namely increasing protection for consumers, this does not necessarily mean that the use of criminal sanctions is normatively justified in this context. The values at stake might be sufficiently important to deserve protection in each Member State, but not so essential as to justify protection though the use of criminal law. Alongside the socialising rationale, this is thus a second exception to the rule that harmonisation justified on values-based grounds implies a deontological approach to EU criminal law. A fortiori, if the harmonisation of internal market rules does not pursue any particular normative aim and only aims to create a level playing field on technical details, if the EU decides to introduce criminal law sanctions in this area, this would only be based on utilitarian grounds. In these cases, EU citizens would see their dignity and liberty further restricted for the sake of the effectiveness of EU norms or, to use terminology referred to in Chapter 1, for the sake of the managerial functions of the EU as a regulatory entity. VIII. CONCLUSIONS

This chapter has clarified when we can say that the EU harmonisation of definitions of crimes implies a deontological or utilitarian approach to criminal law. Chapter 1 illustrated how criminal law can be justified either on a values-based, deontological basis or on utilitarian grounds, that is, in light of the objectives

94  Rationales for the Harmonisation and Legitimacy of EU Criminal Law it will achieve. Harmonisation of national legislation can similarly be justified on different rationales. It is per se a neutral process which might be pursued to ensure that all Member States adopt a ‘qualitatively better’ norm – what was defined as ‘values-based harmonisation’. Yet it can also simply aim to create a legal level playing field which is instrumental to achieving other ends – what was defined as ‘instrumental harmonisation’. However, it would be incorrect to equate a values-based approach to harmonisation with a deontological approach to EU criminal law, as rationales for harmonisation and for criminalisation do not always overlap. Admittedly, the EU harmonisation of criminal law can be straightforwardly justified on the basis of a values-based agenda which considers certain interests so important as to deserve protection via criminal law (what was called the values-based criminalisation rationale). In this case, a normative approach to harmonisation implies a deontological justification to EU criminal law. Yet, harmonisation of the definition of crimes can also be pursued for different reasons. The EU might decide to introduce criminal sanctions in a certain field to secure the enforcement of its policies (the regulatory rationale) or to highlight important values to which members of a community would ideally relate and thus foster a sense of belonging through that (the socialising rationale). In both cases, the EU might pursue a values-based normative agenda when harmonising the relevant behavioural rules. It might safeguard important interests such as consumer protection or respect for EU symbols, as in the hypothetical case of a provision prohibiting the burning of the EU flag. Yet, what justifies the introduction of criminal law is its enforcement and socialising potential rather than the values at stake. In these two contexts, a values-based harmonising agenda does not imply a deontological approach to EU criminal law, but rather a utilitarian one. Finally, when the EU harmonises definitions of crimes with an instrumental agenda, this automatically implies a utilitarian approach to EU criminal law. This happens when harmonisation is pursued to spread a sense of justice among EU citizens (the justice rationale), to encourage free movement (the free movement rationale) or to support judicial cooperation (the cooperation rationale). In practice, legislative instruments are most likely to be adopted with the aim of achieving several objectives, both normative and instrumental, possibly combining utilitarian and deontological approaches to EU criminal law. The aim of the book is nonetheless first to rely on this analytical framework to analyse which approach to the legitimacy of EU criminal law has been the prevailing one (if any) in EU primary and secondary law and EU policy documents, and second, to critically assess the coherence of the EU approach to criminalisation with EU constitutional principles and values. As was mentioned in the previous chapter, a purely values-based justification for EU criminal law, which neglects to examine whether criminal law is actually effective in deterring the undesirable conduct, would create tensions with the principle of subsidiarity. However, solely relying on the effectiveness of criminal law, or any other objectives that harmonisation of criminal law can achieve, would conflict with the EU founding values.

4 EU Competences on Securitised Criminalisation From a Utilitarian to an Integrated Approach to EU Criminal Law

T

his chapter marks the beginning of the analytical part of the book, addressing the second and third research questions mentioned in the Introduction. In particular, it looks at how the EU primary law justifies resorting to criminal law and to what extent these choices are consistent with EU values and principles. The scope of EU competence to harmonise definitions of crimes has admittedly been fragmented and unclear for a long time. The scope of EU action in criminal matters was somehow blurry in the Treaties of Maastricht and Amsterdam. Both Treaties listed a number of areas of crime where the EU had competence, but the Council adopted harmonising instruments in a number of areas for which a legal basis could only indirectly be found in the Treaty.1 Moreover, the Court of Justice also contributed to expanding the range of EU competences beyond what was stated in the Treaties.2 The Treaty of Lisbon brought some clarity, codifying for the most part the pre-existing legislative and judicial understanding of the EU range of action, although questions on competence boundaries still remain.3 Pinpointing the policy areas in which the EU had and has the competence to identify common definition of crimes is thus not an easy task. Among the various available classifications,4 the discussion here relies on Mitsilegas’ distinction between EU competence in the field of ‘securitised criminalisation’, which is discussed in this chapter, and ‘functional criminalisation’, which is discussed in Chapter 5.5

1 See ch 7, s III. 2 See especially ch 5, ss II and III. 3 See s III.A below. 4 For other classifications, see C Sotis, Il diritto senza codice (Milan, Giuffrè, 2007) 69–98; M Wade, ‘Developing a Criminal Justice Area in the European Union’, European Parliament Study 2014, Study PE 493.043, 23–24, www.europarl.europa.eu/RegData/etudes/etudes/join/2014/493043/ IPOL-LIBE_ET(2014)493043_EN.pdf. 5 V Mitsilegas, EU Criminal Law after Lisbon (Oxford, Hart Publishing, 2016) 54.

96  EU Competences on Securitised Criminalisation This classification is better illustrated in section I; section II illustrates EU securitised criminalisation competences in the Treaties of Maastricht and Amsterdam and investigates what approach to the legitimacy of EU criminal law they entail; section III presents EU criminalisation competences in the Treaty of Lisbon, namely Article 83(1) TFEU. The interest in such historical recollection lies in the possibility of identifying any specific trends which help better understand the approach to legitimacy underpinning the current EU competence in the Treaty of Lisbon. Finally, section IV provides some concluding remarks on the consistency, or lack thereof, with EU constitutional values and principles. I.  EU CRIMINALISATION COMPETENCES: SECURITISED v FUNCTIONAL CRIMINALISATION

With the expression ‘securitised criminalisation’, Mitsilegas refers to the use of criminal law to respond to security threats posed by serious criminality.6 He observes that a number crimes on which the EU has adopted legislation (terrorism, trafficking in human beings, sexual exploitation of women and children, illicit drug trafficking, illicit arms trafficking, money laundering, corruption, counterfeiting of means of payment, computer crime and organised crime) are harmful behaviours which puts the security of EU citizens in danger7 and have been elevated to global security threats by the international community after the Cold War.8 As will be further clarified in the following sections, EU securitised criminalisation was initially dealt with in the context of the third pillar, a specific intergovernmental institutional framework, where rules other than the community method applied.9 The Treaty of Lisbon eliminated the pillar structure and EU securitised criminalisation is currently regulated by Article 83(1) TFEU following, for the most part, the community method.10 With the expression ‘EU functional criminalisation’, Mitsilegas refers to the use of criminal law to safeguard EU policies, objectives and interests.11 In this case, the EU uses criminal law as a response to its own victimisation. Poor implementation of EU policies can be seen as affecting the EU public order. Moreover, criminal behaviours affecting specific EU interests, such as EU financial interests, challenge not only the functioning of the EU but also the existence of the EU.12 As will be discussed in Chapter 5, institutionally, the objective of 6 ibid 54. 7 Mitsilegas does not define security in this context, but intuitively it can be understood as ‘a low probability of damage to acquired values’, as in DA Boldwin, ‘The Concept of Security’ (1997) 23 Review of International Studies 5. See also the discussion on security in s II. B below. 8 Mitsilegas (n 5) 58. 9 See extensively s II. 10 See s III below. 11 Mitsilegas (n 5) 54. 12 Sotis (n 4) 76 and 82. George Fletcher speaks in this context, especially with respect to the protection of the financial interests of the EU, of ‘parochial criminal law’, meaning criminal law that

Securitised v Functional Criminalisation  97 enforcement of EU policies and protection of EU financial interests was initially dealt with through administrative law under the first pillar, where the traditional community method applied. In 2005, the Court of Justice introduced the possibility of enacting first pillar criminal law.13 At present, functional criminalisation is regulated by Article 83(2) TFEU, following for the most part the community method.14 Like any classifications, that between functional and securitised criminalisation is of course one that can be improved and some nuances should be introduced. First, obviously each single criminal norm can have multiple purposes. For instance, legislation mainly aimed at protecting the EU public order – that is, the enforcement of EU policies – might also serve the purpose of protecting the security of EU citizens’ and vice versa. For instance, the criminalisation of trafficking in human beings has as its first aim the protection of the physical integrity and self-determination of individuals. Yet it also has strong links to the EU fight against illegal immigration and is thus an instrument to support a specific EU policy. Similarly, criminal law to enforce EU policies (eg, environmental policies) can also contribute to protecting the security of EU citizens by providing them with a non-polluted environment.15 Second, the overlap between policy areas and institutional arrangements – the first v the third pillar, and Article 83(1) v Article 83(2) TFEU – is not a perfect one. For instance, the protection of the financial interests of the EU has been dealt with in both institutional frameworks. Already under the third pillar, the EU adopted a Convention on the corruption of EU officials16 and a Convention on the protection of the financial interests of the EU.17 Yet, in the 1968 Greek Maize case, the Court of Justice also discussed the implication of loyal cooperation, a first pillar principle at the time,18 for the enforcement of EU policies

only protects interests specific to the system and not universal values: G Fletcher, ‘Parochial versus Universal Criminal Law’ (2005) 3 Journal of International Criminal Justice 20, 31. 13 Case C-176/03 Commission v Council [2005] ECR I-7879; Case C 440/05 Commission v Council [2007] ECR I-9097. 14 On this provision, see extensively ch 5. 15 J Alix, ‘Les Frontières de l’harmonisation autonome’ in G Giudicielli-Delage and C Lazerges (eds), Le droit pénal de l’Union Européenne au Lendemain du Traité de Lisbonne (Paris, Société de Législation Comparée, 2012) 147, 150. Estella Baker and Christopher Harding observe that the third pillar ‘had somehow a frangible and mobile content’ and that in a number of ways it is not the content of third pillar activity that is distinctive, but the reasons for using it instead of the first pillar, namely Member State preference for the intergovernmental method: E Backer and C Harding, ‘A Longitudinal Study of the Third Pillar’ (2009) 34(1) European Law Review 25, 38. 16 Convention drawn up on the basis of Article K.3(2)(c) of the Treaty on European Union on the fight against corruption involving officials of the European Communities or officials of Member States of the European Union [1997] OJ C195/1. 17 Convention drawn up on the basis of Article K.3 of the Treaty on European Union, on the protection of the European Communities’ financial interests [1995] OJ C316/49. 18 The Court of Justice expanded the principle to the third pillar in Case C-105/03 Criminal Proceedings against Maria Pupino [2005] ECR I-05285.

98  EU Competences on Securitised Criminalisation and the protection of EU financial interests.19 Under the current framework, Article 83(1) TFEU mentions money laundering, corruption and counterfeiting of means of payment, which are all offences that can potentially affect EU financial interests. It is still the case that ad hoc legislation on the protection of the EU budget, such as the PIF Directive20 (discussed in Chapter 7), is being adopted on the basis of Article 83(2) TFEU.21 Interestingly, given the special position of EU criminal law protecting the financial interests of the EU, some authors have even suggested that it should be treated as a separate category from securitised and functional criminal law.22 Nevertheless, leaving aside these aspects, the distinction between securitised and functional criminalisation remains a helpful one and will be used to organise the discussion in this chapter and Chapter 5. It has the merit of simplicity and clarity. Moreover, it provides a good fit with the theme of this book. Indeed, the various rationales for harmonisation discussed in Chapter 3 have played a different role in these two policy areas. It is thus sensible to treat them separately when investigating the weight of each rationale and the resulting approach to the legitimacy of EU criminal law. II.  A UTILITARIAN LEGITIMACY FOR EU CRIMINAL LAW UNDER THE MAASTRICHT AND AMSTERDAM THIRD PILLAR

The founding Treaties did not contain any reference to criminal law. This silence could be interpreted in two ways. The first option was to consider criminal law as a special field, which would have deserved a special mention in the Treaty. The omission on the matter thus prevented the European Economic Community (EEC) from enacting any criminal legislation.23 A second interpretation was that criminal law was a tool at the disposal of the European legislator to enforce EC law. The competence to use criminal law would thus be implicit whenever the EEC was granted the power to act in a certain policy area.24 19 Case C-68/88 Commission of the European Communities v Hellenic Republic [1989] ECR I-02965. This case is further discussed in ch 5, s II. 20 Directive 2017/1371/EU of the European Parliament and of the Council on the fight against fraud to the Union’s financial interests by means of criminal law [2017] OJ L198/29. 21 On the competence to enact harmonising legislation in the field of the protection of EU financial interests, see s III. 22 Marianne Wade (n 4) in particular makes a further distinction within the category of ‘functional criminal law’ between EU criminal law specifically enacted to protect EU interests, such as EU financial interests, and EU criminal law enacted in the context of the exercise of EU regulatory powers, which would be EU criminal law enacted in order to enforce EU policies. 23 See Mitsilegas (n 5) 65; and M Wasmeier and N Thwaites, ‘The “Battle of the Pillars”: Does European Community Have the Power to Approximate National Criminal Laws?’ (2004) 29(5) European Law Review 613, who argue that criminal law was not be seen as a separate Community policy (like the internal market or the environment), but as a field of law which can horizontally advance the Community objectives. 24 Mitsilegas (n 5) 65.

A Utilitarian Legitimacy for EU Criminal Law   99 In fact, no EU criminal law was adopted before the mention of criminal law in the Maastricht Treaty, somehow confirming the first interpretation. As is well known, in 1992 with the Treaty of Maastricht, the above-mentioned ‘pillar structure’ was introduced. Various policy areas were regulated within different institutional frameworks where different rules applied, especially concerning the legislative procedure. These diverging settings were respectively contained in the Treaty Establishing the European Community (the first pillar)25 and the Treaty on European Union (the second and third pillars).26 EU action in criminal justice matters was to be regulated by the third pillar, where instruments were adopted by unanimity in the Council, with no role for the European Parliament and a reduced role for the Court of Justice.27 The Treaty of Maastricht did not include any specific competence for the EU to harmonise national criminal law. Article K.1.7 and K.1.9 TEU (consolidated in Maastricht) simply listed judicial and police cooperation for the purposes of combating terrorism, unlawful drug trafficking and other serious forms of international crime as areas of common interest for Member States. Harmonisation made its way in the Treaty only with the Treaty of Amsterdam. Article 29 TEU (consolidated in Amsterdam) gave a new objective to the EU, namely that of providing EU citizens with an Area of Freedom, Security and Justice. This was to be achieved by developing common action among the Member States in the fields of police and judicial cooperation in criminal matters and by preventing and combating racism and xenophobia. Further, Article 29 TEU listed the areas of crime, organised or otherwise, which should be prevented and combated. These were in particular terrorism, trafficking of persons and offences against children, illicit drug trafficking and illicit arms trafficking, corruption and fraud. Finally, Article 29 TEU listed the tools through which the fight against crime should be pursued, mentioning police cooperation, judicial cooperation and, as a last option, approximation, where necessary, of rules on criminal matters in the Member States. Harmonisation was to be pursued through a new legal instrument introduced by the Treaty of Amsterdam, namely Framework Decisions. Article 31(1)(e) TEU specified that the EU had the competence to harmonise national legislation only in the field of drug trafficking, terrorism and organised crime. Neither the Treaty of Maastricht, which did not even mention harmonisation, nor the Treaty of Amsterdam provided any specific indications as to the functions of this strategy of legal integration in criminal justice matters. Nonetheless, it is possible to draw some conclusions from the subsidiary position given to harmonisation in the Treaty (see section II.A) and from the objective of the Area of Freedom, Security and Justice (see section II.B). 25 Treaty Establishing the European Community (consolidated in Maastricht) [1992] OJ C224/6. 26 Treaty on European Union [1992], OJ C191/1. 27 For a description of the third pillar institutional settings, see S Miettinen, Criminal Law and Policy in the European Union (London, Routledge, 2013) 23 ff.

100  EU Competences on Securitised Criminalisation A.  The Position for Harmonisation in the Treaty: Evidence of a Cooperation Rationale The fact that harmonisation has been inserted into the Treaties at a later stage than judicial cooperation and that it is put in a subordinate position – only ‘when necessary’ – can be interpreted as establishing an instrumental link between these two strategies of integration. Resort to harmonisation would be justified when ‘necessary’ to support judicial cooperation. As Chapter 6 will explain in more detail, the policy discourse around harmonisation in both the Tampere and The Hague multi-annual programmes, adopted respectively in relation to the Treaties of Maastricht and Amsterdam, confirms such instrumental ‘cooperation rationale’ for harmonisation.28 Moreover, post-Lisbon scholarship qualifies the Treaty of Lisbon as ‘revolutionary’ for envisaging an autonomous criminalisation function for harmonisation, independent of cooperation.29 This would imply that this was not the case under the previous institutional framework,30 confirming a contrario the interpretation of the Amsterdam and Maastricht framework above as only envisaging a cooperation rationale. Briefly, in the first years of EU criminal law, the institutional framework envisaged harmonisation only as an instrument to create a level playing field for cooperation to work smoothly. As was explained in Chapter 3, such an instrumental approach to harmonisation implies a utilitarian understanding for the legitimacy of EU definitions of crimes. Member States might be requested to introduce new crimes in their national legal order to ensure more efficient interstate cooperation. This is regardless of whether there are legitimate normative reasons in some Member States for leaving the specific activity regulated by administrative or civil norms. Flore observes that the primary focus within the Treaties of Maastricht and Amsterdam on the objective of judicial cooperation, rather than on harmonisation, was somehow in contrast with what occurred in the international arena. Already by the 1960s, the UN had adopted the first conventions criminalising transnational crimes, developing a proper international criminal policy.31 Rules governing judicial cooperation, like those on extradition, were inserted in these substantive conventions rather than in ad hoc instruments.32 Judicial

28 On the multi-annual programme, see extensively ch 6. 29 See s III.B.iii below. 30 Weyembergh observe that the Amsterdam Treaty already opened the door for harmonisation as an autonomous strategy of integration, independent of cooperation: A Weyembergh, ‘Introduction’ in A Weyembergh and F Galli (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Editions de l’ULB, 2013) 9, 11. 31 On this, see D Flore, ‘Fondements et objectifs de la coopération internationale en matière pénale’ in D Bernard, Y Cartuyvels, C Guillain, D Scalia and M van de Kerchove (eds), Fondements et Objectifs des incriminations et des peines en droit Européen et International (Brussels, Anthémis, 2013) 427, 432. 32 See art 9 of Convention C195/1 (n 16).

A Utilitarian Legitimacy for EU Criminal Law   101 cooperation and criminalisation were perceived as complementary strategies to fight crime; however, the emphasis was put on criminalisation, contrary to what happened with the EU Treaties. Yet, interestingly, despite the EU Treaties’ strong focus on cooperation and a marginal role for harmonisation, in the 1990s, under the Maastricht Treaty framework, the Council adopted conventions and joint actions containing both criminalisation obligations and rules on judicial cooperation, for instance, in relation to extradition. In this, it replicated in practice the approach of international law instruments. Examples include the joint action on racism and xenophobia,33 the joint action on trafficking in human beings,34 the joint action on participation to a criminal organisation,35 the Convention on corruption in the private sector36 and the Convention on the protection of EU financial interests.37 Later, under the Treaty of Amsterdam, some of these instruments were repealed by Framework Decisions which only retained the part on criminalisation, leaving the aspects on cooperation to be regulated in separate instruments.38 And actually, values-based and in some cases socialising rationales, as opposed to the instrumental cooperation rationale, justified the adoption of these instruments, especially the Framework Decision on racism and xenophobia. Chapter 7 further discusses rationales for the harmonisation of definitions of crimes in the EU secondary law. It interestingly shows that EU criminal legislation was justified on deontological grounds from the start, amounting in some cases to symbolic EU criminal law. B.  The Objective of an Area of Freedom, Security and Justice: A Free Movement and Justice Rationale for Harmonisation? As mentioned, the Treaty of Amsterdam also introduced the objective of building an ‘Area of Freedom, Security and Justice (Area of Freedom, Security and Justice)’. This is listed in Article 2 TEU (consolidated in Amsterdam) and then

33 Joint Action/96/443/JHA of 15 July 1996 adopted by the Council on the basis of Article K.3 of the Treaty on European Union, concerning action to combat racism and xenophobia [1996] OJ L185/5. 34 Joint Action 97/154/JHA of 24 February 1997 adopted by the Council on the basis of Article K.3 of the Treaty on European Union concerning action to combat trafficking in human beings and sexual exploitation of children [1997] OJ L63/2. 35 Joint Action 98/733/JHA of 21 December 1998 adopted by the Council on the basis of Article K.3 of the Treaty on European Union, on making it a criminal offence to participate in a criminal organisation in the Member States of the European Union [1998] OJ L351/1. 36 Joint Action 98/742/JHA of 22 December 1998 adopted by the Council on the basis of Article K.3 of the Treaty on European Union, on corruption in the private sector [1998] OJ L358/2. 37 Convention C195/1 (n 16); and Convention drawn up on the basis of Article K.3 of the Treaty on European Union, on the protection of the European Communities’ financial interests [1995] OJ C316/49. 38 For an exhaustive list of all the adopted instruments, see Miettinen (n 27) 34–37.

102  EU Competences on Securitised Criminalisation in Article 29 TEU (consolidated in Amsterdam) placed in Title VI of the Treaty entitled ‘Judicial Cooperation’, which also includes the EU competence to harmonise definitions of crimes. This novel ‘legal brand’ covers a range of policy areas in the field of justice and home affairs on which a significant volume of EU hard and soft law already existed under the previous informal intergovernmental arrangements and later the Treaty of Maastricht, of which the post-Amsterdam developments are a direct continuation.39 The introduction of this objective nonetheless had an important symbolic and polity-building meaning for the EU, in that, according to Fichera, it re-interprets Europe ‘not merely as a legal and political, but mainly as a moral space’.40 In the Treaty of Amsterdam, the EU makes the claim ‘that it represents a common or shared commitment to realise the goods of freedom, security and justice … the Area of Freedom, Security and Justice symbolises a common political experience and a sense of civic interest by claiming that the EU is to be a place of freedom, security and justice’.41 The Treaty of Amsterdam does not define the term ‘freedom, security and justice’, which contains rather broad and multi-faceted concepts. This has led some authors to refer to this objective as an ‘aspirational tagline’ more than anything.42 Nonetheless, this book argues that the introduction of this broad constitutional objective still has implications for the function of the harmonisation of criminal law. Insights on the meaning of these terms have come from the multi-annual programmes accompanying the adoption of the Treaties and from the resulting scholarly discussion. While EU criminal policy documents are more extensively discussed in Chapter 6, a brief analysis of the Tampere Programme is given here to interpret the triad of concepts introduced by the Treaty of Amsterdam. The security limb of the Area of Freedom, Security and Justice is probably the one which has been more widely discussed, being the most influential of the three dimensions, at least in the early stages.43 Security is a multi-faceted concept, relevant to EU justice and home affairs policies (internal security), to the EU external action in the context of the Common Foreign and Security

39 N Walker, ‘In Search of the Area of Freedom, Security and Justice: A Constitutional Odyssey’ in N Walker (ed), Europe’s Area of Freedom, Security and Justice (Oxford, Oxford University Press, 2004) 3. 40 M Fichera, ‘Sketches of a Theory of Europe as an Area of Freedom, Security and Justice’ in M Fletcher, E Herlin-Karnell and C Matera (eds), The European Union as an Area of Freedom, Security and Justice (Abingdon, Routledge, 2017) 34, 35. 41 AH Gibbs, ‘Reasoned “Balance” in Europe’s Area of Freedom, Security and Justice’ (2011) 17(1) European Law Journal 121, 125. 42 M Fletcher, R Lööf and B Gilmore, EU Criminal Law and Justice (Cheltenham, Edward Elgar, 2008), 20. 43 Fichera (n 40); F Geyer and E Guild (eds), Security vs Justice? Police and Judicial Cooperation in the European Union (Abingdon, Routledge, 2016); E Herlin-Karnell, ‘The Domination of Security and the Promise of Justice: On Justification and Proportionality in Europe’s “Area of Freedom, Security and Justice”’ (2017) Transnational Legal Theory 79.

A Utilitarian Legitimacy for EU Criminal Law   103 Policy, and to the EU internal market law as a possible justification for restriction to free movement.44 In the context of criminal justice, both the Tampere Programme45 and scholars46 have interpreted this concept as referring to ‘crime prevention’, which does not provide an indication on why the behaviours to be prevented are considered to be criminal. In other words, it does not tell us much in terms of deontological or utilitarian legitimacy in relation to EU criminal law. Alongside this, the references to freedom and justice arguably provide a constitutional foundation for two other rationales for harmonisation: the ‘free movement’ rationale (ensuring that all Member States have a similar criminal law to encourage free movement) and the ‘justice rationale’ (ensuring that all Member States have a similar criminal law so that EU citizens feel they are treated equally in different Member States). While the two concepts can be broad ones, the following discussion provides examples of possible restrictive interpretations. Both the free movement and the justice rationales imply a utilitarian understanding of criminal law. The introduction of new crimes is justified in light of the social engineering goals that penal law can achieve. This confirms the argument that in the Treaty of Amsterdam, the focus was very much on the objectives that criminal law could achieve (and thus on its utilitarian legitimacy) rather than on the interests it is meant to protect (that is, on its deontological legitimacy). i.  The Treaty of Amsterdam and the ‘Free Movement’ Rationale for Harmonisation This concept of freedom is as multi-faceted as security. The Treaty of Amsterdam mentions it, but does not define it. Yet, the Tampere Programme for the Area of Freedom, Security and Justice, adopted immediately after the Treaty of Amsterdam, provides some more specific indications. It can be mentioned here that the Tampere Programme states that ‘the Amsterdam Treaty is now to ensure that freedom, which includes the right to move freely throughout the Union, can be enjoyed in conditions of security and justice accessible to all’.47 The main focus is thus on freedom as ‘free movement’. This is consistent with the traditional justification and conceptualisation of the Area of Freedom, Security and Justice as a compensatory measure for the side-effects of opening internal borders. The argument is that in an internal market, criminals could

44 For an overview of the place of security within the EU legal order, see E Herlin-Karnell, ‘The EU as a Promoter of Preventive Criminal Justice and the Internal Security Context’ (2016) 17(2) European Politics and Society 215, 216. 45 Tampere European Council Conclusions, 15 and 16 October 1999, pts 2 and 40. 46 M Fichera, ‘Criminal Law beyond the State: The European Model’ (2013) 19(2) European Law Journal 174, 176. 47 Tampere Conclusions (n 45) pt 2.

104  EU Competences on Securitised Criminalisation travel as freely as goods. As a consequence, heightened external border controls and police and judicial cooperation to counteract potential rises in cross-border crime are necessary.48 Interestingly, a slightly broader understanding of freedom is included in the Action Plan to implement the Treaty of Amsterdam as ‘freedom to live in a law-abiding environment’.49 Fletcher et al nonetheless observe that the interpretation of freedom as ‘freedom from fear of crime’ is unlikely to have been the main focus of the Treaty of Amsterdam, as it is part of that preventive justice rhetoric, typical of the War on Terror, which post-dates the Treaty of Amsterdam.50 Fletcher et al accordingly excludes the idea that freedom in this context should be interpreted otherwise as, for instance, the absence of unjustified constraints imposed by the public authority, which was discussed in Chapter 1 under the label of ‘liberty’.51 This is unlikely to be what the Treaty drafters had in mind, since in the Treaty of Amsterdam, the focus was still mainly on the repressive side of criminal justice and on the security dimension of other justice and home affairs policies like migration.52 If anything, considerations on the relationship between the individual and the state are included in the reference to justice (discussed below), albeit narrowly intended.53 Thus, a fortiori, in order to give distinct meaning to each of the concepts mentioned in Article 29 TEU, freedom must be interpreted differently. Fichera similarly argues that the concept of freedom in the Treaty of Amsterdam mainly refers to ‘freedom of movement’ by reading the objectives of freedom and security jointly. He argues that the Area of Freedom, Security and Justice is pervaded by a security discourse. Accordingly, it is meant to perform two complementary functions: securing the functioning or smooth operation of the internal market (this being the interpretation of freedom as free movement) and ensuring a secure marketplace (referring to security, intended strictu sensu, as freedom from crime).54 If one agrees with this interpretation, the harmonisation of criminal law being listed as one of the objectives of the Area of Freedom, Security and Justice could be justified on the ground that it promotes the internal

48 See art 2 TEU, which connects the objective of the Area of Freedom, Security and Justice to that of free movement. On this, see, Wade (n 4) 24; J Monar, ‘The Dynamics of Justice and Home Affairs: Laboratories, Driving Factors and Costs’ (2001) 39(4) Journal of Common Market Studies 747, 754; S Douglas-Scott, ‘The Rule of Law in the European Union: Putting the Security into the Area of Freedom, Security and Justice’ (2004) 29(2) European Law Review 219. However, it is worth recalling that the connection between free movement and cross-border criminality has not been exhaustively proven from an empirical point of view; see Fletcher et al (n 42) 24. 49 See the Action Plan of the Council and the Commission on how best to implement the provisions of the Treaty of Amsterdam on an area of freedom, security and justice [1999] OJ C19/1, paras 6–7. 50 Fletcher et al (n 42) 22. 51 See extensively ch 1. 52 Fletcher et al (n 42) 21. 53 See s II.B.ii. 54 Fichera (n 40) 34.

A Utilitarian Legitimacy for EU Criminal Law   105 market freedoms of citizens. Article 29 TEU (consolidated in Amsterdam) and now Article 67(1) TFEU would thus provide a constitutional basis for the ‘free movement’ rationale of the harmonisation of criminal law. ii.  The Treaty of Amsterdam and the ‘Justice’ Rationale for Harmonisation The concept of ‘justice’ is also a multi-faceted one, which was extensively debated in philosophy, as well as legal science55 and EU law scholarship.56 The Tampere Programme speaks of an: ‘Area of justice, where people can approach courts and authorities in any Member State as easily as in their own … Judgements and decisions should be respected and enforced throughout the Union, while safeguarding the basic legal certainty of people and economic operators.’ In this context, it seems that justice should be intended as the ‘technical administration of justice’57 – that is, ensuring the correct functioning of criminal justice actors, courts and prosecutors, for instance, within the EU and ensuring citizens’ access to justice,58 which has little relevance to harmonisation of the definitions of crimes. Douglas-Scott expands this interpretation, connecting justice also to access to courts and accountability of Justice and Home Affairs players, especially Europol and Eurojust.59 Commentators have introduced other interpretations for the justice limb of the Area of Freedom, Security and Justice. As disclosed above, Fletcher et al suggest thinking of justice in this context as a balance between the dimensions of the repression of crime and specifically defence rights,60 while stressing that this was only a partial opening to the liberal dimension of criminal justice, and that the real momentum for rebalancing repression and defence came with the Treaty of Lisbon. Herlin-Karnell interprets the reference to justice in Article 67(3) TFEU in broader and more conceptual terms, closely relating to the concept of justifications for public action, which should be ensured through the respect of proportionality.61 Finally Vander Baken

55 For an account of the historical origins of the term and a summary of the relevant literature, see I Rakar and B Ticar, ‘Normative Principles in the System of Criminal Justice’ (2016) 24 European Journal of Crime, Criminal Law and Criminal Justice 135, 139–48. 56 On the interpretation of justice in other areas of EU law beyond art 67(1) TFEU, see broadly D Kochenov, G de Búrca and A Williams (eds), Europe’s Justice Deficit (Oxford, Hart Publishing, 2015). 57 The expression is from S Douglas-Scott, ‘The Problem of Justice in the European Union’ in J Dickson and P Eleftheriadis (eds), Philosophical Foundations of European Union Law (Oxford, Oxford University Press, 2012) 413, 415. 58 Fichera (n 46) 177. 59 Douglas-Scott enumerates the lack of accountability of these bodies as problematic from a rule of law perspective, which this author consider inherently connected to the concept of justice: S Douglas-Scott, ‘The Problem of Justice in the European Union’ in Dickson and Eleftheriadis (n 57) 437. 60 Fletcher et al (n 42) 49. 61 E Herlin-Karnell, The Constitutional Structure of Europe’s Area of ‘Freedom, Security and Justice’ and the Right to Justification (Oxford, Hart Publishing, 2019).

106  EU Competences on Securitised Criminalisation sees justice as equality of treatment, namely the application of the same rules to all EU citizens across the EU.62 He explicitly refers to harmonisation of definitions of crimes as one of the legal strategies to reach the EU Treaty objective of ‘justice’.63 If one agrees with this interpretation, Article 29 TEU would then be the constitutional basis for the ‘justice rationale’ of harmonisation. III.  INTEGRATED LEGITIMACY FOR EU CRIMINAL LAW UNDER THE TREATY OF LISBON

The Treaty of Lisbon brought about an institutional revolution for EU action in criminal matters, since it eliminated the third pillar structure. But for a few exceptions,64 EU action in criminal matters is regulated by the ‘community method’, which includes, among other things, qualified majority in the Council, the involvement of the European Parliament as co-legislator and the full role of the Court of Justice.65 In terms of material scope of EU action, the amendments brought about by Lisbon are perhaps less revolutionary, but are still interesting from a legitimacy of EU criminal law perspective. The Treaty of Lisbon maintains the objective of an Area of Freedom, Security and Justice given in Article 67(1) TFEU. When looking at the Stockholm Programme, like the Tampere Programme, security is often mentioned in connection with citizens’ safety and crime prevention.66 The concept of justice is linked to the broad objective of the rule of law67 and to access to justice.68 This replicates the interpretation of the concept in Tampere, which was thus not directly connected to justice as equality of treatment to be achieved through the harmonisation of definitions of crimes. The Treaty thus cannot be interpreted as providing an explicit constitutional stronghold for the justice rationale for harmonisation of criminal law, namely as a way to obtain equal treatment for all citizens.

62 See T vander Beken, ‘Freedom, Security and Justice in the European Union: A Plea for Alternative Views on Harmonisation’ in A Klip and H van der Wilt (eds), Harmonisation and Harmonising Measures in Criminal Law (Amsterdam, Koninklijke Nederlandse Akad van Wetenschappen, 2002) 95–96. 63 As discussed by Elholm, this rationale has also been used for the harmonisation of criminal penalties: Thomas Elholm, ‘Does EU Criminal Cooperation Necessarily Mean Increased Repression?’ (2009) 17 European Journal of Crime, Criminal Law and Criminal Justice 191, 223–24; A Weyembergh, ‘The Functions of Approximation of Penal Legislation within the European Union’ (2005) 12 Maastricht Journal of European and Comparative Law 149, 165–67; G Vermeulen, ‘Where Do We Currently Stand with Harmonisation in Europe?’ in Klip and van der Wilt (n 62) 65, 75. 64 See, for instance, art 86 TFEU. 65 Mitsilegas (n 5) 5–6. 66 European Council, The Stockholm Programme – An open and secure Europe serving and protecting citizens [2010] OJ C115/1, 4 (hereinafter the Stockholm Programme). 67 ibid 76. 68 ibid 21.

Integrated Legitimacy for EU Criminal Law under the Treaty of Lisbon   107 The Stockholm Programme then emphasises the need to ensure full enjoyment of free movement rights.69 It seems likely that, as in the Tampere Programme, the objective of freedom is to be interpreted as freedom to ‘move freely’. The Stockholm Programme does not directly connect free movement and harmonisation of criminal law. Yet, interestingly, the Commission Communication which would come later in 2011 explicitly on harmonisation of criminal law makes this connection.70 This arguably give some extra support to the interpretation of the mention of the goal of freedom in the Treaty as providing a constitutional stronghold for the ‘free movement’ rationale as a legitimate one for the harmonisation of criminal law. This aspect is further discussed in Chapter 6 on EU criminal policy. Next to the general objective of the Area of Freedom, Security and Justice, the TFEU then includes a number of provisions mentioning different areas of crime, which scholars and the Commission have considered as a possible legal basis for EU securitised criminalisation. However, the following discussion demonstrates that Article 83 TFEU, which explicitly mentions that the harmonisation of the definitions of crimes should be lex specialis for the adoption of criminal law, with respect to all other provisions (see section III.A). The extent of the EU competence to legislate in the field of securitised criminalisation has not been radically amended; conversely, it is in continuation with the areas of crime where the EU has legislated under the Treaties of Maastricht and Amsterdam. However, the Treaty of Lisbon innovatively includes some explicit criteria limiting the boundaries of EU action, namely the serious (and cross-border nature (see section III.B) of the crime, which provide useful indications as to the envisaged legitimacy of criminal law. The inclusion of a seriousness threshold in particular arguably acknowledges a values-based criminalisation rationale for harmonisation and thus the need for deontological justifications for EU criminal law. Moreover, the position of Article 83(1) TFEU in the Treaty shows us that the cooperation rationale, which implies a utilitarian legitimacy for EU criminal law, has a less central place under the Treaty of Lisbon. A.  EU Criminalisation Competence in the Treaty of Lisbon: An Overview Article 83 TFEU, the main and, as demonstrated below, only EU legal basis for harmonisation of definitions of crimes, is made up of three paragraphs. The first two paragraphs describe the extent of the EU competence. The third paragraph provides some special institutional arrangements, traditionally referred to 69 ibid para 2.2. 70 European Commission, ‘Towards an EU Criminal Policy: ensuring the effective implementation of EU policies through criminal law’ COM (2011) 573 final, para I. See more extensively on this Communication ch 6, s IV.B.

108  EU Competences on Securitised Criminalisation as the ‘emergency brake’, which allow one Member State to halt the legislative procedure and basically pull out of the measure which is being negotiated:71 (1) The European Parliament and the Council may, by means of directives adopted in accordance with the ordinary legislative procedure, establish minimum rules concerning the definition of criminal offences and sanctions in the areas of particularly serious crime with a cross-border dimension resulting from the nature or impact of such offences or from a special need to combat them on a common basis. These areas of crime are the following: terrorism, trafficking in human beings and sexual exploitation of women and children, illicit drug trafficking, illicit arms trafficking, money laundering, corruption, counterfeiting of means of payment, computer crime and organised crime. On the basis of developments in crime, the Council may adopt a decision identifying other areas of crime that meet the criteria specified in this paragraph. It shall act unanimously after obtaining the consent of the European Parliament. (2) If the approximation of criminal laws and regulations of the Member States proves essential to ensure the effective implementation of a Union policy in an area which has been subject to harmonisation measures, directives may establish minimum rules with regard to the definition of criminal offences and sanctions in the area concerned. Such directives shall be adopted by the same ordinary or special legislative procedure as was followed for the adoption of the harmonisation measures in question, without prejudice to Article 76. (3) Where a member of the Council considers that a draft directive as referred to in paragraph 1 or 2 would affect fundamental aspects of its criminal justice system, it may request that the draft directive be referred to the European Council. In that case, the ordinary legislative procedure shall be suspended. After discussion, and in case of a consensus, the European Council shall, within four months of this suspension, refer the draft back to the Council, which shall terminate the suspension of the ordinary legislative procedure. Within the same timeframe, in case of disagreement, and if at least nine Member States wish to establish enhanced cooperation on the basis of the draft directive concerned, they shall notify the European Parliament, the Council and the Commission accordingly. In such a case, the authorisation to proceed with enhanced cooperation referred to in Article 20(2) of the Treaty on European Union and Article 329(1) of this Treaty shall be deemed to be granted and the provisions on enhanced cooperation shall apply.

The two first paragraphs are the most relevant for this discussion. Article 83(1) TFEU deals with EU securitised criminalisation and can be considered to be the ‘heir’ of the third pillar.72 The scope of this provision is further discussed below. Article 83(2) TFEU conversely represents a partial codification of the case law of the Court of Justice on first pillar criminal law, dealing with functional criminalisation,73 which will be discussed in Chapter 5.

71 Mitsilegas 72 ibid

58. 73 ibid.

(n 5) 6.

Integrated Legitimacy for EU Criminal Law under the Treaty of Lisbon   109 Alongside Article 83 TFEU, there are a number of other provisions which do not explicitly mention criminal law, but whose potential as a legal basis for EU criminalisation has been discussed. These include Articles 86 and 325(4) TFEU and on the protection of the financial interests of the EU, Article 79 TFEU on trafficking in human beings and two nearly open-ended provisions, namely the internal market legal basis (Article 114 TFEU) and the flexibility clause (Article 352 TFEU). Yet, no criminal law legislation has been proposed on these legal bases so far and it is argued here that Article 83 TFEU should also be considered as lex specialis with respect to each of them in the future. In more detail, Article 325(4) TFEU establishes that: The European Parliament and the Council, acting in accordance with the ordinary legislative procedure, after consulting the Court of Auditors, shall adopt the necessary measures in the fields of the prevention of and fight against fraud affecting the financial interests of the Union with a view to affording effective and equivalent protection in the Member States and in all the Union’s institutions, bodies, offices and agencies.

This provision grants the competence to the EU to adopt measures in the field of prevention of fraud. For a number of historical and textual reasons, part of the scholarship has interpreted the provision as giving the competence to also adopt criminal law measures. The main argument was that the pre-Lisbon version of the Treaty explicitly ruled out the possibility of having criminal measures adopted on this provision. Article 280(4) TEC (Consolidated in Amsterdam) stated ‘The Council … shall adopt the necessary measures in the fields of the prevention of and fight against fraud affecting the financial interests of the Community with a view to affording effective and equivalent protection in the Member States. These measures shall not concern the application of national criminal law or the national administration of justice.’ The Commission had already proposed criminal law legislation on this Amsterdam legal basis, arguing that the application of national criminal law, interpreted as the actual administration of justice, was not directly affected, as the EU instruments would only define the substantive rules.74 However, the proposal was not supported in the Council.

74 ‘Proposal for a directive of the European Parliament and of the Council on the criminal-law protection of the Community’s financial interests’ COM (2001)272 final 2001/0115 (COD), 5. ­Tiedemann and Delmas-Marty had pleaded in favour of having criminal law adopted on the basis of this provision, arguing that national criminal law would be left untouched, and thus the requirement of the provision would be respected, if the selected criminal law instrument is only a directive, which in any case requires an implementation instrument. Conversely, Spencer, and Grasso to a certain extent, opposed this view. See the contributions by these authors in M Delmas-Marty and J Vervaele, The Implementation of the Corpus Juris in the Member States, vol 1 (Antwerp, Intersentia, 2000). For a broader view which includes the possibility of having criminal law directives, but not regulations, enacted on this basis, see R Sicurella, ‘Setting up a European Criminal Policy for the Protection of EU Financial Interests: Guidelines for a Coherent Definition of the Material Scope of the European Public Prosecutor’s Office’ in K Ligeti (ed), Toward a Prosecutor for the European Union: A Comparative Analysis – Vol 1 (Oxford, Hart Publishing, 2012) 870, 877–78.

110  EU Competences on Securitised Criminalisation Given that the mention of the exclusion of criminal law had disappeared in the Treaty of Lisbon, authors have inferred that the provision in its current form can be a legal basis for EU criminal law,75 including regulations, given that it only mentions ‘measures’.76 It was also observed that since Article 325(4) TFEU deals specifically with the fight against fraud, it should be lex specialis for any measures (including criminal law ones) in this policy area with respect to the general legal basis in Article 83 TFEU.77 Finally, the language of the provisions and especially the term fight against fraud were held up in support of construing Article 325(4) TFEU as a potential criminal law legal basis.78 Yet this interpretation was not unanimously accepted. Asp and other scholars challenged the idea that the elimination of the criminal law sentence in the Lisbon version of the provision was only meant to pre-emptively rule out a far-fetched interpretation post-Lisbon that no criminal law could be adopted in the field of fraud against the EU budget at all, not even on the specific Lisbon criminal law competences (Article 83 TFEU).79 Interestingly, the Commission initially endorsed the criminal law interpretation of Article 325(4) TFEU and proposed the Directive for the protection of the financial interests by means of criminal law on this legal basis.80 Interestingly, when advocating Article 325(4) as the legal basis, the Commission also relied on some considerations on the function of criminal law, as will be elaborated upon further in Chapter 7. However, the European Parliament81 and the European Council82 did not share this position, and the text was adopted on the basis of Article 83(2) TFEU.83 Both institutions highlighted that adopting the text on the basis of Article 325(4) TFEU would have made it possible to circumvent the special institutional arrangements provided by Article 83(2) TFEU, such as the opt-outs84 and the possibility for one Member State to halt

75 Sicurella (n 74) 896. 76 H Satzger, International and European Criminal Law (Oxford, CH Beck/Hart Publishing, 2012) 56. 77 Mitsilegas (n 5) 67. 78 P Asp, The Substantive Criminal Law Competence of the EU (Stockholm, Stiftelsen Skrifter utgivna av Juridiska fakulteten vid Stockholms universitet, 2012) 142, which includes a summary for other additional teleological and constitutional arguments in favour of interpreting art 325(4) TFEU as a criminal law competence (at 146, 149). 79 These arguments are exhaustively summarised in ibid 150, 151. 80 Commission, ‘Proposal for a Directive of the European Parliament and the Council on the protection of the Union’s financial interests by means of criminal law’ COM (2012)363 final, 6. 81 European Parliament legislative resolution of 16 April 2014, P7_TA(2014)0427. 82 Council Legal Service Document 8604/15. Available at: www.statewatch.org/news/2015/may/ eu-council-fraud-dir-trilogue-8604-15.pdf. 83 Directive 2017/1371/EU of the European Parliament and of the Council of 5 July 2017 on the fight against fraud to the Union’s financial interests by means of criminal law [2017] OJ L198/29. 84 The UK and Ireland can decide on an ad hoc basis which measures of the Area of Freedom, ­Security and Justice in which they wish to participate; see Protocol (No 21) on the position of the United Kingdom and Ireland in respect of the area of freedom, security and justice [2016] OJ C202/295. Conversely, Denmark has an automatic opt-out from each police and judicial cooperation measure adopted pursuant to Title V of pt III of the TFEU, while it can negotiate them on a case-by-case basis; see Protocol (No 22) on the position of Denmark [2012] OJ C326/299. Adopting criminal law

Integrated Legitimacy for EU Criminal Law under the Treaty of Lisbon   111 the legislative procedure by pulling the above-mentioned ‘emergency brake’.85 Given these legislative developments, it seems unlikely that the provision will be relied upon in the future in terms of adopting criminal law legislation.86 A similar debate surrounded Article 86(2) TFEU. This provision gives the competence to the Council to establish, via a regulation, the EPPO.87 The provision defines the mission of the EPPO of prosecuting the offences against the EU’s financial interests, ‘as determined by the regulation that establishes it’ (emphasis added).88 This wording led authors in the past to debate whether the EU had competence to also define crimes on this legal basis by including such definitions directly in the EPPO Regulation. Ligeti argued that the EPPO needs a uniform definition of its material scope of action for pragmatic reasons and it cannot work within a framework of diversified understanding of ‘crimes affecting the financial interests of the EU’, as would result from a scenario leaving such definition to a directive or even directly to national legislators.89 Conversely, Sicurella suggested that ‘determined’ had to be interpreted as ‘indicated’ rather than ‘defined’.90 The EPPO Regulation should have contained a list of areas of crimes defining its competence, based on the model of the Framework Decision on the European Arrest Warrant, but the specific offences falling within each measures on a different legal basis from those in the Area of Freedom, Security and Justice, would automatically bind the UK, Ireland and Denmark, arguably bypassing their will to regulate their participation to integration in the field of criminal justice on a more ad hoc basis. 85 Article 83(3) TFEU. 86 Interestingly, there is another provision within the Treaty that has a similar history to that on EU fraud. Article 135 TEC, consolidated in Amsterdam [1997] OJ 340/02, concerning customs cooperation, similarly included a reservation stating that no measures could impact on the application of national criminal law, which disappeared in the Lisbon version (art 33 TFEU). This provision is not traditionally discussed among the potential criminal law legal basis in the Treaty, not least because the subject matter – customs cooperation – has a much weaker connection with the criminal field than fraud against the EU budget. For a more detailed discussion, see Asp (n 78) 160. 87 The idea of a European Public Prosecutor was also at the centre of the Corpus Juris project financed by the Commission and led by Mireille Delmas-Marty; see M Delmas-Marty, Corpus Juris Introducing Penal Provisions for the Purpose of Protecting the Financial Interests of the European Union (Paris, PUF Economica, 1997). The competence to establish an EPPO was already included in the Constitutional Treaty: art III-274 of the Treaty Establishing a Constitution for Europe [2004] OJ C310/01. 88 Article 86(4) TFEU also includes the possibility of amending the scope of action of the EPPO by means of a Council Decision so as to include serious crimes having a cross-border dimension. However, it seems quite clear that the Council Decision amending the EPPO scope of action would only contain the enumeration of such crimes, which would have been defined elsewhere; it would not be a harmonisation instrument. See Mitsilegas (n 5) 67. 89 K Ligeti, ‘Approximation of Substantive Criminal Law and the Establishment of the European Public Prosecutor’s Office’ in Weyembergh and Galli (n 30) 73, 81. See similarly, but with reference to provisions in the Constitutional Treaty, A Bernardi, ‘Europeizzazione del diritto penale e progetto di costituzione europea’ (2004) Diritto Penale e Processo 5, 8 ff; L Piccotti, ‘Il Corpus juris 2000. Profili di diritto penale sostanziale e prospettive di attuazione alla luce del progetto di Costituzione per l’Europa’ in L Piccotti (ed), Il Corpus juris 2000. Nuova formulazione e prospettive di attuazione (Padua, Cedam, 2004) 85. 90 Sicurella (n 74) 895. See also, contra, G Grasso, Il Trattato di Lisbona e le nuove competenze penali dell’Unione europea, in Studi in onore di Mario Romano, vol 4 (Naples, Jovene, 2011) 2308, 2344–46.

112  EU Competences on Securitised Criminalisation area should be determined elsewhere. Sicurella in particular raises the question of the less democratic procedure foreseen by Article 86 that requires only the consent of the European Parliament. This would make it a less suitable procedure for enacting criminal law, as it would clash with the principle of legality in its substantial form. Eventually, the material scope of action for the EPPO was designed in a separate instrument, namely the Directive on the protection of the financial interests of the EU by means of criminal law,91 which ruled out the possibility of Article 86(2) TFEU-based criminal law. Article 79 TFEU was also mentioned as a possible lex specialis with respect to Article 83(1) TFEU, considering that it gives the EU the competence to adopt ‘measures to combat … trafficking in human beings’, and it does not specify if these need to be administrative or criminal.92 However, no legislation harmonising criminal law was proposed on this provision. It seems reasonable to believe that Article 83 TFEU, which explicitly includes trafficking in human beings among the relevant policy areas, should be lex specialis when criminal law is involved. Conversely, Article 83 TFEU did not explicitly mention fraud, hence the debate summarised above on whether this provision or Article 325(4) TFEU should be lex specialis on the criminalisation of fraud.93 Finally, Herlin-Karnell and Ioncheva have discussed whether Articles 114 TFEU and 352 TFEU could work as a criminal law legal basis.94 This is because both are very broad provisions: Article 114 grants the competence to approximate national legislation regulating the market and Article 352 grants the competence to adopt the specific measures necessary to achieve one of the objectives of the Treaties, even in the absence of the explicit competence. However, also in this case, similar institutional arguments mentioned above with respect to Article 325(4) TFEU rule out these two provisions as a potential criminal law legal basis, leaving Article 83 TFEU as lex specialis.95

91 Commission, ‘Proposal for a Directive of the European Parliament and the Council on the protection of the Union’s financial interests by means of criminal law’ COM (2012)363 final. 92 Compare art 79 TFEU with art 75 TFEU, which similarly deals with a specific area of crime, namely terrorism. Article 75 TFEU nonetheless clarifies that its scope is only limited to administrative measures. For a discussion on art 75 TFEU and its relationship with the criminal law provisions of the Treaty, see E Herlin-Karnell, The Constitutional Dimension of EU Criminal Law (Oxford, Hart Publishing, 2012) 84. 93 On the point, see largely Asp (n 78) 142; and Sicurella (n 74) 897. 94 Herlin-Karnell observes that given the previously very broad interpretation of the internal market legal basis, there are reasons to believe that art 114 TFEU has not been wiped off the criminal law agenda: Herlin-Karnell (n 92) 109. Moreover, Turner discusses whether art 352 TFEU could be used to adopt criminal law legislation on combating racism and xenophobia. The Treaty on European Union lists combating racism and xenophobia among the objectives of the Treaty. However, it does not include an explicit competence to adopt legislation in this area. Thus, the possibility exists in principle, but given the controversial character of art 352 TFEU, it is unlikely that this provision will be used in practice. See also JI Turner, ‘The Expressive Dimension of EU Criminal Law’ (2012) 60 American Journal of Comparative Law 555, 558. 95 Satzger (n 76) 54; and Herlin-Kernell (n 92) 87 and 109.

Integrated Legitimacy for EU Criminal Law under the Treaty of Lisbon   113 This book maintains this perspective and considers Article 83(1) TFEU as the sole legal basis for adopting the harmonisation of definitions of crimes, mainly on the basis of the institutional arguments mentioned above. Having said that, Mitsilegas points out an inherent tension between this interpretation and the Treaty objective of ensuring the effective enforcement of EU law (which is more extensively discussed in Chapter 5), which is worth recalling.96 Article 83(2) TFEU provides the EU with the competence to adopt criminal law in order to ensure the effectiveness of EU policies, and the objective of the effective enforcement of EU law is given particular visibility in the Lisbon framework.97 In light of this, it seems paradoxical to prevent resorting to other legal bases to adopt EU criminal law, outside the title on judicial cooperation, which could actually serve the end goal of effective EU law. In other words, while the interpretation of the extent of EU competence in criminal matters based on institutional arguments is convincing, there is at least some tension with a theological interpretation of the EU scope of action in light of the objectives of the Treaties. B.  Article 83(1) as an Example of Integrated Legitimacy for EU Criminal Law Article 83(1) TFEU, which as established above is the lex specialis for securitised EU criminal law, does not explicitly mention any function for the harmonisation of substantive criminal law; it only identifies the material boundaries of EU action in this field. It establishes that the EU only has the competence to harmonise the definitions of crimes in areas of particularly serious and crossborder crimes, and it includes a list of 10 relevant areas, which can be expanded through a Council decision. The material scope for EU action roughly coincides with that available under the Treaty of Amsterdam,98 and it is fair to assume that already under Amsterdam, the rationale for EU legislation in these areas of crime was to a great extent their transnational character, which required a joint effort from all EU Member States.99 However, the explicit acknowledgement of some general criteria limiting EU action is still meaningful on at least two levels. First, on a practical level, the two criteria bind the Council in the possible

96 Mitsilegas (n 5) 68. 97 On this, see extensively ch 5. 98 In the Treaty of Lisbon, there is only a newly mentioned area of crime (money laundering) where there had not been EU criminal law legislation beforehand, while one area where there has been criminal law in the past (racism and xenophobia) is not included. See L Arroyo Zapatero and M Muñoz de Morales Romero, ‘Droit pénal européen et Traité de Lisbonne: le cas de l’harmonisation autonome (article 83.1 TFEU)’ in G Giudicielli-Delage and L Christine (eds), Le Droit Pénal de l’Union Européenne au Lendemain du Traité de Lisbonne (Paris, Société de Législation Comparée, 2012) 113, 122. 99 Notably, the focus on cross-border criminalisation was recalled in the multiannual programs. On this, see extensively ch 6.

114  EU Competences on Securitised Criminalisation expansion of the list of crimes that the EU could harmonise beyond the list of 10 areas.100 The Council cannot add minor crimes to the list which traditionally do not occur in a cross-border manner. Moreover, arguably, each single definition of crime included in legislation enacted on the basis of Article 83(1) TFEU in the areas of crime already listed in the Treaty must respect the two criteria. Admittedly, a literal interpretation of Article 83(1) TFEU might suggest that the listed 10 areas of crime already meet these standards. Indeed, Article 83(1) TFEU first grants the competence in areas of crime that are serious and cross-border in nature, and then states: ‘These areas are …’ (emphasis added).101 Yet, the listed areas of crime in which the EU has competence areas are very broadly defined. For example, organised crime or computer crime can encompass a wide variety of criminal conduct. Therefore, some guidance on which conduct falling within these areas deserves harmonisation is necessary. The seriousness and cross-border criteria are there to provide such guidance.102 Second, and more importantly for our discussion, the mention of seriousness and cross-border criteria has conceptual implications hinting at the functions for, and thus the legitimacy of, EU criminal law. Long observes that ‘from a “no criterion” situation pre-Lisbon, we have clearly entered into a new era, namely an attempt to explain the harmonisation purposes of substantial criminal law’.103 The following paragraph illustrates in more detail that the explicit mention of the seriousness criterion acknowledges a criminalisation values-based rationale for harmonisation, implying a deontological justifications for EU criminal law. Evidence of a complementary cooperation rationale for harmonisation, pointing in the direction of a utilitarian legitimacy for EU criminal law, can still be indirectly linked to the cross-border criterion included in Article 83(1) TFEU. However, the position of the provision in the Treaty shows us that the link between harmonisation and cooperation is nuanced in Lisbon. i.  The Seriousness Criterion: Deontological Legitimacy for EU Criminal Law The Treaty of Maastricht already limited areas of EU interests to ‘serious’ crime. Yet this limitation did not refer to harmonisation, given that this strategy was

100 Alix interestingly speaks of the internal and external boundaries of art 83(1) TFEU. The former relate to the meaning to be attributed to the broadly defined areas of crime, while the latter are the boundaries to the future expansion of the provision through the criteria of ‘cross-border dimension’ and ‘seriousness’. See Alix (n 15) 148. 101 See in this sense H Labayle, ‘Le Traité de Lisbonne et l’entraide repressive dans l’Union européenne’ (2007/2008) 2 Revue des Affairs Européens 209, 217. 102 Weyembergh (n 30) 9, 14; Asp (n 78) 83; Zapatero and Munoz de Morales Romero (n 98) 123; M Delmas-Marty, Les forces imaginantes du droit. Le relatif et l’universel (Paris, Seuil, 2004) 413; P De Hert, G Boulet and I Wieczorek, ‘Fondements et objectifs des incriminations au niveau Européen: le cas de la criminalité informatique’ in Bernard et al (n 31) 267, 273; A Klip, ‘EU Criminal Policy’ (2012) 20(1) European Journal of Crime, Criminal Law and Criminal Justice 3, 7; Sicurella (n 74) 58. 103 N Long, ‘Towards a European Criminal Law Code?’ (2011) 01(49) EipaScope Bulletin 49, 50.

Integrated Legitimacy for EU Criminal Law under the Treaty of Lisbon   115 not listed as being among the possible strategies to combat crime.104 Conversely, the Treaty of Lisbon lists it with reference to the harmonisation of definitions of crimes. However, the Treaty itself does not specify what is meant by ‘serious crime’, nor can any general definition be found in the national criminal legislation105 or EU secondary law.106 A recent case heard by the Court of Justice indirectly provides some insight. In PI v Oberburgermeisterin der Stadt Remscheid, the Court was faced with a citizenship-related question. The national court asked how to interpret the expression ‘imperative grounds of public security’ contained in Article 28(3) of the Citizens’ Rights Directive, which could justify the expulsion of foreign EU citizens.107 The facts of the case meant that it had to be determined whether the presence on the national territory of a foreign EU national condemned of sexual assault and rape of a minor might qualify as a direct threat to the calm and physical security of the population that would justify his expulsion on the grounds of public security. The Court first recalled that the offence at stake was among the areas included in the list in Article 83(1) TFEU; therefore, it was a serious crime. The Court continued that it is for the national court to then determine if the offences mentioned within Article 83(1) TFEU all constitute ‘a particularly serious threat to one of the fundamental interests of society, which might pose a direct threat to the calm and physical security of the population’.108 Briefly, the Court of Justice leaves the ultimate choice as to which crimes affect the fundamental interests of society to Member States. Yet, it also points out that the crimes listed in Article 83(1) TFEU could be good ‘candidates’. In simpler terms, while not affirming it directly, the Court seems to imply that Article 83(1) TFEU crimes potentially affect the fundamental interests of society. By limiting EU harmonisation powers only to those forms

104 See s II above. 105 For a comparative analysis of Member States’ criminal legislation, see Wade (n 4). 106 The Europol decision limits Europol’s scope of action only to serious crime. The decision does not provide a definition of what should be understood as ‘serious crime’. The text only refers to a list of crimes; art 4 of Council Decision 2009/371/JHA of 6 April 2009 establishing the European Police Office (Europol) [2009] OJ L121/37. This approach is not particularly helpful in narrowing down the concept of serious crime as established in art 83(1) TFEU, given that the list of crimes is significantly longer than the areas of crimes mentioned in the Treaty provision. Long argues that the definition of serious crimes in art 83(1) TFEU should have as a basis the offences listed in EU secondary norms, especially the Eurojust and Europol decisions: Long (n 103) 50. Yet, even though there must be some coherence between EU primary and secondary law, I believe there is a difference in terms of understanding ‘serious crime’ for the purposes of harmonising definitions of crimes (a more intrusive strategy) and defining the scope for action of an agency supporting police and judicial cooperation, or within instruments establishing a judicial cooperation mechanism (a less intrusive strategy). As a consequence, an interpretation of serious crime as covering all the crimes listed in the Europol and Eurojust decisions should be rejected. 107 Directive 2004/38/EC of the European Parliament and of the Council 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 L158/77. 108 Case C 348/09 PI v Oberbürgermeisterin der Stadt Remscheid [2012] ECR I-00000, para 28.

116  EU Competences on Securitised Criminalisation of behaviour that harm very important interests – what was called a valuesbased criminalisation rationale for harmonisation – Article 83(1) TFEU makes it clear that the legitimacy for EU criminal law should rest, among other factors, on deontological grounds. ii.  The Cross-border Criterion: An Indirect Reference to the Utilitarian Legitimacy of EU Criminal Law? Article 83(1) TFEU also limits the EU competence to harmonisation of crossborder crimes due to the nature or impact of the given offences or to a special need to combat them on a common basis. In order to give effet utile to each aspect of the provisions, the three expressions should be interpreted as referring to different situations. The ‘transnational impact of the given offences’ intuitively refers to crimes that might be perpetrated nationally, but whose harmful consequences spread across borders. The typical example is environmental crimes,109 as well as corruption, which, even if perpetrated nationally, can have a cross-border economic impact.110 Similarly, acts of terrorism can be carried out nationally, but with an aim to spread terror as broadly as possible and thus naturally across borders.111 This criterion requires us to look at the effect of crimes, which by being cross-border can be better assessed by the EU than the Member States alone, and thus indirectly at the interests that criminal law should protect. It thus suggests considering a values-based criminalisation rationale for the harmonisation of criminal law. The ‘transnational nature’ is a less straightforward criterion, but it can be interpreted as referring to the way in which a crime is perpetrated. For instance, crimes that are traditionally perpetrated through a transnational criminal network, or through the Internet, which is borderless, can be said to be cross-border in nature.112 When crimes are carried out through a transnational network, national responses are insufficient to disrupt them, and Member States need to coordinate and pool their efforts.113 EU action is needed to design a framework for interstate cooperation to fight cross-border crime. Harmonisation in this context can play a role to the extent that it facilitates such cooperation. The reference to the cross-border nature of the crime would thus indirectly connect harmonisation to cooperation, acknowledging the cooperation rationale and thus a utilitarian legitimacy for EU criminal law.

109 Zapatero and Munoz Morales de Romero (n 98) 128. 110 Mitsilegas (n 5) 59. 111 This is the case for international terrorism, while of course there can also be cases of local terrorism carried out, for instance, by separatist movements. On the definition of terrorism, see B Saul, Defining Terrorism in International law (Oxford, Oxford University Press, 2008), especially ch 1. 112 Zapatero and Munoz de Morales Romero (n 98) 129. 113 However, this case is different from the hypothetical scenario sketched above, namely where crimes are orchestrated nationally and can thus be fought by national authorities, but cause harm beyond national territories.

Integrated Legitimacy for EU Criminal Law under the Treaty of Lisbon   117 Finally, the last requirement – considering crimes to be cross-border when there is a special need to combat them on a common basis – seems to be a circular argument that common action (in this case harmonisation) is needed when these offences need to be fought by joining forces. Authors have interpreted this category in different ways. Zapatero and Muñoz de Morales Romero construe it as referring to the need to use harmonisation to support judicial cooperation in the fight against crime.114 They thus find a reference to the cooperation rationale in this part of the provision. Turner connects this requirement to the symbolic potential of harmonisation. She argues that the ‘need’ to combat crimes on a common basis, through adopting an EU-wide definition of crimes, can stem from the need to use criminal law to re-affirm the EU normative identity, stressing what values it stands for, although she admits that this might be an unduly expansive interpretation.115 In this case, the wording of Article 83(1) TFEU would refer to the socialising rationale for harmonisation discussed in Chapter 3. Briefly, the cross-border requirement in Article 83(1) TFEU suggests a valuesbased criminalisation rationale, a cooperation rationale, either linked to the question of the nature of the crimes or the need to combat them on a special basis, and possibly a socialising rationale. For the purposes of our analysis, it is interesting that in stressing the importance of focusing on cross-border crimes, the Treaty of Lisbon seems to indirectly suggest that utilitarian considerations should also be factored in when considering whether to adopt EU criminal law. Incidentally, the socialising rationale can also imply both a deontological and a utilitarian legitimacy for EU criminal law. However, the importance of the link between harmonisation and cooperation within the Treaty, and the need for utilitarian justifications of EU criminal law, should not be overestimated. When looking at the position of Article 83(1) TFEU in the title on judicial cooperation, it seems clear that harmonisation is given a more autonomous function with respect to cooperation than was the case in the past. iii.  The Place for Article 83(1) TFEU in the Treaty: A More Autonomous Role for Harmonisation Article 67(1) TFEU, the Lisbon provision corresponding to the former Article 29 TEU, sets the objective of the Area of Freedom, Security and Justice. Then in paragraph 3, it lists the strategies to ensure a high level of security. As in the Amsterdam framework, these are measures to prevent and combat crime, racism and xenophobia, judicial and police cooperation, mutual recognition of judgments in criminal matters and, again if necessary, the approximation of



114 Zapatero 115 Turner

and Munoz de Morales Romero (n 98) 127. (n 94).

118  EU Competences on Securitised Criminalisation criminal laws. Harmonisation thus remains in a ‘default’ position with respect to judicial cooperation and mutual recognition. Moreover, Article 83 TFEU dealing with harmonisation of definition of crimes is inserted into Chapter IV, which is devoted to judicial cooperation. Finally, Article 82(1) TFEU, which is the legal basis for judicial cooperation, affirms that cooperation shall include harmonisation on the basis of Article 82(2) TFEU, dealing with procedural criminal law, as well as of Article 83 TFEU, dealing with substantive criminal law, namely definitions of crimes. Article 82(2) TFEU then delimits the EU competence to harmonise procedural rules, for instance, rules on the right of the defendant, or rights of victims during the criminal trial. Zapatero and Muñoz de Morales Romero, among others, have interpreted this institutional setting as suggesting that harmonisation remains in a subordinate position and should be resorted to as a means to an end only to support judicial cooperation.116 Nonetheless, and contrary to what happened in the Treaty of Amsterdam, under the Treaty of Lisbon, the harmonisation of substantive criminal law is granted an ad hoc independent provision, Article 83 TFEU, which does not mention judicial cooperation or mutual recognition at all. Moreover, the Final Report of Working Group X on ‘Freedom, Security and Justice’, one of the working groups which composed the Convention on the Future of Europe, the body entrusted with the drafting of the Lisbon Treaty, had called for further consideration to the explicit inclusion of the ‘cooperation rationale’ for harmonisation of the definition of crimes in the text of the Treaty. It suggested adding a competence to harmonise substantive criminal law, which includes the definitions of crimes, ‘when approximation is required to generate sufficient mutual confidence to enable the full application of mutual recognition of judicial decisions or to guarantee the effectiveness of common tools for police and judicial cooperation created by the Union’.117 However, the Treaty drafters decided not to include it in the current version of the provision. This suggests that the Lisbon framework envisages harmonisation as partially more autonomous from cooperation, a choice which, as mentioned earlier, was deemed to be revolutionary by some in comparison to the previous Amsterdam framework.118 Moreover, Article 82(2) TFEU, which sets the competence for harmonisation of procedural law, explicitly mentions the link with mutual recognition. This can be interpreted as implying, a contrario, that the harmonisation of substantive criminal law based on Article 83 TFEU, which does not include such a mention, should be independent from cooperation. Sicurella in particular calls the collocation

116 Zapatero and Munos Morales de Romero (n 98) 127; E Rubi-Cavagna, ‘Le domaine et les methods de l’harmonisation autonome’ in Giudicielli-Delage and Lazerges (n 15) 141, 143; Labayle (n 101) 209, 216. 117 CONV 426/02, Brussels, 2 December 2002, WG X 4. 118 Weyembergh (n 30) 21; J Tricot, ‘L’harmonisation pénale accessoire: question(s) de method: observations sur l’art et la manière de légiférer pénalement selon l’Union Européenne’ in GiudicielliDelage and Lageres (n 15) 185, 190.

Conclusions  119 of Article 83 TFEU in the Treaty ‘critical’,119 whereas Bernardi describes the formulation of Article 82 TFEU, which includes the harmonisation of substantive criminal law as a tool to achieve mutual recognition, as unfortunate.120 Incidentally, Alix notes that the fact that the harmonisation of procedural norms, granting rights to individuals, is subordinated to mutual recognition purposes, while the harmonisation of the substantive criminal law also has an autonomous function that expands the range of criminal law, is evidence of an illiberal approach to criminal law. The sword function of criminal law (to suppress crimes) can autonomously develop, while the shield function (to protect fundamental rights) is circularly subordinated to cooperation in suppress crimes. Nonetheless, still in some cases the texts containing norms of substantive criminal law sometimes also include procedural norms.121 This makes the distinction between these forms of harmonisation somehow blurred. Briefly, both the text of Article 83(1) TFEU and its position in the Treaty point to an existing link between harmonisation and cooperation, which, as was explained above, requires the factoring in of utilitarian considerations when deciding whether to resort to criminal law. However, this function of the harmonisation of criminal law is not as prominent as it was under the Treaty of Amsterdam. And actually, the insistence on serious crime tells us that under the Treaty of Lisbon, the legitimacy of EU criminal law must rest on an integrated utilitarian and deontological framework. The adoption of EU securitised criminal law is justified when important interests are at stake and when the use of harmonised criminal law also serves future goals, such as also securing enforcement of the law in cross-border cases. IV. CONCLUSIONS

This chapter has discussed the legitimacy Article 83(1) TFEU envisages for EU definitions of crimes when granting the EU with the competence to harmonise definitions of crimes in the field of ‘securitised criminalisation’. It has looked at both the definitions of competences and the criteria legitimating the exercise of competences that can be inferred from the position of the relevant legal basis in the Treaty. It has focused on the history of the values-based criminalisation rationale and the cooperation rationale, which played a major role in the field of EU securitised criminal law. The analysis here has shown that the approach to the legitimacy of EU criminal law evolved from a utilitarian approach focused on the enforcement and socialising potential of criminal law in the Treaties of Maastricht and Amsterdam to an integrated utilitarian and deontological



119 Sicurella

(n 74) 38 (n 89) 44. 121 For examples and a discussion, see Alix (n 15) 147, 148–49. 120 Bernardi

120  EU Competences on Securitised Criminalisation approach in the Treaty of Lisbon. The latter approach requires justification of resorting to criminal law not only in light of the future objectives that criminal law can achieve, but also of the interests it protects. The acknowledgement in the Treaty of Lisbon of complementary deontological rationales for EU criminal law, alongside the traditional utilitarian rationales, is particularly important in terms of internal coherence of the EU legal order. Chapter 2 discussed the normative standards that the EU constitutional principles and values set for the legitimacy of EU criminal law. It explained that, on the one hand, the principle of subsidiarity requires EU action to be efficient in achieving EU policy objectives. This was interpreted to mean that resorting to EU criminal law needed to be justified on utilitarian grounds, namely on its potential to secure effective enforcement of the law in general and especially in cross-border cases, where harmonised definitions of crimes are necessary to ensure smooth judicial cooperation. As has been mentioned above, the position of the provisions on harmonisation within the Treaty of Lisbon, among other factors, has traditionally suggested that cooperation should be a core rationale for the exercise of these competences. Though this aspect is attenuated in Lisbon, limiting EU competence only to cross-border crime was also interpreted as connecting harmonisation to cooperation. On the other hand, the Treaty of Lisbon acknowledged fundamental rights as EU founding values, giving particular visibility to the EU values-based identity. It is thus not surprising that the deontological acknowledgement of the valuesbased dimension of criminal law was introduced with the same institutional reform. Limiting EU competences only to serious crime ensures greater coherence with the EU constitutional legal order. Interestingly, the same cannot be said for the field of EU functional criminalisation, where conversely the Treaty of Lisbon identifies a stand-alone regulatory rationale for the harmonisation of criminal law and therefore a utilitarian, effectiveness-based, rationale for EU criminal law. This will be examined in the next chapter.

5 EU Competences on Functional Criminalisation The Route to Utilitarian EU Criminal Law

A

longside EU ‘securitised criminalisation’, the EU legal order has also progressively acquired a ‘functional criminalisation’ competence. This is the competence to resort to criminal law to prevent its own ‘victimisation’, namely to safeguard EU policies, objectives and interests.1 However, the route to establish this competence was a tortuous one. Vervaele points out that the founding fathers ‘overlooked the importance of the enforcement of Community law’.2 As a consequence, in the early stages of EU integration, enforcement of core EU policies, such as common agricultural and fishery policy, the Community customs code, European financial services regulations, the European subsidy fraud rule, European environmental policy and European rules on corporate law, was entirely left in the hands of the Member States, with no EU rule regulating this crucial aspect.3 When it became clear that this legislative gap could actually jeopardise the effective implementation of EU policies, the need to set clear obligations for Member States to ensure citizens’ compliance with EU law started to receive more attention by the Commission and the Court of Justice. A key question in this debate was what role criminal law should play as a tool to enforce EU law – that is, whether the EU should have the competence to set harmonised standards in terms of if and when a violation of EU norms should amount to a criminal offence. Importantly for the discussion in this book, relying on the harmonisation of criminal law to secure enforcement of EU policies – what Chapter 3 defined as grounding harmonisation in a regulatory rationale – implies a utilitarian legitimacy for EU criminal law. 1 For more on the distinction between securitised and functional, see ch 4, s I. 2 J Vervaele, ‘Harmonised Union Policies and the Harmonisation of Substantive Criminal Law’ in A Weyembergh and F Galli (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Editions de l’ULB, 2013) 43. 3 ibid.

122  EU Competences on Functional Criminalisation This chapter retraces the evolution of EU constitutional law regulating Member States’ obligations to ensure effective enforcement of EU law. It starts from the early case law of the Court in the 1960s until the codification of EU functional criminalisation competence in the Treaty of Lisbon. Looking into the historical background of the current competence can indeed provide interesting insights on how to interpret them. More specifically, section I further explains the scope of the analysis in the chapter, clarifying the links between the question of enforcement of EU norms or policies, and the protection of EU interests, which are both included under the EU ‘functional criminalisation’ umbrella. Section II discusses the early case law of the Court on enforcement of EU norms, namely the 1977 Amsterdam Bulb case and the 1989 Greek Maize case, which introduced general sanctioning obligations for Member States. Section III discusses the second phase of the Court’s case law, namely the landmark 2005 Environmental Crime case and 2007 Ship Source Pollution case, where the Court’s introduces the competence for the EU to impose obligations onto Member States to criminalise violation of EU law. Section IV analyses the understanding of legitimacy of criminal law which stems from Article 83(2) TFEU, including an analysis on whether the envisaged legitimacy for EU criminal law lives up to European values and principles. Section V summarises the main arguments. I.  THE SCOPE OF EU FUNCTIONAL CRIMINALISATION

The uses of EU criminal law to secure the enforcement of EU policies and protect EU interests (especially EU financial interests) both fall into ‘EU functional criminalisation’, because in both cases what is at stake is the safeguarding of the EU public order. Rather than to protect the security of citizens,4 EU criminal law is used to preserve the existence and the functioning of the EU. Alongside this common purpose, EU competences in these two areas also have an intertwined ‘institutional history’, so it thus makes sense to discuss them jointly. Indeed, one of the first landmark cases – discussed in the next section – in which the question of imposing criminalisation obligations on Member States arose, the 1989 Greek Maize case, concerned a violation of EU subsidies norms. In this context, the EU was doubly ‘victimised’, first because EU norms were not respected and second because its financial interests were harmed. Interestingly, in the same period, the Commission proposed two protocols concerning EC fraud and corruption by EC officials. However, neither of them was approved in the Council5 and

4 However, see what was said in ch 4, s I on the objective of security and effectiveness of EU policies as overlapping. 5 On the legislative history of these protocols, see J Vervaele, Fraud against the Community: The Need for European Fraud Legislation (Boston, Kluwer Law and Taxation Publishers, 1992) 85 ff.

The Early Case Law  123 eventually EU legislation criminalising fraud against the EU was adopted under the third pillar, as was discussed in Chapter 4.6 The Environmental Crime and Ship Source Pollution cases discussed below then addressed the broader question of whether the EU could resort to criminal law to ensure effective implementation EU policies.7 Yet, legislation on the protection of the financial interests of the EU was adopted in the Treaty of Lisbon on the basis of Article 83(2) TFEU, which codifies EU functional criminalisation competence, as will be discussed in section IV.8 However, from a criminal legal theory perspective, it is necessary to distinguish between legitimating harmonisation of criminal law on the basis that important EU financial interests are at stake and on the basis that there is an enforcement gap in the relevant EU policy area, which criminal law could correct. In the former case, the harmonisation of criminal law is justified on a values-based criminalisation rationale and criminal law thus has deontological foundations. In the latter case, harmonisation is justified on a regulatory rationale and criminal law has utilitarian foundations. The aim of this chapter is to investigate to what extent the EU primary law has acknowledged a self-standing competence to use criminal law where there are enforcement gaps or whether this is only the case when important interests, including EU financial interests, are at stake. II.  THE EARLY CASE LAW: SANCTIONING OBLIGATIONS AND AN INTEGRATED APPROACH TO CRIMINAL LAW

The Court first mentioned the possibility of enforcing EU law through criminal law in the 1977 Amsterdam Bulb judgment.9 This case concerned the effective application of a Community regulation in the field of agriculture. The Court established that the obligation of loyal cooperation enshrined in (at the time) Article 5 TEC required Member States to take all possible measures to ensure the effective application of Community law and that these could also be of a criminal nature. The 1989 Greek Maize case10 pushed the reasoning further. This case concerned a fraud against the Community budget committed by the Greek 6 Council Act of 26 July 1995 drawing up the Convention on the protection of the European Communities’ financial interests [1995] OJ C316/03; Council Act of 27 September 1996 drawing up a Protocol to the Convention on the protection of the European Communities’ financial interests [1996] OJ C313/1; Council Act of 19 June 1997 drawing up the Second Protocol of the Convention on the protection of the European Communities’ financial interests [1997] OJ C221/11. 7 See s III. 8 See s IV. 9 Case C-50/76 Amsterdam Bulb BV v Produktschap voor Siergewassen [1977] ECR I-137, para 32. 10 Case C-68/88 Commission of the European Communities v Hellenic Republic [1989] ECR I-2965 (hereinafter the Greek Maize case).

124  EU Competences on Functional Criminalisation authorities. Within the framework of the infringement proceedings brought against Greece, the Court commented in more depth on the Member States’ obligations to ensure the effectiveness of Community law. The judges reiterated that (at the time) Article 5 TEC required Member States to take all measures guaranteeing the application and the effectiveness of Community law, but that the ultimate choice as to the choice of penalties remained at the discretion of Member States. Yet – and this is the importance of the judgment – the judges added two further conditions. The first condition was that national legislators had to ensure that ‘infringements of Community law are penalised under conditions, both procedural and substantive, which are analogous to those to infringements of national law of a similar nature and importance’.11 This principle is referred to as the ‘assimilation principle’ because violations of supranational norms have to be assimilated into violations of national norms.12 Setting obligations of ‘functional equivalence’ for Member States is common in other areas. For instance, a similar assimilation principle governs the protection of the good functioning of the EU judiciary. The Statute of the Court of Justice of the European Union establishes that any violation of witnesses’ or experts’ oaths before the Court have to be considered similar to correspondent behaviours when realised before national jurisdictions.13 Moreover, on a procedural level, the case law of the Court has traditionally held that Member States must provide remedies for violations of EU law under procedural conditions which should not be less favourable than those related to similar domestic actions and must not make the exercise of the Community right impossible in practice.14 Interestingly, the wording of the Greek Maize case seems to establish the assimilation principle as a general transversal principle that is applicable to all policy areas. The Treaty of Maastricht codified the principle of assimilation by including it in the Treaty. However, it limited its scope only to the field of the protection of the financial interests of the EU.15 A second condition was that, regardless of their nature, the chosen penalties must always be effective, proportionate and dissuasive.16 These three 11 ibid para 24. 12 V Mitsilegas, EU Criminal Law (Oxford, Hart Publishing, 2009) 63. 13 Article 30 of Protocol No 3 on the Statute of the Court of Justice of the European Union [2012] OJ C326/210 (art 27 at the time of the Greek Maize case). On the principle of assimilation to protect the good functioning of the EU judiciary, see R Sicurella, ‘Setting up a European Criminal Policy for the Protection of EU Financial Interests: Guidelines for a Coherent Definition of the Material Scope of the European Public Prosecutor’s Office’ in K Ligeti (ed), Toward a Prosecutor for the European Union: A Comparative Analysis, Vol 1 (Oxford, Hart Publishing, 2012) 870, 889. 14 Case 33/76 Rewe Zentralfinanz v Lanwirtschaftskammer für das Saarland [1979] ECR I-1989; see further on the development of Court’s case law M Dougan, ‘The Vicissitudes of Life at the Coalface: Remedies and Procedures for Enforcing EU Law before the National Courts’ in P Craig and G de Búrca (eds), The Evolution of EU Law (Oxford, Oxford University Press, 2011) 407. 15 See art 209A TEU (consolidated in Maastricht) [1992] OJ C224/6, now art 325 TFEU. See the discussion on this provision in ch 4, s III.A. 16 Greek Maize case (n 10) 24.

The Early Case Law  125 requirements for Community penalties have been confirmed in numerous other cases and have become a ‘Community mantra’ also reproduced in legislation, although regrettably the Court never actually defined these concepts in detail.17 If the concept of proportionality was known to the Community legal order, ‘dissuasiveness’ and ‘effectiveness’ were not. Moreover, it was argued that proportionate and dissuasive sanctions might be two oxymoronic concepts.18 The first is an expression of a retributive logic, according to which the severity of the penalty depends on the degree of the harm caused, whereas dissuasive sanctions are calculated based on a deterrence logic, where the severity of the penalty depends on its capacity to persuade individual not to offend in the future.19 The Court did not clarify what kind of balance should be struck between these two. Admittedly, neither in the Amsterdam Bulb nor in the Greek Maize case did the Court explicitly set criminalisation obligations or rules on the function for EU criminal law.20 However, a joint reading of the two conditions established in the Greek Maize case arguably establishes an indirect obligation for Member States to adopt an integrated, utilitarian and deontological approach to criminal law. On the one hand, the Court requires effective sanctions for violations of EU law. While it does not establish that these necessarily be criminal, it is clear that if the choice is made in favour of criminal law, its effectiveness in securing the enforcement of EU law must be one of the aspects to be taken into consideration. This suggests a utilitarian understanding of criminal law. Similarly, when insisting on the dissuasiveness of the sanction, the Court in practice insists on the need to look at the future objective of compliance with the law. If the choice in favour of criminal law, as an alternative to administrative law, for instance, is to be made on these grounds, this implies a utilitarian approach to criminal law.21

17 For the expression ‘Community Mantra’ and for a discussion of the contours of EU law sanctions as established by this case law, see E Herlin-Karnell, The Constitutional Dimension of EU Criminal Law (Oxford, Hart Publishing, 2012) 17. These three conditions have also been reproduced in a vast number of legislative instruments; for an overview, see H Labayle, ‘L’ouverture de la jarre de Pandore : réflexions sur la compétence de la Communauté en matière pénale’ (2006) 3 Cahiers de Droit Européen 382, 387. 18 Herlin Karnell (n 17). 19 See the discussion on proportionality in ch 2, s II.B. 20 Such restraint from issuing definitive judgments on the use of criminal law was most likely due to constitutional reasons. The Treaty did not include any criminal law competence at the time and the Court probably did not want to go as far as to create one at that time. As will be explained in the next section, in 2005 when the Court introduced an EU criminal law competence, the Constitutional Treaty had already envisaged such a possibility. In 2005, there was thus ‘some’ institutional background to the Court’s judgment, even though the Treaty had been rejected by France and Ireland. See M Bergstrom, ‘The Dynamic Evolution of EU Criminal Law’ in M Bergstrom and AJ Cornell (eds), European Police and Criminal Law Co-operation (Oxford, Oxford University Press, 2014) 185. 21 The reference to ‘proportionate’ sanctions is conversely a neutral one in terms of the envisaged legitimacy of criminal law. The Court does not specify whether the sanction, and thus the choice in favour of criminal law, should be proportionate to the future objective of compliance with the law

126  EU Competences on Functional Criminalisation On the other hand, by tying the choice in favour of criminal law to the nature and importance of the infringements of EU law – which must be compared with similar infringements of national law – the Court also arguably suggested incorporating deontological considerations into criminalisation decisions. Indeed, looking at ‘nature and importance’ of the infringements of EU law could be interpreted as referring to the values that the relevant EU norm protects and the sanctioned behaviour harms. This interpretation is particularly fitting given the facts of the Greek Maize case. In this context, the interests at stake are financial interests, so the comparison to be made is between the sanctioning regime for fraud against the EU budget and fraud against the national budget. The message that the Court sends with the Greek Maize case is precisely that EU public money should be as important as national public money. In the years that followed the Greek Maize case, the Commission tried several times to have EU instruments including criminalisation obligations for Member States adopted in the field of money laundering and insider dealing, with the aim of ensuring the effectiveness of internal market policies. Yet it did not succeed.22 Finally, the Treaty of Maastricht introduced the competence for the EU to adopt criminal law legislation. However, this was only in the field of securitised criminal law, and within a very intergovernmental setting, without providing the Court with the competence to enforce EU Criminal Law ­instruments.23 Despite there not being an explicit competence for it, criminal law in the field of the protection of financial interests would be approved within the framework of the third pillar, as mentioned above.24 However, the question of whether the EU could require Member States to use criminal sanctions to enforce EU policies remained an open one. III.  THE CASE LAW IN THE 2000s: CRIMINALISATION OBLIGATIONS AND AN INTEGRATED APPROACH TO EU CRIMINAL LAW

Determined to expand the EU competence to adopt criminal norms beyond the institutionally and materially limited scope of the third pillar, the Commission opted for the judicial route. In 2005 and 2007, it brought two actions before the Court of Justice seeking the annulment of two framework decisions, third pillar instruments including criminal law sanctions for environmental crimes.25 or to the seriousness of the infringement. Compare the Court’s formulation with that in the current art 49(3) of the EU Charter of Fundamental Rights, where it is stated that the severity of the penalty should be proportionate to the offence. See ch 2, s II.B.ii. 22 This is further discussed in ch 7, s IV. 23 See ch 4, s II. 24 See n 6 above. 25 Council Framework Decision 2003/80/JHA of 27 January 2003 on the protection of the environment through criminal law [2003] OJ L29/55 (hereinafter the Framework Decision on Environmental Crime); Council Framework Decision 2005/667/JHA of 12 July 2005 to strengthen the criminal law framework for the enforcement of the law against ship-source pollution [2005] OJ L255/164 (hereinafter the Framework Decision on Ship-Source Pollution).

The Case Law in the 2000s  127 It argued that because of the subject matter, the texts should have been adopted on a first pillar legal basis. These actions gave the European judges the chance to pick up where they had left off some 15 years earlier with the 1977 Amsterdam Bulb and the 1989 Greek Maize cases, where they had envisaged sanctioning obligations grounded in the Treaty norm on loyal cooperation. In the 2005 Environmental Crime26 and the 2007 Ship Source Pollution27 cases, the Court went as far as establishing a self-standing competence for the EC to enact first pillar directive harmonising the definitions of crimes with the aim of enforcing EC norms, bringing about a true institutional revolution. Given their capital importance in constitutional law terms – that is, in terms of the shape of EC competences and especially of the delimitation of competences between the former first and third pillars – these decisions have become the subject of a vast literature.28 This section, like the previous one, focuses on one specific aspect of the debate: the criteria according to which the expansion of competence takes place.29 More specifically, it discusses the rationales for the harmonisation of definitions of crimes that these two cases introduce and the relevant legitimating factors for EU criminal law they imply. A.  The 2005 Environmental Crime Case: The Utilitarian Dimension of EU Criminal Law In the 2005 Environmental Crime case, the Court recalled that the protection of the environment is one of the objectives of the Community, which falls within the first pillar. Therefore, it argued, despite the fact that criminal law normally fell within the third pillar, the EC had the competence to impose on Member States the adoption of effective, proportionate and dissuasive criminal penalties if this proved essential to ensuring the effective application of EU environmental norms. The key passage, which constitutes the real innovation of the case, is as follows: The fact that criminal law and criminal justice are in principle competence of Member States does not prevent the Community legislature, when the application of effective, proportionate and dissuasive criminal penalties by the competent national authorities is an essential measure for combating serious environmental offences, from taking measures which relate to the criminal law of the Member States which it

26 Case C-176/03 Commission v Council [2005] ECR I-7879. 27 Case C-440/05 Commission v Council [2007] ECR I-9097. 28 See the discussion in Mitsilegas (n 12) 70–84; and Herlin-Karnell (n 17) 29–32; as well as E Herlin-Karnell, ‘Commission v Council: Some Reflection on Criminal Law in the First Pillar’ (2007) 13 European Public Law 69; Vervaele (n 2) 45–46. See also S White, ‘Harmonisation of Criminal Law under the First Pillar’ (2006) 31 European Law Review 81; and S Peers, ‘The Community’s Criminal Law Competence: The Plot Thickens’ (2008) 33 European Law Review 399. 29 Herlin-Karnell hints at the fact these cases have a double importance, both from the perspective of the scope of competence and from the legitimacy of EU criminal law: Herlin-Karnell (n 17) 30.

128  EU Competences on Functional Criminalisation considers necessary in order to ensure that the rules which it lays down on environmental protection are fully effective. (Emphasis added)30

The Framework Decision was annulled and replaced with a directive.31 Interestingly for the analysis in this book, by focusing on the effectiveness potential of criminal law,32 the 2005 Environmental Crime case introduced the regulatory rationale for harmonisation of criminal law, thus implying a utilitarian approach to EU criminal law. Admittedly, the Court was far from specific as to what is meant by effectiveness in this context and it did not provide any empirical data to substantiate this concept.33 Moreover, it was observed that precisely in the field of environmental protection, criminal law is not necessarily the most effective weapon to secure effective law enforcement.34 However, this does not negate the point that, as a matter of principle, effectiveness is acknowledged as a relevant criterion justifying the adoption of EU criminal law. The question is whether effectiveness is a sufficient condition for the adoption of EU criminal law or only a necessary one, which must be accompanied by complementary deontological considerations on the interests at stake. The answer to this lies in the importance of the objective of the protection of the environment for the EU, and on whether this new competence is intended as only applying to this specific policy area or in any EU policy area where there is an enforcement deficit. Both aspects are discussed in the next sections. B.  The 2005 Environmental Crime Case: The Deontological Dimension of EU Criminal Law The Court’s 2005 judgment in the Environmental Crime case did not explicitly establish a broad competence to adopt criminal sanctions in order to enforce any EU policy, and it is not clear to what extent the findings of the Court can be applied beyond environmental policies and, if so, on which basis. As was the case for the early case law, where the Court did not explicitly take a position on the role of criminal law, the Court’s reluctance to designate the EU competence to use criminal law for enforcement purposes as a cross-policy competence has constitutional reasons – that is, it derives from the need to avoid an excessive case law-based expansion of the EU range of action in absence of Treaty provisions. It is worth recalling that the draft Constitutional Treaty included

30 Commission v Council (n 26) 47–48. 31 Directive 2008/99 of the European Parliament and of the Council of 19 December 2008 on the Protection of the Environment through Criminal Law [2008] OJ L328/28. 32 The 2005 case speaks of ensuring ‘that rules which it lays down on environmental protection are fully effective’: Commission v Council (n 26) 47–48; Commission v Council (n 27) 66. 33 Herlin-Karnell (n 17) 59. 34 M Faure, ‘European Environmental Criminal Law: Do We Really Need it?’ (2004) 13 European Environmental Law Review 18.

The Case Law in the 2000s  129 an EU functional criminalisation competence.35 The Court thus probably felt confident in introducing into the EU legal order a constitutional innovation that had already been agreed upon during the Intergovernmental Conference and that failed to enter into force due to the referendum in two countries,36 but did not want to excessively expand the new EC competence. Leaving aside the constitutional question of the extent of the competence for a moment, it is still interesting to look more closely at the arguments that the Court relied on in the specific case to introduce the competence for the EU to use criminal law to enforce EU environmental policy. These are arguably values-based ones and thus are an expression of a deontological approach to EU criminal law. This stems very clearly from Advocate General Ruiz-Jarabo Colomer’s sophisticated analysis on the subject of the legitimacy of criminal law. While it did not reproduce all the relevant arguments, the reasoning of the Court seems nonetheless very much in line with it. It is thus worth recalling both the reasoning in the Advocate General’s Opinion and of the Court. The Advocate General first summarised the previous case law and legislation as to the competence for the Community to impose EC law enforcement obligations on Member States.37 He concluded that until that moment, there had not been any explicit mention of the EC competence to impose criminalisation obligations. The decision on whether to penalise the violation of EU law or not was left to Member States, provided that they would envisage effective, proportionate and dissuasive sanctions.38 Yet, he continued that if there are self-evident criteria for determining what such ‘effective, proportionate and dissuasive penalty’ should look like, there is no substantive reason preventing the party (namely the Community) that has competence in that sphere from taking the decision to make such penalties criminal.39 The interesting part, for the purposes of our analysis, is the discussion on how to carry out such an assessment of whether the required effective, proportionate and dissuasive requested sanctions need to be criminal. The Advocate General stated that ‘if the legal interests protected in such offences were one of the objectives of the Community’ (emphasis added),40 then nothing would impede the ability of the Community

35 Article III-271 of the Treaty establishing a Constitution for Europe [2004] OJ C310/1. 36 Maria Bergstrom puts forward an interesting theory which she calls ‘sensitive interpretation’ to interpret the Court’s judicial activism, especially in those cases where the Court has expanded EU competences. She argues that in this context, the Court basically respects the Member States’ will, which has manifested itself in Treaty amendments, but which are yet not in force. She applied this theory to Titanium Dioxide and Waste (Case 300/98 Commission of the European Communities v Council of the European Communities [1991] ECR I-02867), where she argued that the Treaty of Maastricht played a role even before it entered into force, and more pertinently for our analysis to the 2005 and 2007 criminal law cases which arguably resonated with the Constitutional Treaty, which had been agreed upon at the time but whose ratification had failed. Bergstrom (n 20). 37 Commission v Council (n 26) Opinion of AG Ruiz-Jarabo Colomer, paras 32–57. 38 ibid para 38. 39 ibid para 49. 40 ibid para 50.

130  EU Competences on Functional Criminalisation law-making bodies to insist that Member States prosecute using criminal law.41 The Opinion devotes an important number of paragraphs to demonstrating not only that environmental policies fall within the competence of the Community, but also that the protection of the environment is a fundamental objective of the Community.42 The Advocate General then broadened his perspective by recalling the importance of the protection of the environment through criminal sanctions in general. He resorted to fundamental rights arguments, as well as comparative criminal law and general legal theory arguments.43 He also argued that if criminal law is to be used to protect values that sustain co-existence, it seems reasonable that Community law avails itself of criminal penalties as the only ‘effective, proportionate and dissuasive’ penalties to attain a level of protection and to improve the quality of life through safeguarding the environment.44 This approach reflects a clear deontological approach to the legitimacy of EU criminal law. The importance of the protection of the environment is given proper account as a suitable Rechtsgut worthy of criminal law protection.45 Lastly, the Advocate General also interestingly spoke of an ‘ethical dimension’ of criminal law. He argued that ‘when an act is sanctioned in criminal terms, it is held to merit the most severe reproach because it transgresses the fundamental tenets of the legal system’.46 It is only after this elaborate discussion on environmental protection that the Advocate General turned to the utilitarian side of criminalisation, stating that the power to attach sanctions (be they administrative, civil or criminal) to violations of Community law is an power that is instrumental to the effectiveness of Community law. He concluded that only criminal law can afford the ‘effective, proportionate and dissuasive’ response to the relevant conduct. The extensive discussion on the importance of the protection of the environment, and the final recall of the instrumental character of criminal law to the effectiveness of Community norms, should be read in a complementary manner, as expression of an integrated agenda. Criminal law should be read as an instrument to secure the effectiveness of Community norms, but only those norms that pursue the achievement of some fundamental objective of the community safeguarding highly ranked interests, in this case the protection of the environment. Although in less detail, a similar emphasis on the importance of environmental policies can be found in the Court’s reasoning. The judges began by acknowledging that it is ‘common ground that protection of the environment constitutes 41 ibid. 42 ibid paras 51–59. 43 ibid paras 69–73. 44 This can be inferred from ibid para 72. 45 It should be mentioned that such a detailed description and justification of the importance of the interest protected by the criminal norms is not often found in national constitutional court decisions or in national legislation. Indeed, national constitutional courts rarely embark on an evaluation of the specific object protected by the criminal norms, but mainly focus on a proportionality analysis; see ch 1, s VI. 46 Commission v Council (n 26) Opinion of AG Ruiz-Jarabo Colomer, para 74.

The Case Law in the 2000s  131 one of the essential objectives of the Community’ (emphasis added),47 also relying on previous case law,48 and that ‘it is a transversal objective, which must be integrated in the implementation of each Community policy and activity’.49 Such a broad nature of the objective arguably also emphasises the fundamental character of environmental protection within the EU. After having stressed the importance of the objective, the Court lay down the important statement that is considered the first step towards the communitarisation of criminal law. This is that the EC has the competence to resort to criminal sanctions when criminal law is necessary to ensure the effectiveness of EU environmental policy. In a similar fashion to the Advocate General, the Court also adopts an integrated criminalisation agenda. While the Court does not make the link between the two as explicit as the Advocate General had done in his more detailed opinion, it is plausible that the assessment of the importance of the objective at stake represents in the Court’s reasoning a pre-condition for the discussion on the effectiveness of criminal law as an enforcement tool. The protection of the environment is a highly ranked interest for the EU (the Court starts with a value-based, deontological, perspective), so criminal law must be used when it is essential to secure the effective enforcement of EU environmental policies (the Court complements its approach with an instrumental, utilitarian based perspective). In conclusion, even if it is admitted that the criminal law competence established by the Court should be applicable beyond environmental policy, if the approach of the Advocate General and of the Court must be followed, a thorough analysis of the importance of the relevant policy objectives and of the interests at stake must take place. Interestingly, the Commission and the European Parliament each provided an interpretation of this landmark case, both suggesting a broad reading of it. Yet, they suggested different criteria to select which policies deserve criminal law enforcement, and especially whether complementary deontological justifications are necessary. C.  The Commission’s and the Parliament’s Diverging Interpretations of the Environmental Crime Judgment Directly after the judgment, the Commission published a communication interpreting the Court’s decision in a broad manner.50 It considered that criminal law 47 ibid para 41. 48 The Court recalled Case C-240/83 ADBHU [1985] ECR I-531, para 13; Case C-302/86 Commission v Denmark [1988] ECR I-4607, para 8; and Case C-213/96 Outokumpu [1998] ECR I-1777, para 32. 49 Commission v Council (n 26) para 42. 50 Communication from the Commission to the European Parliament and the Council on the implications of the Court’s judgment of 13 September 2005 (Case C-176/03 Commission v Council) COM(2005)583 final/2, para 6. For a comment on this, see Vervaele (n 2) 51–53.

132  EU Competences on Functional Criminalisation could be a tool to be used in all EU policies which met a necessity and a coherence test.51 The substance of the Commission’s necessity test is particularly instructive for the purposes of our analysis.52 The test only requires assessing the suitability of the criminal sanctions to ensure the effectiveness of the EU policy at stake.53 This means that the suitability of criminal law to achieve the objective of enforcement is the necessary and the sufficient criterion for the introduction of criminal law sanctions. Briefly, the harmonisation of definitions of crimes is justified in all cases in which there is an enforcement deficit. The Commission thus endorsed the regulatory rationale as a self-standing rationale for harmonisation, which implies a purely utilitarian interpretation of the legitimacy of criminal law. Interestingly, the Commission had also introduced a proposal for extending the Community method applying in the first pillar also to the policy areas covered by the third pillar by means of the bridging clause (ex Article 42 TEU).54 This proposal was nonetheless put on hold while the potential entry into force of the Constitutional Treaty, which would have brought about such de-pillarisation, was still debated.55 The Parliament’s resolution echoed the Commission’s position that the newly introduced criminal law competence could go beyond environmental policies.56 However, it called for caution in relation to how such expansion should take place. A case-by-case analysis of each policy area should take place. Criminal law should be used to correct enforcement deficits only in terms of the main principles, objectives and competences of the Community. In other words, the harmonisation of definitions of crimes is considered to be justified in those policy areas where there is an enforcement deficit (the regulatory rationale) and which involves some particularly important interests for the EU (the valuesbased criminalisation rationale).57 The Parliament thus envisaged an integrated, utilitarian and deontological legitimacy for EU criminal law. Briefly, at this stage, there were a range of opinions among the various institutions on the subject of a deontological or utilitarian agenda for the use of criminal law to enforce EU policies. The Court had a second opportunity to clarify this point in 2007 in the Ship Source Pollution case, but did not fully seize it, as will be explained in the next section.

51 COM(2005)583 final/2 (n 50) paras 12–13. 52 The consistency test requires avoiding fragmentation between criminal law adopted in different policy areas and especially criminal law adopted under the first and third pillar: ibid para 13. 53 ibid paras 8, 9, 12. 54 ibid para 32. 55 For an analysis of the content of the proposal and its fate, see Herlin-Karnell (n 17) 33. 56 European Parliament, Resolution on the consequences of the judgment of the Court of 13 September 2005 (2006/2007(INI)), P6_TA(2006)0260, pt J. 57 ibid.

The Case Law in the 2000s  133 D.  The 2007 Ship Source Pollution Case: A Confirmed Integrated Agenda for the Legitimacy of EU Criminal Law? The 2007 Ship Source Pollution case concerned a Framework Decision requiring Member States to approximate criminal laws in order to combat ship-source pollution caused with intent or as a result of serious negligence with a view to strengthening maritime safety.58 Like the Environmental Crime case, the Commission had brought an action for annulment claiming that the text touched upon the EC competence in the field of transport and thus should have been adopted under the first pillar. The Court noted that the common transport policy is one of the foundations of the Community, but it also linked it to the objective of environmental protection.59 On the basis of these considerations, it came to the same conclusion as in the 2005 judgment. It annulled the Framework Decision on the grounds that the relevant provisions (including the criminal law provisions) aimed at ensuring effective maritime safety as well as environmental protection could have been adopted under the first pillar. The Framework Decision was replaced with a directive.60 Given that the Court still insisted on the link between transport and environmental policy – be it artificial or genuine – it remained unclear whether the new EC criminal law competence could be interpreted broadly, as the Commission suggested, or would only be limited to environmental policy in light of its importance. This did not help clarify whether this new competence envisaged a utilitarian or an integrated utilitarian and deontological legitimacy. A closer look at both the Advocate General’s and the Court’s reasoning does not provide particularly insightful points to interpret the Court’s judgment. Advocate General Mazak considered that criminal sanctions should be used when necessary to also ensure the effective enforcement of EU policies beyond the environment protection field. The Opinion extensively elaborates on the subject of the EU policies covered by the Community criminal law competence. It first dwells on the special nature of criminal law and on its connection with the culture and the society of a community.61 It then recalls how one of the main features of criminal law is its potential as a deterrent. However, Advocate General Mazak stressed that dissuasion cannot be the sole objective for which criminal law is used and that criminal law should be used only as a last resort, considering its inherent connection with the identity of the Community.62

58 Framework Decision on Ship-Source Pollution (n 25). 59 Commission v Council (n 27) para 60. 60 Directive 2005/35 of the European Parliament and of the Council of 7 September 2005 on ShipSource Pollution and the Introduction of Penalties for Infringements [2005] OJ L255/11. 61 Commission v Council (n 27) Opinion of AG Mazak, paras 68–69. 62 ibid paras 71–72.

134  EU Competences on Functional Criminalisation He also highlighted how protection of the environment is one of the fundamental objectives of the European Community and that at least all the objectives envisaged in (at the time) Articles 2 and 3 TEC were equally fundamental.63 As a consequence, it would not do justice to the identity of the Community to exclude policies such as the establishment of an internal market characterised by the fundamental freedoms, the Common Agricultural Policy or the common rules on competition, which, it can be inferred, are as important as the environmental freedoms.64 All these observations reflect an integrated approach, which understands criminal law as an instrument of deterrence, but also as a field directly connected to social norms, where a selection of interests is necessary, and which includes (at least) all the fundamental objectives of the EU as interests worthy of protection. Yet, the Advocate General also put forth some other considerations, which are arguably hard to reconcile with his previous (alleged) deontological approach. He also stated later in his Opinion that a situation in which the enforcement via criminal law of some other EU policies would be disregarded because such policies are less important would not be acceptable.65 Criminal law should be a corollary to the general principle of effectiveness of Community law (the effet utile principle) and it should be envisaged for every policy whether this test is met. What emerges from these latter statements is an unequivocally utilitarian perspective: the only test that the use of criminal law should pass is the effectiveness test. Briefly, both the deontological and the utilitarian legitimacy for criminal law are mentioned in the opinion of the Advocate General. Nonetheless, they are presented in a contradictory manner rather than a complementary one leading to an integrated agenda. Interestingly, after this case, a first pillar criminal law directive was approved in a field not connected to the environment, namely illegal migration, requiring Member States to impose dissuasive, proportionate and effective criminal sanctions to employers of illegally staying third-country nationals.66 This showed the will of the political institutions to consider the competence established by the Court as a broadly applicable one. When looking at the justification for adopting the Directive and, in particular, the rationales for adopting criminal law, a strong focus on effectiveness emerges, whereas there is little emphasis on the deontological reasons why criminal law should be used in the field of illegal migration. The preamble does mention on one occasion that criminal law should be used for the most serious cases,67 which hints in the direction of a deontological approach.68 However, the main rationale for criminal law 63 ibid paras 94–95. 64 ibid. 65 ibid para 98. 66 See art 9 of Directive 2009/52 providing for Minimum Standards on Sanctions and Measures against Employers of Illegally Staying Third-Country Nationals [2009] OJ L168/24. 67 ibid Preamble, recital 22. 68 See the discussion on serious crime in ch 4, s III.B.

The Treaty of Lisbon: The Final (Utilitarian) Word  135 stemming from the preamble is that administrative sanctions are not sufficient to dissuade employers from hiring illegally staying third-country nationals, hence the need for further legislation and in particular for criminal law.69 The Treaty of Lisbon dissolved any doubt as to the approach to criminalisation, embracing a purely utilitarian perspective. It constitutionalised the effectiveness-based regulatory rationale for harmonisation of the definition of crimes as a stand-alone rationale, without any complementary deontological rationale. IV.  THE TREATY OF LISBON: THE FINAL (UTILITARIAN) WORD

The Treaty of Lisbon codifies and clarifies the extent of the EU competence to harmonise criminal law in order to enforce EU policies in Article 83(2) TFEU, which states that: If the approximation of criminal laws and regulations of the Member States proves essential to ensure the effective implementation of a Union policy in an area which has been subject to harmonisation measures, directives may establish minimum rules with regard to the definition of criminal offences and sanctions in the area concerned. (Emphasis added)

Criminal law directives adopted on the basis of Article 83(2) TFEU shall follow the legislative procedure that was followed for the adoption of the relevant directives harmonising the policy area where criminal law enforcement is being introduced. This can be the ordinary legislative procedure with the full involvement of the Parliament, but it can also be a procedure requiring only consultation of the Parliament.70 Article 83(2) TFEU represents the ‘quasi-codification’71 of the two leading cases on environmental crime, which were discussed in the previous section. While codifying them, it also brings further clarification to the extent of the competences ratione materiae, extending it beyond the Court’s case law. First, it clarifies that the competence is also extended to the definition of sanctions, which was not clear in the case law. Second, it speaks of the effectiveness of EU policies and not of EU norms or objectives as did the Court, which arguably is a broader terminology.72 Lastly, and most importantly, the Treaty clarifies that

69 Directive on Illegal Employment of Third-Country Nationals (n 66) Preamble, recitals 2, 21, 22, 23, 24, 28. 70 An example is the field of approximation of legislation on turnover taxes, excise duties and other forms of indirect taxation regulated by art 113 TFEU, or competition law. In all these cases there is only an obligation to consult the Parliament; see art 103 TFEU. 71 A Weyembergh, ‘Introduction’ in Weyembergh and Galli (n 2) 16; P Asp, The Substantive Criminal Law Competence of the EU (Stockholm, Stiftelsen Skrifter utgivna av Juridiska fakulteten vid Stockholms universitet, 2012) 127. 72 Mitsilegas argues along these lines: Mitsilegas (n 12) 108.

136  EU Competences on Functional Criminalisation this criminal law competence is indeed applicable to each EU policy regardless of any link to environmental policy.73 Article 83(2) TFEU stirred up a significant debate. Questions were raised as to the meaning of the requirement of ‘essential for effective enforcement’ or the necessary degree of ‘prior harmonisation’ to justify relying on this competence,74 what the implications of the exercise of this competence for the principle of legality and democracy in criminal law are,75 and naturally what legitimacy for EU criminal law this competence implies. The following sections focus on this aspect, the core theme of the book, and it looks at the coherence between the inherent utilitarian legitimacy for criminal law in this provisions and EU values (see section IV.A), and its relationship with the principle of subsidiarity (see section IV.B). A.  Article 83(2) TFEU: The Tension between Utilitarian EU Criminal Law and EU Values Article 83(2) TFEU clearly states that approximation of criminal law is justified in light of its potential to ensure the objective of effective implementation of EU policy,76 codifying a regulatory rationale for harmonisation. It conversely drops the complementary values-based criminalisation rationale, which one could infer from the Court’s decision and was acknowledged by the Parliament. Any reference to the environment or to any other specific policy areas was dropped. Moreover, what has disappeared is the reference to any qualitative features of the selected policy areas. The Treaty does not speak of fundamental objectives, as did the Court, or of main policy and principles, as did the Parliament in its resolution published after the 2005 criminal law case.77 73 A Bernardi, ‘La competenza penale accessoria dell’Unione Europea. Problemi e Prospettive’ (2012) 1 Diritto Penale Contemporaneo 43, 50. 74 See, among others, Valsamis Mitsilegas, EU Criminal Law after Lisbon (Oxford, Hart Publishing, 2016) 61; S Peers ‘EU Criminal Law and the Treaty of Lisbon’ (2008) 33 European Law Review 507; Weyembergh (n 71) 16; Asp (n 71) 127 ff; R Sicurella, ‘EU Competence in Criminal Matters’ in V Mitsilegas, M Bergstrom and T Konstantinides (eds), Research Handbook in EU Criminal Law (Cheltenham, Edward Elgar, 2015) 49; Vervaele (n 2) 62; J Tricot, ‘L’harmonisation pénale accessoire: question(s) de method : observations sur l’art et la manière de légiférer pénalement selon l’Union Européenne’ in G Giudicielli-Delage and C Lazerges (eds), Le droit pénal de l’Union Européenne au Lendemain du Traité de Lisbonne (Paris, Société de Législation Comparée, 2012) 185, 186–90; P De Hert, ‘The EU Data Protection Reform and the (Forgotten) Use of Criminal Sanctions’ (2014) 4(4) International Data Privacy Law 262, 267. 75 J Oberg, ‘Union Regulator Criminal Law Competence after the Lisbon Treaty’ (2011) 19 ­European Journal of Crime, Criminal Law and Criminal Justice 289. 76 Interestingly, Gindre highlights how the pursuit of effectiveness on which the provisions put so much emphasis is actually hindered by other provisions in the Area of Freedom, Security and Justice, namely the possibility for one Member State to halt the legislative procedure (the ‘emergency brake’ procedure) and the various opt-outs granted to Ireland, the UK and Denmark: E Gindre, ‘L’Harmonisation pénale accessoire. Eléments de réflexion sur la place du droit pénale au sein de l’Union Européenne’ in Giudicielli-Delage and Lazerges (n 74) 201–02. 77 Juliet Tricot speaks with respect to art 83(2) TFEU as a provision that is not ‘borné materiellement’ (content-wise limited): Tricot (n 74) 192.

The Treaty of Lisbon: The Final (Utilitarian) Word  137 The only requirement for an EU policy to deserve criminal law enforcement is that there has already been some harmonisation in the field. This means that the EU obviously must have competence in the area selected and that it must have already taken a decision to exercise it. However, this is a procedural requirement, uncoupled from the substance of the policies at stake and especially from their importance in relation of the identity and the values of the EU. In other words, as long as criminal law is the only way to secure the effective implementation of any EU policy that the EU has considered worthy of pre-emption, then it can be relied upon as an enforcement tool. This implies a purely utilitarian, effectiveness-based understanding of criminal law. Admittedly, it sets a very high threshold. Resort to criminal law is not permitted simply when criminal sanctions would increase the enforcement rate of that specific policy, but is only allowed when criminal sanctions are essential to ensure the achievement of the relevant policy objective. Yet, this is still a purely instrumental perspective. The Treaty provision does not require looking at the interests that the criminal norm is protecting, namely the Rechtsgut. The neglect of the deontological dimension of criminal law by this competence in Lisbon Treaty is problematic. As was explained in Chapter 2, the EU’s choice to embrace liberal values as its own in Article 2 TEU, and especially liberty and dignity, requires a minimalistic, integrated justification in order to resort to the criminal law. The EU insistence on effectiveness as a self-standing criterion defining EU criminal law competences is at odds with the EU constitutional values. As was clarified in the Introduction, the relationship between two norms of EU primary law – in this case Article 2 TEU and Article 83(2) TFEU – cannot be interpreted as one of constitutional hierarchy, but rather as that of ‘normative coherence’. The argument here is thus that the EU definition of competence is not coherent with the EU founding values. This is naturally not a new debate. Indeed, the EU constitutional framework basically constitutionalises a trend that has already been spreading widely and heavily criticised at a national level, which is defined as regulatory or bureaucratic criminal law.78 Unsurprisingly, criminal legal scholars have not missed the opportunity to also critically address the supranationalisation of this trend.79 It remains to be seen to what extent deontological arguments are factored into the decisions to exercise this competence in practice and if this has led to the actual adoption of regulatory EU criminal law. 78 See the observations of Tricot on the supranational level reproducing the national technique of the ‘legislation par renvoi’: ibid 191. On regulatory criminal law, see ch 1, ss IV and V. 79 The authors of the Manifesto on a European Criminal Policy insisted on having a values-based approach to EU criminal law and harshly criticised a purely utilitarian approach to the criminal law. See European Criminal Policy Initiative, ‘The Manifesto on European Criminal Policy’ (2011) 1(1) European Criminal Law Review 86. With specific reference to art 83(2) TFEU, see also M Kaiafa-Gbandi, ‘Approximation of Substantive Criminal Law Provisions in the EU and ­Fundamental Principles of Criminal Law’ in Weyembergh and Galli (n 2) 85, 96. The critique of the use of effectiveness as a competence definition parameter (and especially in the context of criminal law) is at the core of the work of Ester Herlin-Karnell; see Herlin-Karnell (n 17) ch 4.

138  EU Competences on Functional Criminalisation Conversely, the insistence on the effectiveness of criminal law, already in the definition of the competences, is consistent with the efficiency-based logic of subsidiarity, which regulates when to exercise such competence, which will be discussed below. B.  Article 83(2) TFEU: Utilitarian EU Criminal Law and Subsidiarity Competence definitions and criteria for their exercise, such as proportionality and subsidiarity, are meant to limit EU action in two different stages. There is thus not a need to ensure coherence between the two as there was to ensure consistency between competence definition and the values underlying the EU constitutional identity. Still, it is interesting to note that Article 83(2) TFEU sets a requirement – the effectiveness of EU criminal law – which is also constitutive of the principle of subsidiarity. Chapter 2 clarified that the official interpretation for subsidiarity is that EU action must be more efficient, and thus more effective, than national action. This means that EU criminal law must first be effective in achieving the enforcement of the law. Naturally, the second requirement of subsidiarity, namely the greater effectiveness of EU action, must also be met when exercising Article 83(2) TFEU competence. Chapter 2 also clarified that the clearest case in which EU action in criminal justice is justified is when there is a need for interstate cooperation. EU definitions of crimes are therefore justified when they are instrumental to judicial cooperation. Admittedly, both the regulatory rationale for the harmonisation of criminal law (enshrined in the Treaty) and the cooperation rationale equally imply a utilitarian legitimacy for EU criminal law, and therefore have similar implications for our analysis. The exercise of competence is not the main topic of this chapter; however, for the sake of completeness, it is interesting to investigate what role the cooperation rationale plays in the context of functional criminalisation. The Treaty framework does suggest that the cooperation rationale could also play a role when harmonising criminal law on the basis of Article 83(2) TFEU. The same reasoning that was applied to the position of Article 83(1) TFEU in the Treaty applies here too. However, it is less likely to play a role in this case than it did with respect to Article 83(1) TFEU securitised criminalisation, for at least two reasons. First, historically speaking, functional criminalisation, contrary to securitised criminalisation, did not develop in parallel to the efforts in the area of judicial cooperation. In other words, in the areas of securitised criminal law, such as organised crime or drug trafficking, the EU resorted contemporarily to the instruments of judicial cooperation and harmonisation, which thus developed in a rather interrelated manner. This did not happen, for instance, in the field of environmental crime, which was the first area of criminality where the EU envisaged first pillar criminal law, independently of judicial cooperation. Gindre even suggests that Article 83(2) TFEU-based criminalisation not only

Conclusions  139 lacks strong ties with judicial cooperation, but also more general ties with the whole objective of the Area of Freedom, Security and Justice as, she argued, it has nothing to do with granting security to citizens.80 However, this argument is of course valid only to the extent that we can draw a clear dividing line between the first pillar and the third pillar, or securitised and functional criminal law, which, as mentioned earlier, is not always the case.81 Second, in the discussion on Article 83(1) TFEU, it was also observed that limiting the competence of the EU to harmonise only cross-border crimes was connected to the need for judicial cooperation. In particular, the idea of confining EU action only to cross-border activities and especially to those offences that are cross-border in nature or for which there is a special need to be combated on a common basis, as established by Article 83(1) TFEU, was interpreted as indicating those criminal activities that require judicial cooperation between national authorities.82 As a result, it was argued that harmonisation in the field of cross-border crimes was needed to make such cooperation possible.83 In light of these observations, the fact that Article 83(2) TFEU does not mention the cross-border requirement seems to nuance the instrumental, cooperation rationale for the harmonisation of definitions of crimes. Overall, Article 83(2) TFEU-based harmonisation can be considered relatively autonomous from judicial cooperation.84 Subsidiarity might or should be based on different considerations that support cooperation in this context. V. CONCLUSIONS

This chapter concludes the analysis of the approach to the legitimacy of criminal law that can be derived from EU primary law and contributes to answering the second research question. It has focused on the case law and Treaty provisions on EU functional criminalisation, namely EU criminal law preserving the EU public order, such as its interest and the effective implementation of EU policies. The chapter has shown that the Court of Justice initially adopted an integrated approach in this context. This can be appreciated in the assimilation principle cases. Further, it could be appreciated in the Environmental Crime case, which considers the use of criminal law legitimate to enforce EU policies, like EU environmental policy, only where important interests are at stake, thus

80 Gindre (n 76) 199–200. 81 See ch 4, s I. 82 See ch 4, s III.B.ii. 83 ibid. 84 Tricot highlights how art 83(2) TFEU represents both a rift with and a continuation of the previous institutional framework. It is a rift with the previous tradition of harmonisation of criminal law legislation, given that it attenuates its link with judicial cooperation; conversely, it is in continuation with the previous case law on the enforcement of EU norms, which treats criminal law as a horizontal tool to enforce EU law in various policy areas. See Tricot (n 74) 186–90.

140  EU Competences on Functional Criminalisation including deontological considerations. The Treaty of Lisbon conversely only identifies the effectiveness of criminal law as a self-standing criterion defining EU competence in this area. It thus rules out any obligations to justify resorting to EU criminal law in light of the importance of the policy at stake. Admittedly, the Treaty also included Article 325(4) TFEU, which could have worked as a legal basis for EU criminal law grounded in an integrated legitimacy. However, the Council and the Parliament excluded this possibility in practice. The shift in the Treaty of Lisbon was said to be in tension with EU liberal values, which require a much more careful approach to the use of criminal law, and some complementary deontological reasons for resorting to EU criminal law, beyond the need to correct enforcement gaps. It remains to be seen whether the values-based considerations that can be found in the Court’s more balanced approach are considered when exercising Article 83(2) TFEU competence. The next chapter accordingly turns to the policy documents setting the criteria and the objectives for the harmonisation of EU criminal law, which the Institutions have declared they will use as benchmarks in their policy-making.

6 From Tampere to Stockholm The Path towards Integrated Legitimacy

C

hapter 1 illustrated how the criteria guiding the choice between criminal law, and administrative and civil law, are rarely incorporated into legally justiciable provisions at the national level and are only occasionally spelled out as general criteria in policy documents.1 However, the multi-level structure of the EU implies that the supranational level is particularly far from the citizens and thus it is in need of stronger legitimacy in their eyes. The EU Institutions therefore have a stronger need than nation states to clarify the aim that harmonisation of national criminal law is pursuing; in other words, what the legitimacy is for EU criminal law.2 The EU policy documents are thus a particularly interesting source to investigate how the Institutions have understood what the place for criminal law is within the EU legal order. Their analysis is a another crucial step contributing to the answer of the second research question of this book. The European Council has adopted a number of ‘multi-annual’ programmes in the field of Justice and Home Affairs, including criminal justice matters, in conjunction with each of the various Treaty amendments. Section I discusses the Vienna, Tampere and The Hague pre-Lisbon Justice and Home Affairs Programmes, whereas sections II and III respectively focus on the post-Lisbon Stockholm Programme and ‘strategic guidelines’ Justice and Home Affairs 1 See, for instance, Circolare della Presidenza del Consiglio dei Ministri del 19 dicembre 1983, relativa a ‘Criteri orientativi di scelta tra sanzioni penali e sanzioni amministrative’ in Supplemento ordinario alla G.U. n. 22 del 23 Gennaio 1984 (Document of the Presidency of the Council of Minister of 19 December 1983 on ‘Guidelines for the choice between criminal and administrative sanctions’, published in Ordinary Supplement to the Official Journal n 22 of 23 January 1984) or the declaration of Lord Williams of Mostyn (then Minister of State at the Home Office, now Attorney General) in a written reply to a question by Lord Dholakia, HL Deb, vol 602, WA 57 (18 June 1999). Moreover, the UK Human Rights Act requires any minister who is in charge of a Bill in both chambers to state whether the proposed Act complies with the European Convention on Human Rights and Fundamental Freedoms. Human Rights Act 1998 c 42, s 19. 2 C de Jong, ‘The European Parliament Resolution of 22 May 2012 on an EU Approach to Criminal Law’ in A Weyembergh and F Galli (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Editions de l’ULB, 2013) 37.

142  From Tampere to Stockholm multi-annual programmes. Moreover, directly after the adoption of the Treaty of Lisbon, which ‘communitarised’ EU action in criminal matters, ­depriving Member States to a significant extent of their veto power in this field, the Council of Ministers, the European Commission and the European Parliament adopted three ad hoc EU criminal policy documents. These contain the core of the policy discussion on the legitimacy of EU criminal law and are thus analysed in section IV. Section V brings the discussion together, drawing some general conclusions on the approach in the policy documents. It compares this approach with the one underpinning the definitions of competences discussed in the previous chapters with the aim of investigating into possible common historical trends. Section VI provides some concluding remarks on whether the approach the Institutions declare themselves to be following is consistent with EU general values and general principles. I.  THE UTILITARIAN APPROACH TO CRIMINAL LAW IN PRE-LISBON JUSTICE AND HOME AFFAIRS PROGRAMMES

The role, function and objectives of harmonisation have not been the object of extensive discussion by the Institutions in the pre-Lisbon scenario. The multiannual programmes in the Area of Freedom, Security and Justice did not devote a lot of space to the discussion of the goals and the legitimacy of EU criminalisation. In fact, each programme addressed the topic of judicial cooperation in criminal matters in an ad hoc section. In this context, harmonisation to support coperation was discussed as one of the possible strategies. A first relevant document is the 1999 Vienna Action Plan, which was adopted contextually with the Treaty of Amsterdam.3 This text is not technically enumerated among the multi-annual programmes, since it is a pragmatic document, mainly meant to set the priorities for the implementation of the Treaty of Amsterdam. The Action Plan mentions the goals of bringing the EU closer to its citizens,4 of combating crimes in order to foster free movement5 and of spreading a sense of justice among EU citizens,6 all of which are relevant to the current analysis in that they echo the free movement and socialising rationale discussed in Chapter 3 as potential goals for harmonisation. However, in none of these cases is it established that the harmonisation of definitions of crimes should be the selected strategy to achieve the goal.7 3 Action Plan of the Council and the Commission on how best to implement the provisions of the Treaty of Amsterdam on an Area of Freedom, Security and Justice [1999] OJ C190/1 (hereinafter ‘Vienna Action Plan’). 4 ibid pt 2. 5 ibid pt 5. 6 ibid pt 15. 7 The Vienna Action Plan (n 3) nonetheless considers that the approximation of procedural norms would avoid the situation of EU citizens being treated differently across Europe. This links

Pre-Lisbon Utilitarian Approach to Criminal Law  143 Harmonisation is specifically mentioned as a tool to support judicial cooperation and, in the Action Plan under the heading entitled ‘Approximation’, it is also explicitly recalled that the compatibility of norms is necessary to improve cooperation.8 By putting the focus on the need to use harmonisation to support cooperation, the Vienna Action Plan shows a utilitarian understanding of the legitimacy of the criminal law. The introduction of new crimes is justified by the need for repressive Member States to enforce their criminal law transnationally. Both the 1999 Tampere Programme9 and the 2005 Hague Programme,10 which were multi-annual programmes setting the policy framework for the upcoming five years, respectively following the Amsterdam Treaty and the (failed) Constitutional Treaty, present a similar utilitarian approach to the legitimacy of criminal law. The Tampere Programme recalls the general aim of ‘better compatibility and more convergence between the legal systems of the Member States’11 and it explicitly states that: ‘Enhanced mutual recognition of judicial decisions and judgements and the necessary approximation of legislation would facilitate co-operation between authorities and the judicial protection of individual rights’ (emphasis added).12 It then points out the need to focus on approximation in a limited number of areas, such as financial crime, (money laundering, corruption and euro counterfeiting), drug trafficking, human trafficking, particularly exploitation of women, sexual exploitation of children, high-tech crime and environmental crime.13 Admittedly, given that these are defined as sectors of particular relevance,14 it could be said that there are some substantive, and thus deontological, limits to the areas of crime which should be approximated in order to support judicial cooperation. The objective of fostering free movement is mentioned in the Tampere Programme,15 yet once again is not linked to the approximation of national criminal law, but rather to the need to develop an information campaign.16 The Hague Programme interestingly defines as areas of EU interest those of ‘serious and cross-border crime’, anticipating the formulation of the Treaty of Lisbon.17 It was explained in Chapter 4 that the reference to ‘serious crime’

approximation to equality and thus arguably to a perception of justice: ibid pt 19. On the relationship between justice and equality, see the discussion in ch 3, s V.A. 8 Vienna Action Plan (n 3) pts 17, 18. 9 European Council, Presidency Conclusions of the European Council Meeting in Tampere 15 and 16 October 1999, www.europarl.europa.eu/summits/tam_en.htm (hereinafter ‘Tampere Programme’). 10 European Council, The Hague Programme – Strengthening Freedom Security and Justice in the European Union [2005] OJ C53/1 (hereinafter ‘Hague Programme’). 11 Tampere Programme (n 9) pt 5. 12 ibid pt 33. 13 ibid pt 48. 14 ibid. 15 ibid pt 28. 16 ibid pt 29. 17 Hague Programme (n 10) pt 3.3.2.

144  From Tampere to Stockholm might imply a deontological understanding of criminal law.18 However, the Hague Programme is very clear about approximation being justifiable only on a cooperation rationale of substantive criminal law. The latter should serve the same purpose as the approximation of procedural law, namely that of ‘facilitat[ing] mutual recognition of judgments and judicial decisions and police and judicial cooperation in criminal matters having a cross-border dimension’.19 In short, the legitimacy for EU criminal law remains predominantly a utilitarian one, like in the Tampere Programme. It is somehow surprising that the Hague Programme, which was published contextually with the Constitutional Treaty, does not seem to respect the approach enshrined in the institutional amendment it was meant to accompany. The Constitutional Treaty contained a definition of competences similar to that then included in the Treaty of Lisbon, which granted space to a role for harmonisation that would be more autonomous from cooperation.20 The Hague Programme does not include a trace of either of these, except for the mention of serious crime,21 whereas the reference can be found in the 2009 Stockholm Programme (discussed below). II.  THE INTEGRATED APPROACH IN THE 2009 STOCKHOLM JUSTICE AND HOME AFFAIRS PROGRAMME

The 2009 Stockholm Programme marks the distance with the previous multiannual programmes in many ways.22 The title, ‘The Stockholm Programme: An Open and Secure Europe Serving and Protecting Citizens’, already highlights the key role envisaged for individuals within the system. Such centrality is also recalled in the text when the European Council states that it considers it a priority to focus on the interests and needs of citizens.23 Moreover, the Programme spells out ‘respect for fundamental rights of citizens’ and the ‘concept of the rule of law’ for the first time as political priorities for the Area of Freedom, Security and Justice, and it devotes a whole section to each of them.24 These choices already show a reversal of perspective in comparison with the previous

18 See ch 4, s III.B.i. 19 Hague Programme (n 10) pt 3.3.2. The same is re-stated in the Council and Commission Action Plan implementing the Hague Programme on strengthening freedom, security and justice in the European Union [2005] OJ C198/1, pt 15. 20 Compare art III-271 of the Treaty establishing a Constitution for Europe [2004] OJ C310/1 with art 83 TFEU. 21 A Weyembergh, ‘Approximation of Criminal Laws, the Constitutional Treaty and the Hague Programme’ (2005) 42(6) Common Market Law Review 1567, 1584. 22 European Council, The Stockholm Programme – an open and secure Europe serving and protecting citizens [2010] OJ C115/1 (hereinafter ‘Stockholm Programme’). 23 ibid pt 1.1. 24 ibid.

The Stockholm Programme Integrated Approach  145 EU focus on the repressive side of criminal justice.25 In addition to these general considerations, the importance granted to the values-based perspective in criminal law can also be appreciated when the actual rationales for harmonisation are discussed. The cooperation rationale is present but attenuated, while the values-based criminalisation rationale of using harmonisation to suppress undesirable forms of conduct which affect fundamental interests is granted more space. Admittedly, when discussing mutual recognition in criminal matters, the European Council affirms that: ‘The approximation, where necessary, of substantive and procedural law should facilitate mutual recognition.’26 In the section explicitly devoted to the approximation of substantive criminal law, it reiterates that the EU may adopt common minimum rules ‘[t]o the extent necessary to facilitate mutual recognition of judgements and judicial decisions and police and judicial cooperation in criminal matters’,27 which points in the direction of a utilitarian legitimacy of EU criminal law. However, a few lines below, the Council problematises at least one aspect of this cooperation rationale. Chapter 3 clarified that the harmonisation of definition of crimes can be said to support judicial cooperation to the extent that it favours mutual trust among judicial authorities, that it clarifies the scope of action of judicial cooperation actors like Eurojust or Europol, or that it reduces differences between national legislations as to the definition of crimes, which thus makes compliance with the principle of double criminality more likely. This latter requirement, which is included in traditional judicial cooperation instruments, allow the states from which cooperation is requested to deny such cooperation if the facts at stake are not criminalised in its jurisdiction. The harmonisation of definitions of crimes can be instrumentally relied upon as a strategy to increase the chances of both states having similar legislation in order to overcome this potential hindrance to cooperation.28 The Stockholm Programme invites the Commission to put forward a more elaborate discussion on this connection: ‘The relationship between approximation of criminal offences or their definition and the double criminality rule in the framework of mutual recognition should be further explored. The Commission is invited to make a report to the Council on this issue. One of the issues may be the necessity and feasibility of approximation or definition of criminal offences for which double criminality does not apply.’29

25 M Kaiafa-Gbandi, ‘The Importance of Core Principles of Substantive Criminal Law for a European Criminal Policy Respecting Fundamental Rights and the Rule of Law’ (2011) 1(1) European Criminal Law Review 7, 9. 26 Stockholm Programme (n 22) pt 3.1.1. Interestingly, the Programme links harmonisation to mutual recognition, but not to mutual trust. When discussing the latter, there is no mention of legislation among the tools that might foster it. These listed tools include training or the establishment of networks (ibid pt 3.2). 27 ibid pt 3.3. 28 See the ch 3, s V.A. 29 Stockholm Programme (n 22) pt 3.3.1.

146  From Tampere to Stockholm The European Council admits that harmonisation of substantive criminal law might occur independently from cooperation needs, and thus EU criminalisation could also be legitimated on deontological grounds. However, there is no record at present of the Commission having begun any work on the matter. Moreover, the Stockholm Programme also acknowledges more explicitly the need for a further deontological approach to EU criminal law. It clearly states that: ‘Criminal law provisions should be introduced when they are considered essential in order for the interests to be protected and, as a rule, be used only as a last resort.’30 Even if it is considered that harmonisation should be primarily used to support cooperation, this cannot be done indiscriminately for all the crimes that present a cross-border dimension and on which the legislation in Member States differs. It should only be used as a tool to support cooperation when the EU considers that the interests at stake are worthy of protection via the criminal law. Briefly, the Stockholm Programme admits a need for an integrated approach to the legitimacy of EU criminal law, which combines cross-border enforcement needs and values-based considerations, similarly to what was argued with respect to competence definitions in Lisbon.31 Finally, the Stockholm Programme discusses the goal of fostering free movements, which was listed in Chapter 2 as one of the potential ‘social engineering’ goals for the harmonisation of criminal law. However, it does not link it to the harmonisation of crimes.32 After having laid out what it considers to be the sole general policy framework for EU action in criminal justice,33 the European Council then called on the Council of Ministers, the Commission and the Parliament to reflect together on how to improve the coherence of criminal law provisions in various EU instruments.34 Despite this call for coordination, the Commission and the Parliament felt a need to re-affirm their individual role as agenda-setters, especially in the post-Lisbon environment, where they had been granted a wider role in the law-making process. Hence, the Commission published its own policy document on the Justice and Home Affairs priorities35 and so did 30 ibid. 31 Lorenzo Salazar observes that the Stockholm Programme was adopted a few days after the entry into force of the Treaty of Lisbon and was the first programmatic document of the new ‘Lisbonised’ era because it implemented the new provision of the Treaty, at least formally. Yet, it can still be considered the last multi-annual programme to have been adopted by the European Council under the ‘old’ methodology of listing a number of priorities while not generally providing an overall planning: L Salazar, ‘EU’s Criminal Policy and the Possible Contents of the New Multi-annual Program’ (2014) 1 EuCrim 22, 23. 32 Stockholm Programme (n 22) pt 2.2. See also the discussion in ch 4, s III. 33 See Council of the European Union, Draft Council Conclusions on the Commission Communication ‘Delivering an area of freedom, security and justice for Europe’s citizens – Action Plan implementing the Stockholm Programme’ COM (2010) 171 final, doc 9935/10 (hereinafter ‘Council Conclusions on Stockholm Programme’). 34 Stockholm Programme (n 22) pt 3.3.1. 35 European Commission, Communication, ‘Delivering an area of freedom, security and justice for Europe’s citizens: Action Plan implementing the Stockholm Programme’ COM (2010) 171 final.

The Silence in the 2014 Justice and Home Affairs Strategic Guidelines  147 the Parliament,36 to which the Council in turn also responded, re-stating that the Stockholm Programme should be the only relevant frame of reference.37 These documents present no relevant insight on the role for EU criminalisation. The only exception is the Parliament’s interesting call for further discussion of the possibility of legislating on hate crimes and how to better protect EU values.38 These observations points towards a normative understanding of EU criminal law. III.  THE SILENCE IN THE 2014 JUSTICE AND HOME AFFAIRS STRATEGIC GUIDELINES

The European Council adopted the latest Justice and Home Affairs Strategic Guidelines in June 2014, which were meant to steer EU action in the Justice and Home Affairs field from 2015 to 2020.39 It was observed how the Programme was adopted in something of a rush. This was due to some specific political and institutional context, meaning the then-upcoming European Parliament elections in May 2014, the new President of the Commission being designated in June 2014, with the new College which was due to become operational a few months later, and the election of the new President of the European Council in October 2014. In addition, there was also the uncertain situation as to what would be the position of the UK after the end of the transitional period under Protocol 36. This ‘institutional jam’ arguably meant that the discussion on Justice and Home Affairs priorities was somehow rushed.40 Unlike the 2009 Stockholm Programme, which represented a quite revolutionary document in acknowledging the deontological function of EU criminalisation, the 2014 Strategic Guidelines do not provide any significant insights on the theme of the legitimacy of EU Criminal Law. In terms of horizontal issues, the European Council generally speaks of enhancing procedural rights, arguably through harmonisation, of facilitating cross-border activities and operational cooperation, and of enhancing the mutual recognition of decisions and judgments in criminal matters.41 Then the European Council identifies 36 European Parliament Resolution of 25 November 2009 on the Communication from the Commission to the European Parliament and the Council – An area of freedom, security and justice serving the citizen – Stockholm Programme, P7_TA(2009) 0090, B7-0155/2009 (Hereinafter ‘Parliament Resolution on Stockholm Programme’). For an analysis of the institutional politics behind the adoption of these various documents, see S Carrera and E Guild, ‘The European Council’s Guidelines for the Area of Freedom, Security and Justice 2020: Subverting the “Lisbonisation” of Justice and Home Affairs?’ (2014) 13/14 CEPS Policy Paper, 3–4, https://www.ceps.eu/system/files/ SC%20EG%20AFSJ%20Guidelines%202020.pdf. 37 Council Conclusions on Stockholm Programme (n 33). 38 Parliament Resolution on Stockholm Programme (n 26) pt 26. 39 European Council, Conclusions of 26 and 27 June 2014, doc 79/14 (hereinafter ‘2014 Strategic Guidelines’). 40 Salazar (n 31) 23. 41 2014 Strategic Guidelines (n 39) pt 11.

148  From Tampere to Stockholm some priority areas of crime, such as financial crime or terrorism, but it does not discuss whether further harmonisation of the definitions of crimes in these areas should be one of the suitable legal strategies.42 It was observed how the discourse that has surrounded the adoption of the June 2014 Strategic Guidelines has been filled with terms such as ­‘consolidation’, ‘implementation’, ‘evaluation’ and ‘no new legislation’.43 Similarly to what had happened with Stockholm, while the European Council was discussing the new strategic guidelines, the Commission also felt the need to put forward its position on the future of Justice and Home Affairs policies and it adopted three Communications relating to the future of the Area of Freedom, Security and Justice.44 However, in these documents, there is nothing on the rationales for the harmonisation of criminal law.45 When discussing the subject of new legislation, the Commission, similarly to the European Council in the Strategic Guidelines, also stressed the need for first consolidating what has already been achieved and, when necessary, codifying EU law and practice, and only then complementing the existing framework with new initiatives.46 The focus seems to be very much on the implementation of existing acquis rather than on future harmonisation and its possible functions.47 Similarly, the Parliament held a joint Committee meeting together with the national parliaments on the ‘Future Priorities in the Field of Civil Liberties, Justice and Home Affairs’, the conclusions of which devoted particular attention to the EPPO and judicial cooperation in criminal matters, but nothing specific on harmonisation.48 IV.  THE EU CRIMINALISATION POLICY DOCUMENTS: THE CORE OF THE DISCUSSION

Next to these general programmatic documents, the legislative Institutions also adopted ad hoc policy documents on the subject of the harmonisation of 42 ibid pts 10–11. 43 HG Nilsson, ‘Where Should the European Union Go in Developing its Criminal Policy in the Future?’ (2014) 1 EuCrim 19, 19. This author nonetheless claims that there is still a need to intervene with further harmonisation in certain specific sectors, especially crimes against the EU financial interests, money laundering and organised crime. 44 See European Commission ‘The EU Justice Agenda for 2020 – Strengthening Trust, Mobility and Growth within the Union’ COM (2014)144 (hereinafter ‘EU Justice Agenda for 2020’); E ­ uropean Commission, ‘An open and secure Europe: making it happen’ COM (2014)154. See also the communication on the rule of law, which is arguably also relevant to the Area of Freedom, Security and Justice: European Commission, ‘A new EU Framework to Strengthen the Rule of Law’ COM (2014) 158 final/2. 45 For a comment on these Communications, see E Herlin-Karnell, ‘All Roads Lead to Rome: The New AFSJ Package and the Trajectory to Europe 2020’ (2014) 1 EuCrim 27. 46 ‘EU Justice Agenda for 2020’ (n 44) pt 4. 47 Salazar (n 31) 24. 48 A summary of the meeting conclusions can be found at: https://www.politiekemonitor.nl/ 9353000/1/j9tvgajcor7dxyk_j9vvioaf0kku7zz/vjfg72yzniwt?ctx=vh7ykb8bjvyr&tab=1.

The EU Criminalisation Policy Documents  149 criminal law. In keeping with their ‘uncoordinated’ post-Stockholm approach, the Commission, the European Parliament and the Council enacted three separate EU criminalisation policy documents that are discussed more in detail in the next section. Already in 2005, the Commission enacted first a Communication on the mutual recognition of judicial decisions and mutual trust envisaging a clear instrumental rationale for harmonisation of definitions of crimes, as a tool to build trust, and thus a utilitarian legitimacy for EU criminal law.49 Similarly, in the Communications which followed the two Environmental Crime and Ship-Source Pollution crime judgments in 2005 and 2007, the Commission extensively elaborated on the utilitarian legitimacy for EU criminal law as a tool to secure the enforcement of EU policies, thus attributing a regulatory rationale to harmonisation.50 Then, while the Treaty of Lisbon was being discussed and approved, each of the EU legislative Institutions (namely, the Council, the Commission and the Parliament) adopted more exhaustive policy documents precisely on the topic of the harmonisation of substantive criminal law. The reason why there are three separate ‘EU Criminalisation Policy documents’51 rather than one comprehensive study is arguably linked to inter-institutional politics, and the desire of each Institution to re-affirm its own role in the law-making process. In November 2009, directly before the entry into force of the Treaty of Lisbon, the Council enacted a document on model provisions guiding the Council deliberation in criminal matters (hereinafter the 2009 Council Conclusions).52 Vervaele argues that the Justice and Home Affairs formation of the Council, which had been in charge of adopting criminal law in the third pillar, wished to set some guidelines to re-affirm its key role as a law-maker to the other players in the field. For instance, other formations of the Council (for instance, those including the minister of health or agriculture) could also play a role after Lisbon if criminal

49 European Commission, ‘Communication on the mutual recognition of judicial decisions in criminal matters and the strengthening of mutual trust between Member States’ COM (2005)195 final, pts 18, 28–29. 50 ‘Communication from the Commission to the European Parliament and the Council on the implications of the Court’s judgment of 13 September 2005 (Case C-176/03 Commission v Council)’ COM (2005) 583 final/2 (hereinafter ‘Communication on the Environmental Crime Case’). 51 These documents are referred to as such in P De Hert and I Wieczorek, ‘Testing the Principle of Subsidiarity in EU Criminal Policy: The Omitted Exercise in the Recent EU Documents on Principles for Substantive European Criminal Law’ (2012) 3–4 New Journal of European Criminal Law 394. 52 Council of the European Union, Council Conclusions on model provisions, guiding the Council’s criminal law deliberations, JHA Council, 30 November 2009 (hereinafter ‘2009 Council Conclusions’), included in Weyembergh and Galli (n 2) 226. Interestingly, the Council had in the past already set out documents on the approximation of penalties (Council conclusions on the approach to apply regarding approximation of penalties, JHA Council of 27 May 2002, doc 9141/02) and on the procedure for handling of legislative files containing proposals relevant to the development of criminal law policy (Council conclusions on the procedure for handling of legislative files containing proposals relevant to the development of criminal law policy, JHA Council of 21 February 2006, doc 7876/06).

150  From Tampere to Stockholm law was used to enforce a specific EU policy on the basis of Article 83(2) TFEU.53 Moreover, the extension of the co-decision procedure to the field of EU criminal law naturally meant that the Commission and the Parliament also played a role in the adoption of criminal justice matters. The Council Conclusions were thus enacted in a manner that was more like a summary of past performance than a prospective criminal policy document fully taking into account the substantive changes introduced by the Treaty of Lisbon.54 The Commission first openly criticised the unilateral decision of the Council to establish such a framework for developments in criminal matters.55 Then, in 2011, it enacted its own Communication on EU criminal policy (hereinafter the 2011 Commission Communication).56 The text was on the programme of Commissioner Reading since the beginning of her mandate, but saw the light of day only in September 2011, which leads one to think that it was actually a response to the Council.57 Interestingly, while the Council’s 2009 Conclusions deal with the whole of the EU competence to harmonise substantive criminal law on the basis of Article 83 TFEU, the Commission dedicates an ad hoc communication solely to the principles guiding the exercise of the EU competence in the field of functional criminalisation on the basis of Article 83(2) TFEU. Finally, in May 2012, the Parliament also approved a resolution on the theme of EU criminalisation (hereinafter the 2012 Parliament Resolution) addressing the whole of EU competence in this field, in a similar manner to the 2009 Council Conclusions.58 Likewise, this document has been interpreted as a reaction to the aggressive unilateral move of the Council,59 even though the Parliament seems

53 See contra Sicurella, who interprets the Council’s conclusions as a genuine effort to take a stance in favour of a more liberal and principled-based criminal law. She claims that the text goes beyond being simple law-making guidelines: R Sicurella, ‘“Prove Tecniche” per una metodologia dell’esercizio delle nuove competenze concorrenti in Materia Penale’ in G Grasso, L Picotti and R Sicurella (eds) L’evoluzione del diritto penale nei settori d’interesse europeo alla luce del Trattato di Lisbona (Milan, Giuffrè, 2011) 1, 896. 54 J Vervaele, ‘Harmonised Union Policies and the Harmonisation of Substantive Criminal Law’ in Weyembergh and Galli (n 2) 43, 61. See also A Weyembergh, ‘Introduction’ in Weyembergh and Galli (n 2) 16, 24. Mitsilegas observes that the Member States sitting in the Council wanted to send out a message on the extent and limits of EU competence to criminalise after Lisbon: V Mitsilegas, EU Criminal Law after Lisbon (Oxford, Hart Publishing, 2016) 69. 55 The Commission issued a declaration stating that it considered such guidelines premature and unnecessarily restrictive on the interpretation of art 83 TFEU. In doing so, it continued, the Council unilaterally established a framework for future legislation to which neither the Commission nor the Parliament agreed. The text of the declaration can be found in the minutes of the Council meeting of 30 November – 1 December, doc 16826/09, p 22. 56 European Commission, ‘Towards an EU Criminal Policy: ensuring the effective implementation of EU policies through criminal law’ COM (2011) 573 final (hereinafter ‘2011 Commission Communication on an EU Criminal Policy’). 57 Weyembergh (n 54) 24. 58 European Parliament, Resolution of 22 May 2012 on an EU approach to criminal law, P7_ TA(2012)0208, A7-0144/2012 (hereinafter ‘2012 Parliament Resolution on an EU approach to criminal law’). 59 Weyembergh (n 54) 25.

The EU Criminalisation Policy Documents  151 more open to establishing the inter-institutional cooperation that the European Council was calling for in the Stockholm Programme.60 Be that as it may, regardless of the specific institutional political reasons that led to the adoption of each of these documents, each of them is highly relevant for the discussion on the legitimacy of EU criminal law. The mere fact that the Institutions embarked on the task of defining when the EU competence in criminal matters should be exercised already attests to the fact that they acknowledge the need for EU criminal law to also benefit from legitimacy stemming from the selection of specific subjects and the pursuit of specific interests, and not simply from the respecting of competence boundaries and procedures.61 The rapporteur for the European Parliament Resolution even openly acknowledged this legitimacy-seeking function of the Institutions’ policy documents.62 A quote from his speech was reported in full at the beginning of the book, since it so well illustrates the need to look into the justifications for EU criminal law. Alongside this, the text of each document includes interesting insights on the relevant Institution’s understanding of the legitimacy of EU criminal law. The next three sections discuss how the theme was addressed in the 2009 Council Conclusions on model provisions (see section IV.A), the 2011 Commission Communication on an EU Criminal Policy (see section IV.B) and the 2012 Parliament Resolution on an EU approach to criminal law (see section IV.C). A.  The 2009 Council’s Conclusions: A Deontological, Harm-Based Approach to EU Criminal Law The 2009 Council Conclusions originated from a Swedish presidency initiative. Their declared purpose is to establish guidelines to avoid inconsistency in negotiations, transposition and the legal interpretation of legislative texts containing criminal law provisions.63 To achieve this aim, the text states that particular attention must be granted to the wording of the legislative texts, especially after the Treaty of Lisbon.64 The document is a relatively short, five-page text, clustering the selected guidelines under six headings. The first heading concerns ‘Assessment of the Need for Criminal Provisions’. After this, aspects related to

60 Stockholm Programme (n 22). 61 Klip argues that the fact that the Commission raised the question of the added value of European criminal law shows its awareness that EU criminal law is not self-evident and actually needs some support: A Klip, ‘European Criminal Policy’ (2012) 20 European Journal of Crime, Criminal Law and Criminal Justice 3, 4. 62 See more broadly Introduction, s III. 63 2009 Council Conclusions (n 52) introduction. 64 ibid.

152  From Tampere to Stockholm the content of the provisions and to their structure are dealt with, under the headings ‘Structure of Criminal Provisions’, ‘Intent’, ‘Inciting, Aiding and Abetting and Attempt’, ‘Penalties’ and ‘Model Provisions’. Finally, in line with the announced pragmatic approach, the second part of the document provides proper examples of model provisions. The most relevant heading is the first in which the Council proposes a threefold test to assess whether the legislator should resort to (EU) criminal law: Assessment of the need for criminal provisions (1) Criminal law provisions should be introduced when they are considered essential in order for the interests to be protected and, as a rule, be used only as a last resort. (2) Criminal provisions should be adopted … a) in the areas of particularly serious crime with a cross-border dimension resulting from the nature or impact of such offences or from a special need to combat them on a common basis, or b) if the approximation of criminal laws and regulations of the Member States proves essential to ensure the effective implementation of a Union policy in an area which has been subject to harmonisation measures. (3) When there seems to be a need for adopting new criminal provisions the following factors should be further considered, while taking fully into account the impact assessments that have been made: the expected added value or effectiveness of criminal provisions compared to other measures, taking into account the possibility to investigate and prosecute the crime through reasonable efforts, as well as its seriousness and implications.

Alongside this, the Council makes another relevant remark at point 5 when discussing the ‘Structure of Criminal Provisions’ heading, where it prescribes that: The criminal provisions should focus on conduct causing actual harm or seriously threatening the right of essential interest that is the object of protection; that is avoiding criminalisation of a conduct at an unwarrantably early stage. Conduct which only implies an abstract danger to the protected right or interest should be criminalised only if appropriate considering the particular importance of the right of interest which is the object of protection.

Finally, point 6 within the ‘Intent’ category also interestingly reads as follows: Negligent conduct should be criminalised when a case-by-case assessment indicates that this is appropriate due to the particular relevance of the right or essential interest which is the object of protection, for example in cases of serious negligence which endangers human life or causes serious damages.

At least three observations can be made on the position of the Council. First, as mentioned, the Council makes clear that the discussion applies to both EU criminalisation in the area of cross-border crimes as well as when criminal law is

The EU Criminalisation Policy Documents  153 used to enforce EU policies, namely to the whole of Article 83 TFEU (heading 1, point 2).65 Second, the Council does not speak of harmonisation, but rather of criminalisation and of the use of criminal law. From a methodological point of view, this implies that the Council’s approach to the criminal law does not have to be inferred from the functions attributed to harmonisation, but can be directly gleaned from the Council’s language. From a substantial point of view, this means that the Council does not refer to any of the other specific rationales for harmonisation, namely support for judicial cooperation, spreading a sense of justice, fostering free movements or building a cohesive society. This rules out an understanding of criminal law as instrumental to the enforcement needs of single Member States or of the achievement of any social engineering goals. Third, and most importantly, when specifically discussing the role of criminal law, the Council shows a pronounced inclination towards a deontological understanding for the legitimacy of EU criminal law. It establishes that that criminal law is to be relied upon only to protect a particularly important interest and that it must be used only when essential to this aim (heading 1, point 1). It then provides further detail on what elements should be taken into account in designing EU offences (heading 1, point 3). Among these, admittedly the Council considers the impact of criminalisation (the expected added value or effectiveness of criminal provisions), which is a utilitarian, objectiveoriented perspective. Yet, it places this justification next to the deontological perspective stemming from the nature of the conduct at stake (the seriousness and implications of the crime). The approach of the Council is at most an integrated one. Fourth, the Council also provides at least some details on the characteristics of the interests that criminal law is meant to protect. It makes it explicit that there is a hierarchy in terms of importance between the various interests that might deserve criminal protection and that the structure of the criminal provisions might depend on these interests. For instance, potentially harmful or negligent conduct can be criminalised only if these respectively put in

65 Interestingly, the Council does not mention the wording of art 325(4) TFEU anticipating that the Member States did not wish to include this provision among the suitable legal bases for criminal law. Indeed, when the Commission proposed the PIF Directive on the basis of this provision, the Member States in the Council opted for the alternative of art 83(2) TFEU (see ch 4, s III.A). In any case, should the Council in the future adopt a different position and negotiate criminal legislation to be adopted on the basis of this provision, it is reasonable to think that it would follow these very same guidelines. See by analogy Harding and Bannach Guitierrez, who make a similar observation with respect to the scope and the applicability of the 2011 Commission Communication on an EU Criminal Policy (n 56): C Harding and JB Banach-Gutierrez, ‘The Emergent EU Criminal Policy: Identifying the Species’ (2012) 37 European Law Review 758.

154  From Tampere to Stockholm danger or negligently affect very highly ranked interests (headings 5 and 6). This implies that there are less important interests whose protection by means of the criminal law is justified only when what is at stake are directly harmful forms of conduct. Furthermore, one can appreciate that the Council’s very specific language – ‘actual harm to a fundamental interest’ (heading 6) – is reminiscent of the Anglo-Saxon liberal discourse on criminal law, which focused on the harm deriving as a consequence of the relevant conduct, as opposed to the inherent contrast between the conduct and the current morality. Nonetheless, it was observed that while the Council manifests such a strong inclination towards an interests-based criminal law, it does not provide any examples of what these interests might be.66 Finally, the Council also leaves the door open for some criminalisation of forms of conduct that do not directly harm the protected interest, but threaten to do so, namely preventive criminalisation.67 Briefly, the Council envisages the need for a strong deontological legitimacy for EU criminalisation. Resort to criminal law must be grounded in the harm that a conduct can cause to a specific and particularly important interest and not simply on the positive outcome that can be derived from criminalisation or rectius harmonisation. B.  The 2011 Commission’s Communication: Hinting at an Integrated Approach The Commission Communication was published in 2011, with the declared aim of providing a framework for the further development of an EU criminal policy under the Treaty of Lisbon. The document is a longer text (12 pages) compared to the Council Conclusions. Moreover, contrary to the Conclusions, which were presented as a grid to follow closely when drafting legislation, the Commission Communication openly presents itself, starting from its title, as a broad policy document and attempts to address the theme of EU criminal policy from a broader and more theoretical perspective. The fact that the Commission traditionally holds the right of initiative and that it is thus an agenda-setter could explain this announced and more encompassing approach. However, even if the idea of a policy-shaping exercise is commendable per se, it was rightly observed that the Commission has hardly fulfilled its promise. Moreover, important aspects that one would expect in a Communication ambitiously entitled ‘EU Criminal Policy’ are neglected. The Commission focuses on legislative aspects, and actually only on definitions of crimes, whereas it does not provide any detailed discussion on sanctions or procedural aspects.68



66 Vervaele

(n 54) 61. Council Conclusions (n 52) pt 5. 68 Weyembergh (n 52) 25. 67 2009

The EU Criminalisation Policy Documents  155 For instance, there is no discussion of how to coordinate and regulate the enforcement aspects during the judicial phase.69 In light of this, a more appropriate title for the Communication would have probably been ‘Towards an EU Criminal Law Policy’ or even ‘Towards a EU Criminalisation Policy’.70 In terms of material scope, the Communication opted for a restricted approach. Admittedly, part 1 of the document discusses the added value of EU criminal law in general, in line with the policy-shaping function mentioned above. It mentions both Article 83(1) TFEU-based securitised EU criminalisation and Article 83(2) TFEU-based functional criminalisation, and even the possibility of having Article 325(4) TFEU-based EU criminal law to protect the financial interests of the EU. The Commission would propose the Directive for the Protection of Financial Interests of the EU on this latter legal basis.71 This part is discussed more in detail below. Yet, after this broad introduction, the Commission declares that it aims only at providing guidance with respect to EU criminal law ensuring the effectiveness of EU law, that is, harmonisation based on Article 83(2) TFEU. It thus only concerns itself with framing the use of criminal law to enforce EU policies.72 Within this limited scope, the Commission nonetheless puts forward a number of interesting statements for the debate on the legitimacy of criminal law. The following paragraphs focuses on what insights can be gained respectively from the introduction of the Communication from part 2 on General Principles and from part 3 listing the policy areas deserving criminalisation. Finally, some concluding remarks are drawn on the Commission’s approach. i.  The Introduction: A Strong Focus on a Utilitarian Legitimacy for Criminal Law The Commission acknowledges in the introduction that criminal law is a sensitive policy domain and by nature an intrusive field, and that therefore there should be important limits for EU action in this area.73 Moreover, it acknowledges that this field of law is linked to national traditions.74 Yet, no specific policy implications are derived from the special nature of criminal law at this stage. In other words, at least at this stage, the Commission does not explicitly 69 Klip (n 61) 11. 70 On the subject of what constitutes an EU criminal policy and an EU criminal law policy, see A Weyembergh and I Wieczorek, ‘Is There an EU Criminal Policy?’ in R Colson and S Field (eds), EU Criminal Justice and the Challenges of Legal Diversity (Cambridge, Cambridge University Press, 2016) 29. 71 See ch 4, s III.A and ch 7, s V. 72 It was observed that ‘it is inconceivable that many of the general points of argument in the document would not be applicable to the whole field of EU criminal law, especially in view of the Commission’s stated purpose of providing greater coherence and consistency in this field of legal activity’. See Harding and Banach-Gutierrez (n 65) 759. 73 2011 Commission Communication on an EU Criminal Policy (n 56) introduction. 74 ibid.

156  From Tampere to Stockholm list the selection of specific Rechtsgut among the possible limits to the use of criminal law. Subsequently, when discussing the general added value of European criminal law, the Commission enumerates four reasons in favour of supranational, as opposed to national, criminalisation at the EU level. First, it would foster citizens’ confidence in the fact that they live in an Area of Freedom, Security and Justice, that EU law is correctly implemented and that there is full respect of proportionality and subsidiarity. As a consequence, it would foster their confidence in making use of their right to free movement. Second, having common rules will avoid having crime safe havens across Europe, which criminals could avoid. Third, it will enhance mutual trust among judicial authorities. Fourth, it would ensure the effective enforcement of EU policies.75 André Klip has been very critical towards these arguments, claiming that in substance, there is no evidence supporting the first three points and that possibly only the fourth (ensuring the effectiveness of EU policies) is a valid one.76 Moreover, other authors have condemned the approach relying on citizens’ confidence as a legitimating factor for criminal law. The argument is that the logic of legitimation is in this way upset. Citizens’ confidence should be the effect and not the cause of criminalisation.77 Still, regardless of the actual soundness of these four arguments, their mentioning is already telling in relation to the approach to the legitimacy of EU criminal law. The Commission recalls three rationales for harmonisation that were associated with a utilitarian understanding of criminal law, namely the pursuit of social engineering goals (when it speaks of fostering of free movement), the cooperation rationale (when it speaks of easing mutual trust) and the regulatory rationale (when it speaks of enforcing EU policies). The focus seems to be very much on the objectives that criminal law can achieve (utilitarian legitimacy) rather than on the interests it must protect (deontological legitimacy). Admittedly, the Commission also mentions the forum shopping argument, that is, harmonising definitions of crimes to avoid the creation of safe havens for criminals. This argument was interpreted in Chapter 3 as an expression of the values-based criminalisation rationale for harmonisation, namely using harmonisation of definitions of crimes in order to ensure that particularly undesirable forms of conduct are suppressed throughout Europe. It was argued that in order to speak of ‘safe havens’, one must have first necessarily carry out a normative judgement on which forms of conduct deserve and which do not deserve to be considered lawful.78 In mentioning the safe havens argument, the Commission thus hints at the need for the deontological legitimacy of EU criminal law. Yet, it

75 ibid. 76 Klip (n 61) 4–5. 77 S Manacorda, ‘Le programme pour une politique pénale de l’Union entre mythe et réalité’ (2011) 4 Revue de science criminelle 900, 902. 78 See ch 3, s IV.

The EU Criminalisation Policy Documents  157 does so only implicitly, without using a strong values-inspired language, as did the European Council in the 2009 Stockholm Programme and the Council of Ministers in the 2009 Council Conclusions on model provisions, when explicitly stating that criminal law must only be used to protect selected interests.79 ii.  General Principles Justifying the Resort to Criminal Law: A Hinted-at Utilitarian Approach The second part of the Communication indicates the relevant principles that should constitute the core of the EU criminal policy. The Commission first stresses the importance of the respect of subsidiarity and fundamental rights. Interestingly, it lists the relevant international and European fundamental rights instruments as sources of fundamental rights, but does not mention Article 6 TEU. This means that national constitutional traditions, which are mentioned in the Treaty as sources of EU fundamental rights, thus remain out of the picture. It is not clear whether this omission is intentional and whether it is meant to express a stronger focus on EU sources for general principles of EU law or not.80 After this general remark, the Commission envisages a more specific two-step approach in criminal law legislation: the first step requires that a ‘decision on whether to adopt criminal law measures at all’ must be taken; the second step includes respecting the ‘principles guiding the decision on what kind of criminal law measures to adopt’.81 This sub-division is similar to the one made by the Council between ‘Assessment of the need for criminal law’ and ‘Structure of the provisions’.82 When discussing when ‘to adopt criminal law measures at all’, the Commission starts by recalling the important consequences of criminal investigations on individuals, namely that they may have a significant impact on citizens’ rights and include a stigmatising effect. The decision to criminalise should thus be based on ‘necessity and proportionality’ and it adds that criminal law must always remain a measure of last resort (the ultima ratio principle).83 Interestingly, the Commission establishes an explicit link between last resort and proportionality: ‘criminal law must remain a measure of last resort. This is reflected in the general principle of proportionality (as embodied in the Treaty on European Union [Article 5(4) TEU] and, specifically for criminal penalties, in the EU Charter of Fundamental Rights [Article 49(3) of the EU Charter])’.84 Harding and Banach-Gutierrez maintain that the mention of the important impact that criminal law can have on the position of the individual and of the



79 See

s II, and s IV.A. (n 54) 62–63. 81 2011 Commission Communication on an EU Criminal Policy (n 56) pt 2.2. 82 See s V. 83 2011 Commission Communication on an EU Criminal Policy (n 56) pt 2.2. 84 ibid pt 2.2.1. 80 Vervaele

158  From Tampere to Stockholm principle of ultima ratio are evidence of a change of tone in the Commission Communication, which in the past relied on the ‘familiar supranational call for “effective, dissuasive and proportionate” measures to ensure implementation of international and European policies and principles, insofar as that has often been interpreted as a requirement of tough action to ensure effect’.85 These authors see in the acknowledgement of the special character of criminal law and the weight granted to ultima ratio a shift in emphasis from ‘dissuasive’ (and one could also add effective) sanctions to ‘proportionate’ sanctions.86 If such attenuation of the focus on the effectiveness of the criminal law is combined with the call for the respect of fundamental rights and the inclusion of the principle of retrospective proportionality, which arguably imposes a values-based dimension in the sanctioning judgement,87 a hint towards a deontological understanding of the legitimacy of the criminal law can arguably be seen. Nevertheless, later on in this section, the Commission underlines the need for a detailed impact assessment, and a strong emphasis is placed on the importance of evidence-based justifications for the effectiveness of criminal law and of an ad hoc policy specific analysis.88 In particular, the success of Member States’ sanction regimes in achieving the desired results and the difficulties faced by national authorities implementing EU law must be considered jointly with the specificity of the policy area which is a decisive factor for the success of EU law. Briefly, what matters is not the substance of the policy area – that is, the interests at stake – but the degree of success in implementing EU norms. No values-based Rechtsgut discussion is present. Instead, there is a focus on the desired final result – the implementation of EU law – which points towards a utilitarian approach. This hinted-at utilitarian perspective is nonetheless attenuated in the following section of the Communication, which identifies the criteria for selecting a policy area and which incorporates some content limitations, and arguably points in the direction of a deontological understanding of the criminal law. This is discussed below. iii.  The Selection of Policy Areas Deserving Criminal Sanctions: A Hinted-at Integrated Approach The third part of the Communication lists a number of policy areas in which enforcement via criminal law should be envisaged. The Commission mentions three areas where it has already been established that criminal law will be resorted to, namely the financial sector, the fight against fraud and the protection of the euro against counterfeiting through criminal law. It then adds six

85 Harding 86 ibid. 87 See

and Banach-Gutierrez (n 65) 764.

ch 2, s II.B.2. Commission Communication on an EU Criminal Policy (n 56) pt 2.2.1.

88 2001

The EU Criminalisation Policy Documents  159 areas where the role of criminal law could be explored, namely road transport, data protection, customs rules, environmental protection, fisheries policy and internal market policies. When deciding in favour of criminal law or other alternatives such as administrative measures in these policy areas, the seriousness and the character of the breach of law must be considered, as for most serious crimes penal law is probably the most suitable instrument (the first criterion). In addition, criminal law must be chosen when strong disapproval is needed in order to ensure deterrence (the second criterion) and it must be considered that criminal proceedings provide for stronger protection of the rights of the accused (the third criterion). Conversely, one of the assets of the administrative system is that it offers a more efficient sanction system, which is more varied and more flexible, and therefore sanctions can be better tailored (the fourth criterion). This part of the Communication is probably the most interesting, as it sets a concrete agenda. This effort is undoubtedly the real added value of the Communication. Yet the choices in the agenda and the listed criteria have encountered some criticism in the literature. Vervaele, for instance, observes that areas in which there is already a criminal law acquis or where there have been proposals, such as counterfeiting and piracy of products, feed and food safety and corruption, are surprisingly absent from the list.89 For the purposes of our discussion on the legitimacy of EU criminal law, we can retain the following from this part of the Communication. The Commission starts with what seems a moderate integrated approach. It states that the selection of forms of conduct that deserve to be criminalised must also include ‘seriousness’ and that the ‘character’ of the breach must be taken into account (the first criterion). If we also apply in this context the interpretation of ‘serious crime’ as crime harmful to fundamental interests, which was proposed in Chapter 4,90 the statement of the Commission also incorporates a deontological perspective. However, even if the Commission did not envisage such interpretations of the term ‘serious crime’, the overall approach to the selection of which norms should be enforced by the means of criminal law is telling. Not all the breaches of law deserve criminal sanctions, even when criminal law is the only effective tool to ensure law-abidance. A selection must be made, including only the most serious breaches, whatever this means. As a consequence, even where the use of criminal law would be effective and other means would fail, penal sanctions should still not be envisaged because the breach is not sufficiently serious. However, the Commission quickly turns to a strictly effectiveness-based perspective. It recalls the purely utilitarian wording of the Treaty as to the need to use criminal sanctions to ensure the effective implementation of a policy area. It further speaks of using criminal law, which carries with it a strong sense



89 Vervaele 90 See

(n 54) 64. ch 4, s III.B.i.

160  From Tampere to Stockholm of disapproval, in order to ensure deterrence (the second criterion). Here the perspective is clearly utilitarian. The final goal is deterrence. Criminal law is selected to achieve deterrence because, in light of its symbolic character, it can ensure greater effectiveness. The ethical character of criminal law is only instrumental to law-abidance. A deontological perspective would impose an inverted logic: it would first look at what kinds of conduct deserve strong social disapproval, then it would resort to criminal law to regulate these types of behaviour, which in light of its strongest ethical connotation is also the most effective enforcement tool. According to the deontological perspective, the social disapproval would precede the criminalisation process rather than following it, as the Commission seems to suggest. Herlin-Karnell even sees a contradiction between connecting effectiveness to the ethical character of criminal law and the traditional understanding of effectiveness within EU law as a principle being based on objective criteria.91 Lastly, the final remarks of the Commission on the rights of the accused and on the higher flexibility of administrative sanctions (the third and fourth criteria) cannot be considered expressions of a clear theoretical vision on criminal law. They are instead some pragmatic observations on when it is more convenient to use a particular set of sanctions. In this respect, Manacorda has remarked that the ultima ratio is interpreted in an odd way. The Commission does not interpret the principle as requiring the choice of criminal law only as a last resort, but simply as a criterion implying weighing up the pros and cons of the choice of criminal law. It seems that some arguments can be listed in favour of criminal law and against administrative sanctions such as the presence of stronger guarantees for the accused. If this is undeniable, this commentator argues, this reasoning certainly cannot be used as a ground for criminalisation. Indeed, the ultima ratio logic is not susceptible to also being reversed in favour of criminal law.92 iv.  Concluding Remarks: An Integrated Agenda with an Emphasis on a Utilitarian Approach The Commission Communication has been criticised for failing to follow a self-evident structure and for repeating similar observations in different parts of the document.93 The previous paragraphs similarly showed that relevant insights on the question of the legitimacy of EU criminal law are scattered throughout the whole text. Still, what emerges from the analysis above is that the Commission seems to have a general utilitarian understanding of

91 E Herlin-Karnell, The Constitutional Dimension of EU Criminal Law (Oxford, Hart Publishing, 2012) 58. 92 Manacorda (n 77) 907. 93 Harding and Banach-Gutierrez describe the Communication as repetitive, randomly ordered and elliptical: Harding and Banach-Gutierrez (n 65) 763.

The EU Criminalisation Policy Documents  161 EU criminal law as a means to several ends, namely enhancing mutual trust, gaining citizens’ confidence and enforcing EU policies. This perspective is partially nuanced in favour of a more integrated utilitarian and deontological perspective when actually selecting the specific conduct deserving to be punished through the use of criminal law. The Commission unsurprisingly insisted on the need to use criminal law where it is effective in terms of achieving the enforcement goal, yet only in the most serious cases in which this occurs rather than in all cases. It is interesting to compare this Communication with the one published directly after the Environmental Crime case. As discussed in Chapter 5, in 2005 the Court of Justice acknowledged a regulatory rationale for harmonisation: harmonisation of criminal law is justified if it is necessary to ensure the enforcement of EU norms. Yet, it arguably combined this predominant utilitarian perspective with a deontological one when it insisted on the fundamental character of the EU objectives that the relevant norms pursue. The Commission had nonetheless interpreted this case law quite broadly, only focusing on the utilitarian justifications for the use of EU criminal law. The only test to establish if criminal sanctions should be used was the ‘essentiality to secure enforcement’ test.94 The Treaty framework arguably codified the Court’s case law, but by embracing this purely utilitarian perspective, namely dropping any substantial requirement as to what sort of policies deserve to be enforced through criminal law. Interestingly enough, the deontological perspective returns, at least to a certain extent, in the 2011 Commission Communication, when it restricts the use of criminal law as an enforcement tool to only the most serious breaches of law. This latter approach can be defined as an integrated approach, but with an emphasis on the utilitarian dimension of European criminal law. C.  The 2012 Parliament Resolution: An Integrated, Damage-Based Approach The Parliament is the third Institution that published a ‘EU Criminalisation Policy’ document in 2012, the ‘European Parliament resolution of 22 May 2012 on an EU approach to criminal law’. Cornelis de Jong, the rapporteur for the Resolution, openly acknowledged the influence that the initiatives of the other Institutions had on the decision to publish such a report.95 Indeed, if compared with the 2009 Council Conclusions and the 2011 Commission Communication, the document issued by the Parliament is more comprehensive, showing that it has taken the previous documents into account. Moreover, the Parliament’s Resolution is also more constructive, envisaging future developments in light



94 Communication 95 De

Jong (n 2) 39.

on the Environmental Crime Case (n 50). See ch 5, s III.C.

162  From Tampere to Stockholm of the new Lisbon possibilities to be carried within a framework of interinstitutional cooperation rather than being aimed at stressing what specific role is to be envisaged for the Parliament. In the same piece, de Jong also stresses what can already be derived from the quote given in the Introduction.96 He acknowledges that the EU faced a crisis of social legitimacy – that is, of legitimacy in the eyes of the beneficiaries of the system – when the Constitutional Treaty was rejected. Given that the Treaty of Lisbon re-introduces many of the institutional novelties, especially in criminal matters, without additional referenda in the two countries that had rejected the Constitutional Treaty (namely France and the Netherlands), there is the risk of not acting in accordance with the wishes of the electorate. This is why, he claims, it is of the utmost importance to develop at least a framework for law-making in the field of criminal matters where new competences have been introduced. He states that the Commission has often seen law-making as an end in itself, without considering which criteria should guide the exercise of the competence, and that it is not the case to repeat this approach with respect to the newly introduced competences in criminal matters.97 Concerning the scope of the Resolution, as occurred in the case of the Council, but unlike the Commission, the Parliament makes clear that the considerations in its report apply to the whole of EU criminal law approved on the basis of Article 83 TFEU. The Resolution recalls the wording of both paragraphs: Article 83(1) on cross-border crimes and Article 83(2) on the use of criminal law to enforce EU policies.98 Interestingly, the Parliament stresses that ‘the introduction of EU criminal law provisions is not confined to the Area of Freedom, Security and Justice but can relate to many different policies’. This could refer to the harmonisation of substantive criminal law aimed at enforcing other EU policies as envisaged by Article 83(2) TFEU. Alternatively, it could relate to other possible legal basis in the Treaty falling outside the Area of Freedom, Security and Justice, such as Article 325(4) TFEU on the protection of the financial interests, which had been mentioned as a potential legal basis for EU criminal law.99 However, this second scenario is less likely. Indeed, as was mentioned above, when the Commission actually proposed legislation on financial crimes on the basis of Article 325(4) TFEU, the Parliament conversely insisted on having it adopted on the basis of Article 83(2) TFEU, which it considered lex specialis.100 In terms of length and content, the Parliament Resolution is an extensive text, including 15 ‘recitals’ and 20 recommendations which address approximately



96 ibid

38. See Introduction, s III. Jong (n 2). 98 2012 Parliament Resolution on an EU approach to criminal law (n 58) pt 3.3. 99 See ch 4, s III. 100 ibid. 97 De

The EU Criminalisation Policy Documents  163 five themes: general observations on the role of the harmonisation of definitions of crimes in relation with mutual recognition (the first theme),101 an aspect that was not clarified in the Council Conclusions and only partially in the Commission Communication;102 principles governing the decision to resort to European criminal law (the second theme);103 and principles governing the content of European criminal law provisions (the third theme).104 In addition, the Parliament also makes a brief reference to the importance of also ensuring the enforcement of EU substantive criminal law (the fourth theme).105 The remaining part of the document is devoted to procedural aspects concerning institutional and inter-institutional arrangements to ensure the respect of the identified criminal policy principles, among which the call for an inter-institutional agreement (the fifth theme).106 The themes relevant for our discussion are the first two. When discussing the first theme – the general role of harmonisation – the Resolution recommends that harmonisation measures be proposed primarily with a view to supporting the application of the principle of mutual recognition in practice rather than merely expanding the scope of harmonised EU criminal law.107 This acknowledgement of the cooperation rationale shows a quite bold utilitarian understanding of the criminal law: what matters are the transnational enforcement needs of Member States. Nevertheless, the deontological dimension is given more attention when discussing the second theme – the specific EU criminalisation criteria. The Parliament first recalls the importance of respecting the principles of subsidiarity and proportionality and fundamental rights,108 which are said to play a special role in criminal law,109 and it arguably also refers to the principle of coherence, which scholars have suggested should play a role in this field.110 The Parliament

101 2012 Parliament Resolution on an EU approach to criminal law (n 58) recitals E, F, G, and Q and pt 7. 102 See s IV.A and s V.B.3. 103 2012 Parliament Resolution on an EU approach to criminal law (n 58) recitals D, H, I and N and pts 1, 2, 3, 5, 10. 104 ibid recitals H, K and L, in particular pt 4 and to a certain extent pt 11. 105 ibid pt 6. 106 ibid recitals O and P, in particular pt 8 and pts 12–20. 107 ibid pt 7. 108 ibid pts 1 and 2. 109 ibid recital D. 110 The need to ensure coherence when enacting EU criminal law norms is not openly acknowledged as a specific criterion determining when the EU should resort to criminal law. Nonetheless, the special features of the Member States’ legal systems are recalled several times (2012 Parliament Resolution on an EU approach to criminal law (n 58) recitals E and Q) and so is the need for coherent European developments in criminal matters in general (Recital D). Moreover, the potential of the principle as a true criterion to select which areas deserve criminalisation seems to be partially acknowledged when it is said that the key areas of criminal law must be left to Member States (recital E). Such a choice could indeed be inspired by the desire of preserve the inner coherence of Member States’ legal systems.

164  From Tampere to Stockholm then goes into more detail and devotes a specific point of the Resolution to the actual steps to be followed in the course of the legislative procedure when deciding on whether to adopt criminal law at all. The relevant part reads as follows: [The Parliament] emphasises that it is not sufficient to refer to abstract notions or to symbolic effects but that the necessity of new substantive criminal law provisions must be demonstrated by the necessary factual evidence making it clear that: [T]he criminal provisions focus on conduct causing significant pecuniary or nonpecuniary damage to society, individuals or a group of individuals; there are no other, less intrusive measures available for addressing such conduct.

What can be read from the recitals above is that the cooperation rationale, with the implied utilitarian legitimacy for EU criminal law, is not sufficient as a standalone rationale. Not every type of behaviour can be criminalised on the sole ground that this would help create a level playing field that is instrumental to cooperation. Only those forms of conduct that cause significant pecuniary or non-pecuniary damage should be considered. In other words, a complementary values-based rationale is also necessary to justify harmonisation. This focus on the consequences of the criminalised behaviour thus incorporates a deontological approach into the criminal law. Interestingly, the Parliament uses a similar, if not even more pragmatic, wording to that of the Council: the latter speaks of ‘harm to an interest’, while the former speaks of ‘pecuniary or non-pecuniary damage’ to either the society or the individuals. Both expressions echo the liberal approach to criminal law and especially the Anglo-Saxon harm-based approach. Yet, Chapter 1 clarified that the concept of ‘harm to an interest’ can and has been interpreted in various ways and it might include even a moralitybased approach. It could be said that the Parliament’s use of the term ‘damage’ rather than harm arguably refers even more clearly to the undesirable (pecuniary or not) material consequences of the act rather than its clash with moral conduct norms. V.  THE SHIFT FROM A UTILITARIAN TO AN INTEGRATED APPROACH IN POLICY DOCUMENTS

The analysis of the various Justice and Home Affairs programmes and ad hoc EU criminalisation policy documents in this chapter has shown a historical shift from a solely utilitarian to an integrated approach to the legitimacy of EU criminal law (see section V.A). This evolution in policy language mirrors that of EU Treaty norms in the field of EU securitised criminalisation. Conversely, this is the opposite of the approach in the field of EU functional criminalisation, where EU Treaty norms currently only focus on the utilitarian, effectivenessbased legitimacy of EU criminal law (see section V.B).

The Shift from a Utilitarian to an Integrated Approach   165 A.  The Growing Importance of the Deontological Approach to the Legitimacy of Criminal Law The 1999 Tampere Programme and the 2005 Hague Programme endorsed the cooperation rationale as the key rationale legitimating the harmonisation of definitions of crimes. The EU was understood as a platform for Member States to enforce their own criminalisation policies in transnational cases, and harmonisation as a tool to achieve this goal. The legitimacy of EU criminal law was meant to rest on utilitarian grounds. The criminalisation of new behaviours was justified to the extent that it created a level playing field enabling the criminalisation policy choices of more repressive Member States to be effective beyond their borders. The 2009 Stockholm Programme innovatively mentioned the need for a complementary values-based justification for harmonisation of definitions of crimes, namely the need to focus only on those behaviours which affect important interests. The EU was entrusted with a more complex mission in the field of harmonisation. This was the task of carrying out a selection of interests to be considered so high as to be worth protecting in all Member States, and whose safeguarding requires Member States to cooperate among one another. The understanding of the role for the EU shifted from a mere enforcement-enhancing platform to that of an interests-based agenda-setter. This implied a complementary deontological understanding of the legitimacy of EU criminal law: new forms of behaviour were supposed to be criminalised not only for the sake of creating a level playing field, but also because they deserved to be. A similar importance for the deontological legitimacy of EU criminal law can be appreciated in the EU criminalisation policy documents, which the Institutions enacted in conjunction with the Treaty of Lisbon’s entry into force. Considering the documents of the Council, the Commission and the Parliament as an expression of a single voice on the theme of the legitimacy of EU criminal law would not be correct. This chapter has described how these documents came about in a context of inter-institutional tension, where each Institution seemed to be driven more by the desire to impose its own position on the matter rather than by the motivation for inter-institutional coordination on the subject. Nonetheless, there is at least a common trend which underpins the perspective of each Institution. The mere objective of ensuring enforcement either of national criminal policies (when harmonisation is used to support cooperation) or of EU non-Area of Freedom, Security and Justice policies (when harmonisation is used to secure the enforcement of EU norms) is not sufficient to justify resorting to the harmonisation of criminal law. There need to be further deontological rationales for criminal law. These are variously framed by the Institutions. The interests at stake must play a role (the Council), as does the seriousness of the breach (the Commission) and the damage caused by the conduct to the society and the individuals (the Parliament).

166  From Tampere to Stockholm B.  The EU Policy Documents and EU Treaty Norms: Synchronies and Ruptures The shift from a utilitarian, cooperation-based approach to an integrated approach to the legitimacy of EU criminal law in EU policy documents is very much in sync with what one can find in the Treaty norms regulating EU securitised criminal law. Chapter 4 explained how EU criminalisation in this context was initially justified as a tool to support judicial cooperation under the Treaties of Maastricht and Amsterdam, whereas the Treaty of Lisbon requested further deontological justifications. First the Tampere, The Hague and the Stockholm Programmes and then the EU criminalisation policy documents followed a similar path.111 The situation is somewhat reversed in the field of functional criminalisation. Chapter 5 explained that the case law of the Court of Justice initially maintained an integrated approach. It focused on the need for effective sanctions for violations of EU norms, but it implicitly or explicitly acknowledged the role for values-based considerations when the choice in favour of criminal sanctions was at stake. The Court does not insist on the cooperation rationale for the harmonisation of criminal law. However, the Treaty of Lisbon, in a departure from the previous case law, failed to acknowledge any role for deontological considerations in the choice of which breaches of EU law deserve criminal sanctions. The policy discourse surrounding EU functional criminalisation, which conversely stresses both the importance of the cooperation rationale and the values-based justifications in this field, diverges from the Treaty framework and, interestingly, complements it. Admittedly, in its 2005 Communication reacting to the Environmental Crime case, the Commission had initially postulated a purely utilitarian, effectiveness-based legitimacy for EU functional criminalisation.112 Yet, the Parliament’s Resolution on the Environmental Crime case already advocated the need for complementary deontological rationales, as did the subsequent Commission of 2011 addressing specifically EU functional criminalisation, and the following documents of the Parliament, the Council of Ministers and the European Council analysed in this chapter, which discussed the policy framework for the EU criminalisation as a whole, including functional criminalisation. In short, the deontological aspect that went missing in the Treaty wording in Article 83(2) TFEU, or the cooperation aspect that had not featured in the discussions, have been brought back by the policy documents.

111 Interestingly, the 2009 Council Conclusions (n 52) do not mention the cooperation rationale at all. 112 See ch 5, s III.C.

Conclusions  167 VI. CONCLUSIONS

This chapter has provided a second building block to answer the second and third research questions of this book. It has shown how, in their policy declarations, the Institutions have progressively acknowledged the importance of deontological alongside utilitarian considerations as a legitimating factor for EU criminal law. This ensures greater coherence with EU values. Such an integrated position replicates the Treaty’s integrated approach in Article 83(1) TFEU on securitised criminalisation. It also positively complements the utilitarian-only approach of Article 83(2) TFEU on functional criminalisation. In their policy documents, the Institutions recall the deontological dimension to EU criminalisation, which was present in the Court’s case law, but which had been dropped in the Treaty. Chapter 1 raised the question of the desirability of constitutionalising criminalisation criteria. It was suggested that not enshrining them as legally enforceable principles would give them more flexibility and adaptability to criminal policies. It is also more respectful of democratically elected legislators to allow them to set goals for themselves. However, keeping such indications solely as doctrinal indications or including them in non-binding policy documents has the disadvantage of rendering them legally toothless. In light of this, while it is true that the Institutions embrace an integrated agenda, especially in the field of EU functional criminalisation, it is necessary to assess what the actual impact of such non-binding declarations is. Chapter 7 will turn to the analysis of legislation, which constitutes the third building block to answering the central question raised in this book: how is EU criminal law legitimate in practice and how does this relate to the EU constitutional values and principles?

7 Legitimating EU Criminal Law in Practice The Case of Racism and Xenophobia, Market Abuse and PIF Crimes

H

aving investigated what rationales for the use of criminal law the EU is bound to follow (Treaty norms) and intends to follow (policy documents), this final chapter turns to the analysis of what rationales have, in practice, guided EU criminalisation. This is not an easy task. Since the entry into force of the Treaty of Maastricht, the EU has adopted a considerable number of EU legislative instruments dealing with criminal law, including conventions, Framework Decisions and directives. A large number of these contained definitions of criminal offences,1 notably in the field of organised crime, terrorism, drug trafficking, racism and xenophobia, sexual exploitation of children, human trafficking, corruption, fraud, money laundering, counterfeiting the euro, counterfeiting non-cash means of payment, aiding illegal immigration, ship-source pollution and cybercrime.2 Providing a general theory of the rationales which have guided the adoption of all these instruments is a manifestly impractical task as well as a potentially self-defeating one. Indeed, as the next section will explain, the rationales for the adoption of criminal norms

1 For an analysis of the alternation of the two strategies of integration (harmonisation and mutual recognition) and for the balance between the two, see S Manacorda, ‘La consolidation de l’Espace de Liberté Securité et Justice: vers une mise à l’écart du rapprochement pénal?’ (2007) Revue de science criminelle et de droit comparé 890; S Manacorda, ‘La mutation “à droit constant” du Troisième Pilier: renforcement et élargissement de la cooperation’ (2008) Revue de science criminelle et de droit comparé 995; S Manacorda, ‘Un bilan des dynamiques d’intégration pénale à l’aube du traité de Lisbonne’ (2009) Revue de Science Criminelle 927; S Manacorda, ‘L’age de la Maturité: stabilisation et traits conservateurs dans la politique pénale de l’Union Européenne’ (2012) Revue de Science Criminelle 931. 2 For an exhaustive list of the approved instruments, see L Arroyo Zapatero and M Munoz de Morales Romero, ‘Droit pénal européen et Traité de Lisbonne: le cas de l’harmonisation a­ utonome (article 83.1 TFEU)’ in G Giudicielli-Delage and L Christine (eds), Le Droit Pénal de l’Union ­Européenne au Lendemain du Traité de Lisbonne (Paris, Société de Législation Comparée, 2012) 113, 118–20.

The Patchwork Structure of EU Criminal Law  169 appear to be very disparate, and sometimes a sound theoretical reflection on why the adoption of criminal norms is legitimate is even missing. Conversely, this chapter has two more modest objectives. Through three case studies, it first aims to illustrate how the different rationales for the harmonisation of criminal law play out in practice and what this implies for the legitimacy of EU criminal law. Second, it aims to demonstrate that, in fact, the focus on a utilitarian justification for criminal law that one finds in the Treaties and, to a minor extent, in the policy documents is not fully apparent in the legislation; quite the contrary. Despite the text of the Treaties and what the Institutions have announced in the policy documents, EU criminal legislation relying on the deontological rationale, even as a self-standing rationale, was adopted. In other words, where the Treaty and the Institutions’ policy declarations apparently leaned towards a risk of bureaucratic criminal law that is effective but does not embody or promote particularly important values, in practice the Institutions have in several cases leaned towards the opposite undesirable example of equally poorly legitimate criminal law, namely symbolic criminal law. This is criminal law which declares to protect important values but is, in fact, ineffective. In terms of structure, section I provides some general remarks on EU criminal legislation and explains the choice of case studies. The following sections focus respectively on each of the case studies, notably the 2008 Framework Decision on Racism and Xenophobia3 (section II), the 2014 Market Abuse Directive4 (section III) and the 2017 Directive for the Protection of the Financial Interests of the EU (hereinafter the PIF Directive) (section IV).5 I.  THE PATCHWORK STRUCTURE OF EU CRIMINAL LAW AND THE CHOICE OF CASE STUDIES

Law-making is naturally first and foremost a political process.6 While there might be, and sometimes there is, some theoretical underpinning behind a proposed policy, legal provisions are finally included in the texts because political compromise among the involved actors was reached on them. Speaking of

3 Council Framework Decision 2008/913/JHA of 28 November 2008 on combating certain forms and expressions of racism and xenophobia by means of criminal law [2008] OJ L328/55 (hereinafter Framework Decision on Racism and Xenophobia). 4 Directive 2014/57/EU of the European Parliament and of the Council of 16 April 2014 on criminal sanctions for market abuse (market abuse directive) [2014] OJ L173/17 (hereinafter Market Abuse Directive). 5 Directive (EU) 2017/1371 of the European Parliament and of the Council of 5 July 2017 on the fight against fraud to the Union’s financial interests by means of criminal law [2017] OJ L198/29 (hereinafter PIF Directive). The acronym PIF, which is commonly used to indicate protection of the financial interests of the EU, actually comes from French ‘Protection des Interets Financiers’. 6 On the criminal law as a social and political institution, see RA Duff et al, ‘Introduction’ in RA Duff et al (eds), The Structure of the Criminal Law (Oxford, Oxford University Press, 2011) 1, 8.

170  Legitimating EU Criminal Law in Practice EU legislation,7 Weatherill provocatively claims that it is hard to take seriously the notion that community competence is limited by anything other than the need to secure unanimous support in the Council.8 While this is true for every field of law, the phenomenon is particularly acute in criminal law. Enacting new crimes, especially as a response to public events, is a cheap political currency to show proactiveness on social problems in order to demonstrate that ‘something is being done on the matter’. It is thus relatively easy to find a political compromise on criminal law. Before Lisbon, Member States had even an easier job in adopting such ‘events-responsive’ criminal law at the supranational level, considering that only the agreement between governments in the Council was needed. Oppositions from national parliaments could be easily bypassed in such a way. National legislatures were presented with a fait accompli to be implemented at a national level, in what De Hert qualifies as a phenomenon of ‘Politics of Scale’.9 This phenomenon was referred in Chapter 3 as ‘hidden’ or ‘dirty’ rationales for harmonisation.10 Weyembergh illustrates how the Dutroux paedophilia scandal in Belgium influenced the adoption of EU legislation on trafficking in human beings; how the Douvres case involving the death of a large number of illegal immigrants in a French factory influenced the adoption of EU legislation on illegal migration; how the environmental disaster of the Prestige oil spill, which affected France, Spain and Portugal, led to the adoption of EU legislation on environmental protection; and, finally and most notably, how the events of 9/11 gave a strong impulse to negotiations on the EU legislation on terrorism.11 Weyembergh also illustrates how adopting EU criminal law only as a response to public events directly implies that some forms of behaviour are criminalised, while others, which are equally serious but have less resonance in the media, are not.12 Post-Lisbon, the Council is no longer the sole

7 The Commission has the right to take the legislative initiative. It proposes the text on which the negotiations will be based. It can also play the role of mediator in the course of the legislative procedure if the Council and the Parliament do not reach an agreement. See art 294 TFEU. 8 S Weatherill, Law and Integration in the European Union (Oxford, Clarendon Press, 1995) 51. 9 P De Hert, ‘Division of Competences between National and European Levels with Regard to Justice and Home Affairs’ in M Anderson and J Apap (eds), Police and Justice Cooperation in the New European Borders (The Hague, Kluwer Law International, 2002) 65. 10 See ch 3, s III. 11 A Weyembergh, L’harmonisation des législations: condition de l’espace pénal européen et révélateur de ses tensions (Brussels, Editions de l’Université de Bruxelles, 2004) 261–62. 12 If the main driving factor for adopting new legislation is just a need to reassure the public opinion, there is a risk that the legislator only focuses on those forms of behaviour that actually worry the public opinion, and neglects other behaviours which are similarly reprehensible, but which have not had similar resonance in the media. In the example mentioned in the text concerning trafficking in human beings, the legislature focused only on trafficking for the purposes of sexual exploitation, the type of conduct that was at the heart of the national scandal triggering the European initiative. Conversely, it left out other forms of exploitation, such as forced labour. Secondly, criminalisation responsive to public-events risks of favouring the practice of having very large incriminations that might include also legitimate behaviours that in principle do not deserve criminalisation, with the aim of catching the larger number of conducts possible, in order to impact on public opinion.

The Patchwork Structure of EU Criminal Law  171 legislator;13 the Parliament is now also involved.14 Nonetheless, as section III will explain in more detail, some concerns have been raised in the literature as to the event-driven character of the post-Lisbon 2014 Market Abuse Directive, which was interpreted as a hasty response to the financial crisis. Similarly, the rationale for adopting the 2017 Directive on terrorism was clearly event-driven. The instrument was proposed and adopted in reaction to the 2015 terrorist attacks in France, which led to a very hasty negotiating process and policy choices.15 The consequences of this events-driven policy-making are for some that EU criminal law has a ‘patchwork structure’, leading to some sort of ‘legislative chaos’,16 which ‘almost defies the legal urge to systematise and to categorise’.17 For example, Vervaele questioned why, pre-Lisbon, there has been much debate about the protection of environmental policies and of the financial interests of the EC through criminal law directives, while protection of the interests at stake in other ‘hardcore’ policies, such as monetary policy, has been left to criminal sanctions envisaged in Framework Decisions or have not relied on criminal law at all. The latter was the case of competition policies and used to be the case for the interests affected by market abuse and terrorist financing.18 De Hert also See Weyembergh (n 11) 261–62. Still, it should be mentioned that having civil society influencing the legislative process and having public events triggering debate on some important topics can have some welcome consequences. This is the case, for instance, when NGOs manage to draw legislators’ attention to gross violations of human rights and make them act in that regard; see U Sieber, ‘The Forces behind the Harmonization of Criminal Law’ in M Delmas-Marty, U Sieber and M Pieth, Les chemins de l’harmonisation penale law (Paris, Société de Legislation Comparé, 2008) 385, 407. It is therefore necessary to distinguish between criminalisation choices that are ‘event-triggered’ and those that are event-driven. In the former case, the events trigger the debate, but the choice on whether to criminalise or not follows some clear, pre-established criminal policy guidelines. In the latter case, the occurrence of a public scandal or gross rights violations automatically implies the criminalisation of a certain form of conduct. 13 In the Treaty of Amsterdam, the only other Institution involved was the Commission, which had a shared right of initiative with Member States; art 34 TEU, consolidated in Amsterdam [1997] OJ C340/2. 14 See art 83 TFEU, which mentions the legislative procedure as applicable in this field, jointly with art 294 TFEU, which clarifies the involvement of each Institution. 15 The proposal was drafted within two weeks, directly after the attacks, with no Impact Assessment attached, and was approved in a little over a year. On the Directive, see N Vavoula, ‘Prevention, Surveillance, and the Transformation of Citizenship in the “Security Union”: The Case of Foreign Terrorist Fighters’ in U Sieber et al (eds), Alternative, Informal, and Transitional Types of Criminal Justice and the Legitimacy of New Sanction Models in the Global Risk Society (Freiburg, Max Planck Institute for Foreign and International Criminal Law in collaboration with Duncker & Humblot, 2018). 16 W Perron, ‘Perspective of the Harmonization of Criminal Law and Criminal Procedure in the European Union’ in EJ Husabø and A Strandbakken (eds), Harmonisation of Criminal Law in Europe (Antwerp, Intersentia, 2005) 5, 13. See also Delmas Marty, who argues that supranational harmonisation upsets the inner coherence of national legal systems, without actually creating any supranational coherence: M Delmas-Marty, ‘Introductions objectifs et méthodes’ in Delmas-Marty et al (eds) (n 12) 19, 21. 17 M Fletcher, R Lööf and B Gilmore, EU Criminal Law and Justice (Cheltenham, Edward Elgar, 2008) 173. 18 J Vervaele, ‘Harmonised Union Policies and the Harmonisation of Substantive Criminal Law’ in A Weyembergh and F Galli (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Editions de l’ULB, 2013) 43, 58.

172  Legitimating EU Criminal Law in Practice observes that the lack of EU criminal sanctions in the field of data protection is striking and actually indefensible.19 Selecting representative case studies against this background is thus a challenging exercise. The choice was made in favour of the 2008 Framework Decision on Racism and Xenophobia,20 the 2014 Market Abuse Directive,21 and the 2017 PIF Directive22 for three reasons. A first criterion, closely related to the first descriptive aim of this chapter, is choosing legislative instruments that are able to showcase the greatest number of rationales for the harmonisation of criminal law. This justifies the choice in favour of the Framework Decision on Racism and Xenophobia, which provides an interesting example of legislation justified, among other reasons, on a socialising rationale, and of the PIF Directive, which is included as an example of the positive integration of cooperation, values-based criminalisation and regulatory-based rationales. A second criterion, closely related to the second argumentative aim of this chapter, is to include legislative texts justified on rationales which were not incorporated into their original legal basis. The aim is to show how there can be contradictions between the institutional policy-declaring and policy-making levels as to what criteria legitimate resorting to criminal law. This justifies the inclusion of all three instruments which were adopted on a legal basis incorporating a utilitarian legitimacy for EU criminal law, but which, in fact, seem grounded in complementary or even self-standing deontological legitimacy, albeit to different extents. A final criterion is a chronological one. The choice has been made to include both pre-Lisbon and post-Lisbon instruments and one of the latest texts adopted – namely the PIF Directive – to identify historical trends. II.  THE 2008 FRAMEWORK DECISION ON RACISM AND XENOPHOBIA: SYMBOLIC EU CRIMINAL LAW UNDER AMSTERDAM

The EU started to address the problem of racism and xenophobia back in the 1990s. The attention given to this phenomenon stemmed from the resurgence of extreme-right nationalist parties and ethnic tensions resulting from increasing immigration. The Commission made declarations condemning racism and xenophobia and calling for a ‘vigorous and effective’ response to racism and xenophobia, and it designated 1997 as the European Year against Racism.23 19 P De Hert, ‘The EU Data Protection Reform and the (Forgotten) Use of Criminal Sanctions’ (2014) 4(4) International Data Privacy Law 262, 267. On the need for criminal law in the context of privacy, see also P De Hert and G Boulet, ‘The Co-existence of Administrative and Criminal Law Approaches to Data Protection Wrongs’ in P De Hert and David Wright (eds), Enforcing Privacy: Legal and Technological Approaches (Berne, Springer International, 2016), 357. 20 Framework Decision on Racism and Xenophobia (n 3). 21 Market Abuse Directive (n 4). 22 PIF Directive (n 5). 23 ‘Commission Communication of 13 December 1995 on Racism, Xenophobia and ­Anti-Semitism’, 4–6, COM (1995) 653 (hereinafter Commission Communication on Racism and

The 2008 Framework Decision on Racism and Xenophobia  173 The European Council in its meeting at Corfu in 1994 called for ‘a global strategy at Union level aimed at combating acts of racist and xenophobic violence’.24 In addition to these mainly ‘declaratory initiatives’, the EU soon moved on to the adoption of criminal norms in the field. In 1996, it adopted a Joint Action on Racism and Xenophobia in which Member States committed to either include criminalisation of racist and xenophobic behaviour in their legislation or at least to not subject such offences to the double criminality requirement for extradition.25 In the Treaty of Amsterdam, the commitment of the EU to fight racism and xenophobia was also acknowledged. As mentioned earlier, this area of crime did not feature among those in which the EU was granted explicit competence to harmonise, but it was set as one of the goals of the Area of Freedom, Security and Justice. In the Tampere Programme, the heads of state again called on the EU to take strong steps against racism and xenophobia. The European Parliament then adopted a resolution stating that the EU should pass a ‘Framework Decision’, which unlike Joint Actions (whose legal status was debated) would uncontrovertibly be a legally binding instrument.26 Despite the uncertainty on the presence of a legal basis, the Commission presented a proposal to the Council in 2002 recalling all the provisions relating to judicial cooperation and approximation in the Treaty of Amsterdam, namely Articles 29, 31 and 34(2)(b) TEU.27 The envisaged criminalisation range was wider than that of the Joint Action and, in particular, it included the criminalisation of racist speech and racist conduct as well as denial of serious international crimes. Given this wide scope, its relationship with freedom of speech proved controversial and the negotiations lasted quite some time. The text was finally adopted in 2008. The Framework Decision targets two sorts of conduct. First, Article 1(1)(a) and (b) concerns hate speech and requires Member States to criminalise: (a) publicly inciting to violence or hatred directed against a group of persons or a member of such a group defined by reference to race, colour, religion, descent or national or ethnic origin; (b) the commission of an act referred to in point (a) by public dissemination or distribution of tracts, pictures or other material. Xenophobia); Europa, ‘European Year against Racism is Launched in The Hague’, 30 January 1997, http://europa.eu/rapid/pressReleasesAction.do?reference=IP/97/72&format=HTML&aged=1&la nguage=EN&guiLanguage=en; and see also ‘European Year against Racism is Launched in The Hague’, press release, 30 January 1997, http://europa.eu/rapid/pressReleasesAction.do?reference=IP /97/72&format=HTML&aged=1&language=EN&guiLanguage=en. 24 Corfu European Council of 24–25 June 1994, Conclusions of the Presidency, pt III, 1, https:// www.consilium.europa.eu/media/21207/corfu-european-council.pdf. 25 Joint Action 96/443/JHA of 15 July 1996 adopted by the Council on the basis of Article K.3 of the Treaty on European Union, concerning action to combat racism and xenophobia (hereinafter Joint Action on Racism and Xenophobia) point A. 26 European Parliament Resolution on the European Union’s Position at the World Conference against Racism and the Current Situation in the Union, P5_TA(2003)0012 (2001/2014(INI)). 27 ‘Proposal for a Council Framework Decision on combating racism and xenophobia’ COM (2001) 664 final.

174  Legitimating EU Criminal Law in Practice Second, the Framework Decision addresses the question of the so-called ­‘denialism’, that is, publicly condoning, denying or grossly trivialising heinous crimes whose historical truth and widespread and generalised extremely harmful character are uncontroversial. The Framework Decision in particular targets ‘denialism’ of international crimes as envisaged by the Statute of the International Criminal Court and ‘denialism’ of the Holocaust. Article 1(1)(c) and (d) require Member States to criminalise: (c) publicly condoning, denying or grossly trivialising crimes of genocide, crimes against humanity and war crimes as defined in Articles 6, 7 and 8 of the Statute of the International Criminal Court, directed against a group of persons or a member of such a group defined by reference to race, colour, religion, descent or national or ethnic origin when the conduct is carried out in a manner likely to incite to violence or hatred against such a group or a member of such a group; (d) publicly condoning, denying or grossly trivialising the crimes defined in Article 6 of the Charter of the International Military Tribunal appended to the London Agreement of 8 August 1945, directed against a group of persons or a member of such a group defined by reference to race, colour, religion, descent or national or ethnic origin when the conduct is carried out in a manner likely to incite to violence or hatred against such a group or a member of such a group.

The preamble of the Framework Decision justifies the introduction of new crimes on a values-based rationale (harmonising criminal law to protect important interests) and a cooperation rationale (harmonising criminal law to support judicial cooperation).28 Given that the text was adopted under the Amsterdam framework, when both the Treaty and policy language insisted on the cooperation rationale, one would expect the cooperation rationale to be pre-eminent. Conversely, the next paragraphs show that the normative rationale is the dominant one in the text (see section II.A); that the cooperation one is actually less important if not irrelevant (see section II.B); and that actually a different rationale (namely the socialising rationale) played an important role (see section II.C). This Framework Decision makes a very interesting example of – nearly – symbolic EU criminal law, thus relying heavily on a deontological legitimacy being adopted on mainly utilitarian institutional premises (see section II.D). A.  The Values-Based Criminalisation Rationale of the Framework Decision: EU Values of Liberty, Democracy, and Individuals’ Safety The preamble of the Framework Decision begins by highlighting the values that racism and xenophobia broadly affect, namely the principles of liberty, democracy, respect for human rights and fundamental freedoms, and the rule of law,



28 Recital

12 of the Preamble of the Framework Decision on Racism and Xenophobia (n 3).

The 2008 Framework Decision on Racism and Xenophobia  175 principles upon which the EU is founded and which are common to the Member States.29 It also points out that, more specifically, these crimes represent a threat to the groups targeted by these forms of behaviour.30 In this, it replicates the language of the Commission, which in 1995 already acknowledged that: [T]he need to build the foundations of a wider and deeper community between peoples who had too often opposed each other in violent conflict was central to the ideals that inspire the founders of the Community. The defence of human rights and fundamental freedoms, core values of the European integration project, cannot be separated from the rejection of racism. Indeed, the struggle against racism is a constituent element of the European identity.31

Briefly, the values of liberty and democracy, which are mentioned in the Framework Decision, but also arguably of tolerance, as well as the security of the target group, which are affected by hate speech and denialism, are highly ranked in general. But they are also specifically important because they represent the axiomatic core of the EU project, which is multi-national and multi-cultural in nature. As a consequence of their importance, they need to be protected throughout Europe via the criminal law. The harmonisation of definitions of crimes in this field is thus justified on the basis of these values-based grounds. There are then a number of specific values or legal interests that each criminalisation norm protects, which can be inferred from the text, even though the Framework Decision does not spell them out explicitly. The preamble clearly identifies citizens’ safety as one of the interests protected by the text. Indeed, the safety of a number of people belonging to a target group could be endangered by the conduct of hate speech criminalised by the Framework Decision. The interests protected by the second form of conduct, denialism (Article 1(1)(c) and (d)), are less straightforward. In addressing denialism, the EU entered a very delicate field, where Member States show different levels of sensitivity due to historical and cultural reasons. Condemning Holocaust denial enjoys wide social and political support across Europe. Conversely, qualifying the negation of other mass crimes, such as those committed by the Communist totalitarian regimes as forms of denialism, has strong support, but only in a selected number of countries.32 Furthermore, the criminalisation of denialism raises particular tensions with freedom of expression, especially to the extent that it risks impeding further historical research.33 From a legitimacy of criminal law perspective, the question of what interests are affected by the behaviour of ‘publicly denying historically established truth’ is the subject of debate in doctrinal circles. What is at stake is simply the psychological wellbeing of the victims of mass crimes, 29 ibid Recital 1. 30 ibid Recital 5 of the preamble. 31 Commission Communication on Racism and Xenophobia (n 23) 4. 32 P Lobba, ‘Punishing Denialism beyond Holocaust Denial: EU Framework Decision 2008/913/ JHA and Other Expansive Trends’ (2014) 5 New Journal of European Criminal Law 58, 63. 33 ibid 73 ff.

176  Legitimating EU Criminal Law in Practice which might be affected by denialist conduct. Nevertheless, this interpretation is not unanimously endorsed, and denialism is sometimes given as an example of a crime lacking a legal interest. Indeed when developing the theory of legal interests, Klaus Roxin offers the example of ‘Holocaust denial’ as behaviour whose criminalisation would be particularly hard to justify on the basis of his theory. When discussing Roxin’s work, Lauterwein observes: ‘Despite the fact that, since the opposite is historically proven, anyone who approves, denies or belittles the crimes of the Nazis is either a fool or a liar, the simple stating of such an untruth does not affect the peaceful coexistence of the people.’34 Moreover, the UN Human Rights Committee also considered that the protection of a legally established historical truth cannot per se amount to an interest protected by the crime of denialism. It maintained that: ‘Laws that penalise the expression of opinions about historical facts are incompatible with … freedom of opinion and expression. The Covenant does not permit general prohibition of expressions of an erroneous opinion or an incorrect interpretation of past events.’35 However, the question of the legal interest specifically protected by the ­criminalisation of denialism does not arise as such for the EU Framework Decision, since the text embraces a restrictive approach to the definition of this crime. It adds two specific conditions to the criminalisation provisions, which limit the obligation to criminalise only those forms of denialist conduct that affect further interests than the simple psychological wellbeing of the victims. First, Article 1(3) imposes that denial of the Holocaust and of international crimes is to be criminalised only when the conduct is ‘carried out in a manner likely to incite to violence or hatred against such a group or a member of such a group’. Denialism is not criminalised because it contravenes shared values or allegedly because it threatens the psychological wellbeing of the victims of mass atrocities, but only if it also constitutes a form of hate speech that is likely to threaten the actual safety of the targeted group. In deciding not to conceptualise denialism as an autonomous crime, but as a sub-species of hate speech crime, the Framework Decision makes a similar choice to the one of the previous Joint Action36 and of the protocol to the Convention on Cybercrime of the Council of Europe.37 Second, Article 1(2) of the Framework Decision grants Member States the possibility of limiting their criminalisation of denialism when this is ‘carried out in a manner likely to disturb public order’. This clause draws from German legislation, though German judges have adopted an interpretation which de

34 C Lauterwin, The Limits of Criminal Law: A Comparative Analysis of Approaches to Legal Theorising (Farnham, Ashgate, 2010) 15. 35 General Comment 34 on States parties’ obligations under Article 19 of the International ­Covenant on Civil and Political Rights (ICCPR), para 34, https://www2.ohchr.org/english/bodies/ hrc/docs/GC34.pdf. 36 Joint Action on Racism and Xenophobia (n 25) art A(b), (c). 37 Article 6(1) and (2) of Additional Protocol to the Convention on Cybercrime, 28 January 2003, ETS No 189.

The 2008 Framework Decision on Racism and Xenophobia  177 facto considers any form of denial or minimisation of the Holocaust as a threat to public peace, a concept that is similar to public order.38 Still, the inclusion of this provision in the EU text might allow other jurisdictions to implement this requirement and give it more substantial weight. Briefly, the Framework Decision makes it possible to restrict the punishment of denialism only when the relevant conduct can harm the objective interests of the safety of a vulnerable group, as well as ‘public order’ more generally. This approach makes it possible to ground criminalisation in some other legal interests beyond the psychological wellbeing of victims and thus to avoid entering the debate on whether the latter is a sufficient justification. B.  The Cooperation Rationale of the Framework Decision: Was Transnational Racist and Hate Speech Really the Problem? The second alleged rationale for the adoption of the Framework Decision and imposing an EU-wide criminalisation of racism and xenophobia, is the cooperation rationale. References to the need for harmonisation in this area in order to ensure effective cooperation can be read in recitals 3, 4 and 12 and Article 10 of the Framework Decision. Nevertheless, Turner, who has provided the most extensive discussion of this Framework Decision, has highlighted that at least three aspects in the legislative history and the text of the Framework Decision show that this rationale, albeit often recalled, is not the conclusive one leading to the adoption of the text.39 First, this author observes that despite the insistence on the need for approximation in order to ease cooperation, neither the Commission proposal nor the Parliament resolution on the proposal mentions any evidence of the cross-border nature of racist and xenophobic crimes. Admittedly, the Commission recalls the spread of racist propaganda on the internet. This arguably points in the direction of a transnational dimension of the phenomenon, given the borderless nature of cyberspace. However, the Commission does not extensively elaborate on the transnational spread of these forms of conduct across the EU via the Web and actually only refers to data about racist websites outside the EU.40 Moreover, the Commission also points to the difference in legislative approaches towards the theme of hate speech, and genocide denial, and how this could ‘constitute barriers to international judicial cooperation’.41 Yet, by failing to extensively discuss the cross-border extent of these crimes, they also fail to demonstrate the

38 Lobba (n 32) 66. 39 JI Turner, ‘The Expressive Dimension of EU Criminal Law’ (2012) 60 American Journal of Comparative Law 555. 40 European Commission, ‘Proposal for a Council Framework Decision on Combating Racism and Xenophobia’ COM (2001) 664 final, at 5. See Turner (n 39) 568. 41 Turner (n 39) 568.

178  Legitimating EU Criminal Law in Practice need for such interstate cooperation and the resulting need for approximation. To this first Turner’s argument, one might add that the Framework Decision itself requires Member States to extend the scope of the relevant jurisdiction to online cases, implying that the primary focus is offline cases.42 The cross-border dimension of offline cases is not addressed in any point. Moreover, interestingly, the current version of the Treaty, notably Article 83(1) TFEU, which arguably lists the areas of serious and cross-border crime, does not include racism and xenophobia.43 This would retroactively confirm that the EU focus on the fight against these crimes never depended on their cross-border character. To be clear, neither Turner’s argument nor the one made here are that racist and xenophobic speech is not in fact spread transnationally across the EU, especially through the Web. In recent times, in the context of the refugee crisis, this phenomenon might have even assumed greater proportions than was the case when the Framework Decision was negotiated. The argument is simply that in the preparatory documents that have accompanied the Framework Decision, there does not seem to be much insistence on this point. Second, Turner recalls that the previously existing Joint Action on racism and xenophobia introduced the option of either criminalising racist and xenophobic crimes or eliminating the double criminality requirement in extradition law with respect to these offences.44 It was explained in Chapter 2 that one of the reasons that justified the need for harmonisation in order to support ­cooperation was that it could help meet the double criminality requirement.45 Yet, by the time the Framework Decision on Racism and Xenophobia was adopted, the EU had already abolished the double criminality requirement in cases of racist and xenophobic crimes under the European Arrest Warrant.46 The elimination of double criminality thus weakens the argument for the need for harmonisation to ease cooperation.47 Yet, it is important to bear in mind the other two arguments normally relied upon in the context of the cooperation rationale for harmonisation, namely that harmonisation can create mutual trust and that it can help clarify that the scope of action of the Justice and Home Affairs actors is still valid. Third, Turner finally observes that the way in which the offences were designed is a rather watertight compromise – think of the optional character of the obligation to criminalise denialism – that is unlikely to have very little

42 Article 9(2) and 9(3) of the Framework Decision on Racism and Xenophobia (n 3). 43 Though interestingly, the Stockholm Programme calls for some investigation on whether the Framework Decision should be repealed: European Council, The Stockholm Programme: an open and secure Europe serving and protecting citizens [2010] OJ C115/1, pt 2.3.1. 44 Turner (n 39) 573. 45 See extensively the rationales for harmonisation discussed in ch 2. 46 Council Framework Decision 2002/584/JHA of 13 June 2002 on the European Arrest Warrant and the surrender procedures between Member States – statements made by certain Member States on the adoption of the Framework Decision [2002] OJ L190/1. 47 Turner (n 39) 573.

The 2008 Framework Decision on Racism and Xenophobia  179 harmonisation impact.48 If the Framework Decision is not actually able to bring national legislation together, then the final goal of creating a level playing field to ease cooperation is doomed to fail. As a consequence, Turner concludes that the reason that led to the adoption of the instrument was not connected to the needs of transnational criminal law enforcement. In particular, denialism is indeed criminalised as a form of hate speech. As a consequence, Member States that already had a provision on denialism as an autonomous crime obviously did not have to amend their legislation. Those who did not have such legislation could arguably maintain their more tolerant approach. The requirement of criminalising racist denialism could be implemented by relying on the law simply criminalising hate speech, under which racist denialism could be subsumed. The actual harmonisation potential of the provision on denialism is thus hard to see. Before the adoption of the Framework Decision, there were only three Member States that did not include criminal law legislation on hate speech as an independent crime, namely Cyprus, Slovenia and Estonia, and that in Cyprus at least was introduced as a result of EU legislation.49 However, it seems unlikely that these differences between existing legislation actually represented a concrete barrier to cooperation. It appears improbable that xenophobic groups would routinely base themselves, or their servers, in one of these countries to spread hate speech throughout Europe and that cooperation would be impaired due to the unwillingness of these countries to disrupt the relevant websites. In any case, the Framework Decision does not explicitly state that there is a specific problem with these three countries which requires harmonisation in order to be addressed. In light of Turner’s three arguments presented above, even bearing in mind the caveats identified in each case, it seems that the cooperation rationale, despite being insisted on in the preamble, was not the single and pre-eminent argument for the adoption of the text. C.  A Complementary Social Engineering Rationale for the Framework Decision? The previous two sub-sections have analysed the weight granted to the two rationales referred to in the preamble of the Framework Decision: the values-based criminalisation and the cooperation rationales. Alongside these discussions, it is also interesting to reflect on the role for a third rationale, namely the socialising rationale. Chapter 2 explained that one possible objective for harmonisation can be that of creating a new community through a new harmonised law. Such new common rules would and should indicate the shared common interests 48 ibid 571–72. 49 Implementation Reports are available at: https://eur-lex.europa.eu/legal-content/EN/NIM/?uri= CELEX:32008F0913.

180  Legitimating EU Criminal Law in Practice and values that people can jointly relate to and which would reflect why the new community was created. It is argued that this was one of the implicit rationales for the Framework Decision on Racism and Xenophobia. While this rationale is not acknowledged within the text as such, it can be deduced from the declarations of the Commission and the Parliament that have accompanied the negotiation of the instrument. It has already been mentioned that the Commission connected the fight against racism to the EU identity.50 By criminalising conduct that affects values regarded as close to the EU identity, the EU basically re-affirms what its identity is.51 The Parliament similarly insisted on the symbolic potential of the Framework Decision to show what Europe stands for. In particular, it urged the Council to finalise the text of the Framework Decision so as to ‘send out a strong political message in support of a citizens’ Europe’.52 These declarations can arguably be connected to the socialising rationale identified above. By declaring what values the EU stands for, the EU better defines the nature of the project and thus can also achieve the objective of increasing the attachment of EU citizens to it, thus making the ‘EU community’ more cohesive. The socialising rationale is naturally intended to be complementary to the values-based criminalisation rationale. The Framework Decision does not merely pursue expressive purposes without any actual harmonisation and thus without any ambition to harmonise national criminal laws. It was recalled above that the Framework Decision had at least some (even if reduced) harmonisation impact. Rather, the text seems more like an international human rights instrument setting a minimum common denominator to which all Member States have to abide. Such a minimum criminalisation denominator is chosen first in light of the absolute importance of the values at stake – all Member States should protect the value of liberty, democracy, tolerance and the safety of vulnerable groups because these are important values, second in light of the particular importance of such values for the EU project and third in light of its identitybuilding potential. D.  The Framework Decision as Symbolic EU Criminal Law in a Utilitarian Institutional and Policy Context? To conclude, the analysis shows that the Framework Decision is an interesting example of symbolic EU criminal law adopted within a legal framework that in fact relied on a strong utilitarian legitimacy for criminal law. As Chapter 4 largely illustrated, the Amsterdam framework was underpinned by a 50 Commission Communication on Racism and Xenophobia (n 23) 4. 51 On EU identity, see also Introduction, s III. 52 European Parliament Recommendation to the Council of 21 June 2007 Concerning the Progress of the Negotiations on the Framework Decision on Action to Combat Racism and Xenophobia, P6_TA(2007)0285, A6-0151/2007.

The 2014 Market Abuse Directive  181 cooperation rationale which mainly envisaged harmonisation as a tool to secure judicial cooperation. This was in particular due to the position of the legal basis in the Treaty within the title, and the article specifically devoted to judicial cooperation. Chapter 6 explained that the same approach could be appreciated in the 1999 Vienna Action Plan for the implementation of the Amsterdam Treaty and in the following 1999 Tampere and 2005 Hague Programmes. In terms of criminalisation, this approach was qualified as a utilitarian understanding of EU criminal law, since it considered it legitimate to use EU criminal law to ensure the enforcement of national criminalisation agendas. It was also highlighted how relying solely on a cooperation rationale implied a risk of having a spread of EU criminal law without a thorough preliminary normative discussion on whether the interests at stake were so important as to require protection in each Member States; or whether diversity on the point could be tolerated and respected; or a discussion on whether criminal law was the most effective tool in this policy area. In light of this, it is interesting to see how the Framework Decision’s strong values-based rationale and weak cooperation rationale sit at odds with the relevant institutional and policy framework. This is even more interesting if one considers that the Treaty was not clear on whether the EU had the competence to enact the relevant Framework Decision, yet the Institutions acted nonetheless. The Framework Decision was thus not only approved in ‘disharmony’ with the – at the time – institutional understanding of the legitimacy of EU criminal law, but also arguably stretched the boundaries of EU competence in the field. III.  THE 2014 MARKET ABUSE DIRECTIVE: SYMBOLIC EU CRIMINAL LAW UNDER LISBON?

EU legislation in the field of ‘market abuse’, which is traditionally used as an umbrella label to indicate insider dealing and market manipulation,53 has existed since the early 1990s. Four directives were adopted since then, the last one being the 2014 Directive on criminal sanctions for market abuse54 discussed in this section, which was the first one to include criminal penalties. Analysing the justification for the introduction of criminal law in this policy area is particularly interesting given that the question has been at the heart of inter-institutional tensions between the Commission and the Council until very recently. Interestingly, a similar debate surrounded the legislation in the field of money laundering. Before the entry into force of the Treaty of Lisbon, the Commission

53 E Herlin-Karnell, ‘White-Collar Crime and European Financial Crises: Getting Tough on EU Market Abuse’ (2012) 37(4) European Law Review 481, 483. 54 Market Abuse Directive (n 4).

182  Legitimating EU Criminal Law in Practice had included criminal sanctions in each of the three legislative proposals it had respectively submitted to counteract money laundering.55 However, the Council had routinely stripped each text of its penal character56 and stuck to administrative sanctions in the final texts, often referring to the threshold of effective, dissuasive and proportionate sanctions.57 These texts were approved on an internal market legal basis, namely (at the time) Article 95 TEC, which granted the competence to approximate national legislation for the purposes of establishing and building the market.58 The Anti-Money Laundering Directives were then justified as market-making instruments. The Treaty of Lisbon then provided an ad hoc criminal law legal basis for harmonising legislation on money laundering in Article 83(1) TFEU.59 Yet, interestingly enough, in its latest proposal for a fourth money laundering directive, the Commission did not opt for such a criminal law legal basis. It still relied on the internal market basis and proposed a text which only envisaged administrative sanctions.60 These features were

55 See art 2 of the proposal to the first Anti-Money Laundering Directive: European Commission, ‘Proposal for a Council Directive, on prevention of use of the financial system for the purpose of money laundering’ COM (1990) 106 final. The proposal for the second Anti-Money Laundering Directive leaves the definition of crimes unmodified; see European Commission, ‘Proposal for a European Parliament and Council Directive amending Council Directive 91/308/EEC of 10 June 1991 on prevention of the use of the financial system for the purpose of money laundering’ COM (1999) 352 final. Article 1 of the proposal to the third Anti-Money Laundering Directive also includes criminal law: European Commission, ‘Directive of the European Parliament and of the Council on the prevention of the use of the financial system for the purpose of money laundering, including terrorist financing’ COM (2004) 448 final. 56 Vervaele nonetheless observes that the opposition to the use of criminal law in the Council was not unanimous and that there was a certain level of support for resorting to criminal law to enforce EU policies, but which did not lead to the approval of criminal law directives: Vervaele (n 18) 43, 46. 57 See art 2 of Council Directive 91/308/EEC of 10 June 1991 on prevention of the use of the financial system for the purpose of money laundering [1991] OJ L166/77, which simply grants an obligation to prohibit the relevant forms of conduct. Directive 2001/97/EC of the European Parliament and of the Council of 4 December 2001 amending Council Directive 91/308/EEC on prevention of the use of the financial system for the purpose of money laundering [2001] OJ L344/76 does not modify the sanction regime. See finally art 39 of 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 L309/15. 58 The text of the provision reads as follows: ‘The Council shall, acting in accordance with the procedure referred to in Article 251 and after consulting the Economic and Social Committee, adopt the measures for the approximation of the provisions laid down by law, regulation or administrative action in Member States which have as their object the establishment and functioning of the internal market.’ 59 As has been mentioned several times, art 83(1) TFEU states that: ‘The European Parliament and the Council may, by means of directives adopted in accordance with the ordinary legislative procedure, establish minimum rules concerning the definition of criminal offences and sanctions in the areas of particularly serious crime with a cross-border dimension resulting from the nature or impact of such offences or from a special need to combat them on a common basis. These areas of crime are the following: terrorism, trafficking in human beings and sexual exploitation of women and children, illicit drug trafficking, illicit arms trafficking, money laundering, corruption, counterfeiting of means of payment, computer crime and organised crime.’ 60 See art 55 of ‘Proposal for a Directive of the European Parliament and of the Council on the prevention of the use of the financial system for the purpose of money laundering and terrorist financing’ COM (2013)45 final.

The 2014 Market Abuse Directive  183 maintained in the final version that was adopted in 2015.61 Only very recently, with the fifth Anti-Money Laundering Directive, were criminalisation obligations introduced.62 Legislation in the field of market abuse followed a similar pattern, but with a slightly different outcome, which conversely included criminal law directly after the entry into force of the Lisbon Treaty. The argument developed in this section is that the use of criminal law has been purely justified on deontological grounds in order to highlight the importance of the interests at stake. Conversely, the utilitarian justifications are quite weak in this context. It does not stem from the text that criminal law is considered because it would be an effective instrument to counteract market abuse. The following discussion retraces the historical origins of the 2014 Market Abuse Directive (see section III.A), describing the offences sanctioned in the text (see section III.B), explaining where in the text we can find evidence of a regulatory rationale supporting the harmonisation of criminal law in this policy area (see section III.C), but then highlighting how the actual empirical evidence provided in favour of this argument is fairly weak (see section III.D). Lastly, the importance of the values-based criminalisation rationale in this text is underlined (see section III.E) and the discussion concludes by highlighting the paradox inherent in having mainly deontological, symbolic EU criminal law instruments adopted on a legal basis such as Article 83(2) TFEU, which conversely only envisages a utilitarian legitimacy for EU criminal law (see section III.F). A.  The EU Fight against Market Abuse: The Road to Criminal Law In 2001, the Commission proposed a directive that imposed the obligation to criminalise market abuse forms of conduct on Member States.63 Yet, in the final version of the text approved by the Council and the Parliament, the relevant sanctions were turned into administrative ones.64 Like the Anti-Money Laundering Directives, the 2003 Market Abuse Directive was adopted on an internal market legal basis as it was considered necessary to boost confidence in the market.

61 See art 58 of Directive (EU) 2015/849 of the European Parliament and of the Council of 20 May 2015 on the prevention of the use of the financial system for the purposes of money l­aundering or terrorist financing, amending Regulation (EU) No 648/2012 of the European ­Parliament and of the Council, and repealing Directive 2005/60/EC of the European Parliament and of the Council and Commission Directive 2006/70/EC [2015] OJ L141/73. 62 Article 3 of Directive (EU) 2018/1673 of the European Parliament and of the Council of 23 October 2018 on combating money laundering by criminal law [2018] OJ L284/22. 63 European Commission, ‘Proposal for a Directive of the European Parliament and of the Council on insider dealing and market manipulation (market abuse)’ COM (2001) 281 final, art 14. 64 Article 14 of Directive 2003/6/EC of the European Parliament and of the Council of 28 January 2003 on insider dealing and market manipulation (market abuse) [2003] OJ L96/16.

184  Legitimating EU Criminal Law in Practice Such a choice nonetheless raised some concerns. It was debated whether this objective could fall within ‘market establishing and market building’.65 In 2009, the Treaty of Lisbon entered into force and included an ad hoc legal basis for adopting criminal law directives. However, market abuse did not feature among the relevant areas of crime mentioned in Article 83(1) TFEU,66 which exhaustively list the areas of serious and cross-border crime on which the EU has the competence to legislate. Thus, in order to repeal the 2003 Directive, the Commission relied on the alternative criminal law legal basis, namely Article 83(2) TFEU, which grants the EU the competence to harmonise criminal law legislation to enforce EU policies when there has already been some harmonisation and when this is essential for their effective enforcement. The Commission proposed a text in 2011 including criminal sanctions, which was adopted in 2014 by the Council and the Commission, and criminal sanctions were maintained in the text.67 Alongside the Directive, a regulation was also approved as part of the same regulatory package on insider dealing and market manipulation.68 This second text was adopted on an internal market legal basis, namely Article 114 TFEU, the Lisbon successor to Article 95 TEC. The Regulation envisages administrative sanctions for market abuse related misconduct.69 Briefly, to date, the 2014 Directive is the first instrument to deal with market abuse that envisages criminal law sanctions. This makes it a particularly interesting case study for examining what the justifications are for this choice. B.  The Offences in the Market Abuse Directive: Insider Dealing, Unlawful Disclosure of Information and Market Manipulation Given that the 2014 Directive and the 2014 Regulation were adopted as a part of the same regulatory package on ‘Insider Dealing and Market Manipulation’, they are meant to be read jointly. Both texts identify insider dealing, unlawful disclosure of information and market manipulation as forms of misconduct that Member States have to repress. The Regulation describes these types of

65 Herlin-Karnell (n 53) 484 and the authors referred to therein. 66 Nevertheless, it was observed that the misconduct of insider dealing and market manipulation could be linked to organised crime, fraud or computer-related criminality: ibid 483. 67 Market Abuse Directive (n 4). 68 European Commission, ‘Impact Assessment accompanying the document Proposal for a ­Regulation of the European Parliament and of the Council on insider dealing and market manipulation (market abuse) and the Proposal for a Directive of the European Parliament and of the Council on criminal sanctions for insider dealing and market manipulation’ COM (2011)651 final. 69 Regulation 596/2014/EU of the European Parliament and of the Council of 16 April 2014 on market abuse (market abuse regulation) and repealing Directive 2003/6/EC of the European Parliament and of the Council and Commission Directives 2003/124/EC, 2003/125/EC and 2004/72/EC [2014] OJ L173/1 (hereinafter Market Abuse Regulation).

The 2014 Market Abuse Directive  185 conduct in detail,70 and, as said, it imposes the obligation on Member States of attaching administrative sanctions for violations of the relevant prescriptions.71 The Directive then establishes in Articles 3, 4 and 5 that Member States should ensure that insider dealing, unlawful disclosure and market manipulation are considered criminal offences, at least in the most serious cases and when committed intentionally. Interestingly enough, the Directive reproduces within the text of the provisions the definitions of the various forms of misconduct and does not simply refer to the provisions in the Regulation, but to some specific clauses.72 However, the content of such definitions is almost identical to that which can read in the Regulation.73 Scholars have raised concerns as to how to guarantee the respect of ne bis in idem in the context of this double ­administrative/criminal law track and on the effectiveness of such a double ­sanctions regime.74 The relevant doctrinal question is how the choice in favour of harmonisation of criminal law is justified. When looking at the official justifications for the harmonisation of criminal law within the Directive’s preamble, one can appreciate a strong emphasis on the regulatory rationale, that is, using the harmonisation of criminal law to ensure prescriptions of law are complied with. This is discussed in the following sub-section. Nevertheless, if we look at how the criminalisation provisions are shaped, we can infer the presence of a complementary, values-based dimension. This is discussed below. 70 For insider dealing, see arts 8 and 14 of the Market Abuse Regulation (n 69); for unlawful disclosure of information, see arts 10 and 14; and for market manipulation, see arts 12 and 15. For an analysis of the material scope of the regulation and a comparison with the pre-existing framework in the 2003 Directive, see M Siems and M Nelemans, ‘The Reform of the EU Market Abuse Law: Revolution of Evolution?’ (2012) 19 Maastricht Journal of European and Comparative Law 195. 71 Article 30 of the Market Abuse Regulation (n 69). 72 Article 3(8) of the Market Abuse Directive (n 4). 73 Compare art 3 of the Directive and art 8 of the Regulation addressing the conduct of insider dealing. The two provisions have an almost identical wording, except for the fact that art 8(4) of the Regulation also extends the scope of the prohibition to persons who possess ‘inside information under circumstances other than those referred to in the first subparagraph where that person knows or ought to know that it is inside information’. Conversely, as per art 3(3) of the Directive, the scope of the criminal law prohibition this time extends to any person ‘who has obtained inside information under circumstances other than those referred to in the first subparagraph where that person knows that it is inside information’. The reference to ‘ought to know’ has disappeared, probably in line with the Directive’s aim to criminalise only intentional forms of behaviour. A similar comparison can be made between art 4 of the Directive and art 10 of the Regulation defining ‘Unlawful disclosure of inside information’. They present the same wording, except for the condition of a person recommending to disclose inside information when this person knows (the wording of the Directive) or ought to know (the wording of the Regulation) that it is inside information. Finally, a similar comparison can be drawn between art 5 of the Directive and art 12(1) of the Regulation addressing the conduct of market manipulation. Both provisions present identical wording; however, art 12(2) of the Regulation provides a list of examples of specific forms of behaviour that must be included in the definition of ‘market manipulation’, which are not present in the Directive’s definition. 74 On the fundamental rights issue raised by the double-track regime and on the overall effectiveness of the system, see Herlin-Karnell (n 53) 491; M Luchtman and J Vervaele, ‘Enforcing the Market Abuse Regime: Towards an Integrated Model of Criminal Law and Administrative Law Enforcement in the European Union’ (2014) 5(2) New Journal of European Criminal Law 193, 212–17.

186  Legitimating EU Criminal Law in Practice C.  The Regulatory Rationale in the Market Abuse Directive: Three Arguments in Favour of the Effectiveness of Criminal Law As mentioned above, Article 83(2) TFEU, the chosen legal basis for the Market Abuse Directive, embodies a regulatory rationale for the harmonisation of criminal law. It grants the competence to the EU to set harmonised standards in terms of criminal sanctions, when this can ensure the effective enforcement of EU policies.75 Thus, unsurprisingly, the core of the justification for resorting to criminal law in the Market Abuse Directive focuses on the potential of criminal law to secure the enforcement of the EU policy aimed at counteracting market abuse. In particular, in the preamble of the Directive, there are three different effectiveness-based arguments in support of the choice of harmonised criminal sanctions. The first argument found in the Commission’s proposal explanatory memorandum and in the sixth recital of the Directive’s preamble is that criminal law has a higher deterrence potential because it is able to convey a stronger form of social disapproval. It is stated that: ‘Minimum rules on criminal offences and on criminal sanctions for market abuse, which would be transposed into national criminal law and applied by the criminal justice systems of the Member States, can contribute to ensuring the effectiveness of this Union policy by demonstrating social disapproval of a qualitatively different nature compared to administrative sanctions or compensation mechanisms under civil law’ (emphasis added).76 A similar connection between societal disapproval and deterrence could also be appreciated in the amendment proposed by the Parliament on the publication of final criminal convictions.77 The proposed Article 6(1) in the Parliament’s first reading read: ‘To ensure that sanctions have a dissuasive effect on the public at large, they shall, where appropriate, be published, without undue delay, including at least information on the type and nature of the crime and the identity of persons responsible for it.’ However, this amendment did not make it into the final text. The preamble of the Directive only indicates the publication of a final decision on a sanction as a sanction for legal persons (recital 18). The second argument in favour of criminal law as an effective tool relates to the specific deterrent potential of imprisonment sanctions. Article 7 of the Directive envisages minimum maximum imprisonment sanctions for the forms of market abuse misconduct,78 and recital 16 of the preamble justifies this choice

75 See ch 4, s V. 76 European Commission, ‘Proposal for a Directive of the European Parliament and of the Council on criminal sanctions for insider dealing and market manipulation’ COM (2011) 654 final (hereinafter Commission Proposal for a Market Abuse Directive) 3. 77 See ibid art 6(1). 78 Interestingly enough, the Commission enumerated the inefficiency of the sanction regime that pre-dated the 2014 Directive as being among the reasons for updating the legislative framework. Yet in its original proposal, it only included an obligation to resort to effective, proportionate and

The 2014 Market Abuse Directive  187 in light of their deterrent and effective potential: ‘In order for the sanctions for the offences referred to in this Directive to be effective and dissuasive, a minimum level for the maximum term of imprisonment should be set in this Directive.’79 It appears that the only way to ensure effective enforcement is to include imprisonment sanctions and thus resort to criminal law. The third argument in favour of criminal law as an effective tool to enforce EU policy on market abuse relies on the stronger powers that authorities have in the context of criminal investigations. Recital 7 of the preamble of the Directive states that: ‘Common minimum [criminal] rules would also make it possible to use more effective methods of investigation and enable more effective cooperation within and between Member States.’ In this case, the argument does not relate to the potential of criminal law as an enforcement instrument as such; rather, it relates to the consequences that stem from criminalisation in terms of applicable procedures. It thus looks at criminalisation in a pragmatic way.80 Alongside these three main arguments supporting the regulatory rationale, it is worth recalling that in its explanatory memorandum, the Commission also mentions the traditional cooperation rationale for the harmonisation of definitions of crimes: ‘Common minimum rules on the definition of criminal offences for the most serious market abuse offences facilitate the cooperation of law enforcement authorities in the Union, especially considering that the offences are in many cases committed across borders.’81 The Directive’s preamble also refers to the cooperation rationale, albeit from a different angle. It does not explicitly refer to the traditional double criminality, mutual trust and Justice and Home Affairs actors’ arguments, which were discussed in Chapter 3; rather, it states that cooperation should be improved and that the difference between the powers of the various judicial authorities should not hinder such cooperation

dissuasive sanctions, without any specification on the type of sanctions. The inclusion of an obligation to impose imprisonment sanctions was added by the Parliament. On this, see Luchtman and Vervaele (n 74) 211. 79 Remarkably, the preamble does not mention why enforcement would be a proportionate sanction. This is arguably an instrumental approach to the choice of sanctions. It focuses on the objective they can achieve, deterrence, while it does not consider whether their severity reflects the disvalue of the sanctioned conduct. Such a focus on the deterrence potential on sanctions, and the neglect of the proportion between the sanction and the disvalue of the prohibited conduct, sits at odds with the principle of proportionality of sanctions, as enshrined in art 49(3) of the Charter, which the Directive declares itself to respect (see Recital 27 of the Market Abuse Directive (n 4)). 80 It is interesting to compare this pragmatic approach to the similar one adopted by the Commission in its 2011 Communication: European Commission, ‘Towards an EU Criminal Policy: Ensuring the effective implementation of EU policies through criminal law’ COM (2011) 573 final (hereinafter 2011 Commission Communication on EU Criminal Policy), which is discussed in ch 6, s IV.B. In that context, the Commission had also enumerated the applicable procedure stemming from the decision to consider a certain type of behaviour a criminal offence rather than an administrative offence. The approach was equally pragmatic; however, in the Communication, the Commission focused on the higher procedural safeguards that individuals would benefit from in the context of criminal proceedings, rather than on the powers of the investigative authorities. 81 Commission Proposal for a Market Abuse Directive (n 76) 3.

188  Legitimating EU Criminal Law in Practice (recital 26). In other words, it does not focus on the fact that Member States define the crimes differently and that such differences in legislation hinder cooperation and thus need to be levelled up. Instead, it tells us that if criminal law is introduced in each Member State, each national judicial authority will be equipped with more effective investigative and prosecutorial powers and, as a consequence, stronger authorities will cooperate with one another. This argument should be read jointly with the pragmatic argument mentioned above – that resorting to harmonised criminalisation would enhance the powers of each investigative authority and thus would also foster cooperation (Recital 7). Briefly, the official justifications for the adoption of criminal law in the field of market abuse are effectiveness-based and pragmatic. Yet, the following sub-section illustrates that the evidence raised to support the effectiveness of criminal law in this area is rather weak. D.  The Weakness of the Empirical Data on the Regulatory Rationale In its Impact Assessment, the Commission refers to four public consultations on the subject of the need for criminal sanctions to counteract market abuse, which are intended to provide evidence in support of the use of criminal law: a consultation launched by the Commission in its Communication on reinforcing sanctioning regimes in the financial sector; a public conference held on 12 November 2008 by the European Commission on the review of the market abuse regime; the call for evidence on the review of the Market Abuse Directive launched by the Commission on 20 April 2009; and the public consultation on the revision of the Directive launched on 28 June 2010.82 However, a closer look at these documents shows that none of them provided conclusive evidence in favour of the greater effectiveness of criminal sanctions in the context of market abuse.83 The Impact Assessment itself acknowledges that the contributions offered ‘a mixed response to the option of harmonising criminal sanctions in financial services legislation’.84 In several cases, these documents even raised valid concerns as to the suitability of criminal law to increase the effectiveness of the EU policy, and praised the advantages of administrative sanctions, such as the shorter duration of administrative procedures and their reduced burden

82 These are all mentioned in European Commission, ‘Impact Assessment accompanying the document Proposal for a Regulation of the European Parliament and of the Council on insider dealing and market manipulation (market abuse) and the Proposal for a Directive of the European Parliament and of the Council on criminal sanctions for insider dealing and market manipulation’ COM (2011) 651 final (hereinafter Impact Assessment for the Market Abuse Directive) 7. 83 The analysis of the Impact Assessment and the relevant argument is developed in M Miglietti, ‘The First Exercise of Article 83(2) TFEU under Review: An Assessment of the Essential Need of Introducing Criminal Sanctions’ (2014) 5(1) New Journal of European Criminal Law 5. 84 Impact Assessment for the Market Abuse Directive (n 82) 55.

The 2014 Market Abuse Directive  189 of proof, the fact that prosecutors are not often specialised in financial topics and the fact that supervisory authorities are in a better position to address the relevant types of behaviour.85 Yet, despite these shaky empirical foundations, the Commission nonetheless concluded in favour of the use of criminal law, arguably relying on the assumption, among other reasons, that the higher societal disapproval conveyed by the criminal law would ensure a higher level of enforcement. The Council and the Parliament then supported this choice through the negotiations, and criminal sanctions were included in the final text. Such an approach, which relies on an assumption of effectiveness rather than on objective criteria, was criticised for contradicting the need to base decisions to criminalise on clear and factual evidence as established by the 2011 Commission Communication on an EU criminal policy.86 Miglietti in particular suggested that the real objective of EU criminalisation in this context was not actual law enforcement, but simply re-assuring citizens that ‘something’ was being done to address the roots of the financial crisis, which was at its peak during the time in which the negotiations were being conducted.87 Briefly, the regulatory rationale for the harmonisation of criminal law is significantly relied upon to justify the adoption of the Market Abuse Directive, but is poorly backed up in terms of empirical evidence. Conversely, it seems that the strongest actual rationale for the adoption of the text, though not formally acknowledged in the preamble, is a values-based one. E.  The Normative Rationale of the Market Abuse Directive: The Seriousness Threshold The values-based rationale for the EU criminalisation of conduct that results in market abuse can be indirectly deduced from the introduction of a seriousness threshold limiting the scope of the criminal law provisions. The Regulation and the Directive on market abuse target the same type of behaviour (insider dealing, unlawful disclosure of information and market manipulation), but, as mentioned above, they envisage different kinds of sanctions (administrative and criminal respectively). As per Articles 3, 4 and 5 of the Directive, criminal law should be used only for serious conduct. Interestingly enough, such a seriousness threshold was not included in the initial Commission proposal, which only envisaged the intentional character of the misconduct as a distinguishing criterion for criminal law.88 The majority of the Council delegations nonetheless 85 The analysis of these contributions is summarised in Miglietti (n 83) 18. 86 2011 Commission Communication on EU Criminal Policy (n 80). On this, see extensively ch 6, s IV.B. 87 Miglietti (n 83) 25. See also Herlin-Karnell, who recalls the need to base effectiveness on objective criteria: E Herlin-Karnell, The Constitutional Dimension of EU Criminal Law (Oxford, Hart Publishing, 2012) 58. 88 See arts 3 and 4 of the Commission Proposal for a Market Abuse Directive (n 76).

190  Legitimating EU Criminal Law in Practice considered that the distinction between administrative offences and criminal offences should be based on some further substantial criterion in addition to intent.89 Thus, the Council included the seriousness threshold within its general approach.90 The same wording was then included in the European Parliament First Reading.91 The concept of serious crime was previously linked to a deontological approach when discussing Article 83(1) TFEU in Chapter 4, which limits the EU competence to harmonise criminal law only in the field of serious and crossborder crime. The Market Abuse Directive, which is based on a different legal basis (Article 83(2) TFEU), includes the term, but does not define it. Some guidance on how to interpret it can be found in the preamble. Recitals 11 and 12 list some elements that Member States might consider when deciding whether the relevant forms of misconduct are serious. Among these, there are the degree of the ‘impact on the integrity of the market’ and the level of ‘damage caused to the market’, which concern the harm caused by the conduct. Member States are also encouraged to consider some characteristics of the offender, namely if he or she belongs to and acts within a criminal organisation, if he or she has already committed the offence, or if he or she is a person employed or working in the financial sector or in a supervisory or regulatory authority.92 Of relevance for the current discussion is that the threshold of seriousness is considered to be met when the damage caused by the relevant conduct to the interest protected (in this case market integrity) is significant. This approach is an expression of a valuesbased rationale for EU criminalisation. What justifies requiring all Member States to criminalise those forms of market abuse is the disvalue inherent in the prohibited behaviour. Naturally, one could wonder whether the selected legal interest (namely market integrity) is sufficiently precise and sufficiently highly ranked to represent a Rechtsgut worthy of protection via the criminal law.93 Moreover, it 89 See Council of the European Union, Proposal for a Directive of the European Parliament and of the Council on criminal sanctions for insider dealing and market manipulation – outstanding issues, doc 12089/12, pts 9 and 10. 90 See arts 3, 3a and 4 of the Council General Approach, Council of the European Union, Proposal for a Directive of the European Parliament and of the Council on insider dealing and market mani­ pulation (market abuse) (first reading_– general approach, doc 16820/12. 91 See arts 3, 4 and 5 of the European Parliament First Reading. European Parliament, Position of the European Parliament adopted at first reading on 4 February 2014 with a view to the adoption of Directive 2014/ … /EU of the European Parliament and of the Council on criminal sanctions for market abuse (market abuse directive) (EP-PE_TC1-COD(2011)0297). 92 When interpreting the expression ‘serious crime’ used in the Treaty, ch 4 relied on the understanding of Ashworth and Zedner as being one of the crime affecting important interests and involving a high degree of guilt. ch 4, s III.B.i. The list of criteria for identifying the seriousness threshold seems to mirror this distinction. 93 Herlin-Karnell interestingly observes that the same justification – preserving market integrity – is used for the Market Abuse Regulation, and understandably so, given that the text is based on art 114 TFEU, and thus has an internal market legal basis, but also for the Directive, which conversely is based on art 83(2) TFEU and is thus a criminal law instrument: Herlin-Karnell (n 53) 487. This argument suggests that the Directive does not include a very detailed explanation of the specific legal interests that deserve protection by the criminal law.

The 2014 Market Abuse Directive  191 is debatable to what extent the identified conducts actually impact on market integrity.94 Nevertheless, regardless of these debates, it remains the case that one of the reasons justifying the imposition of harmonised criminalisation of market abuse is the features of the conduct at stake. F.  The Market Abuse Directive and Article 83(2) TFEU: An Integrated/ Symbolic Approach in a Utilitarian Institutional Context? The previous discussion has shown that the legitimacy of the choice to resort to criminal law in the context of market abuse apparently only rests on an integrated, deontological and utilitarian approach. Admittedly, preparatory documents, and the Market Abuse Directive itself, consider both the capacity of criminal law to secure enforcement and the inherent disvalue of the relevant misconduct also plays an important role in legitimating resorting to criminal law, implying a deontological approach. However, considering the very weak evidence raised in support of the utilitarian rationale, it seems that actually it is this latter deontological rationale which plays a bigger role. This is interesting if compared with the relevant legal basis, notably Article 83(2) TFEU, which conversely only insists on effectiveness-based legitimacy. Chapter 5 discussed how the Treaty of Lisbon’s sole utilitarian approach in this provision marked a break with the pre-existing integrated approach of the case law of the Court of Justice, which justified the use of criminal law to secure the enforcement of EU policies, but only in the cases in which important interests were at stake. It was also recalled that Article 83(2) TFEU was criticised for fear that it would lead to pure regulatory criminalisation (or droit pénal bureacratique), namely to the use of criminal law wherever there is an enforcement deficit which penal sanctions could correct, regardless of the importance of the values at stake.95 Chapter 6 illustrated how the policy documents re-introduced the need to also include the deontological, values-based perspective when legislating on the basis of Article 83(2) TFEU. The 2014 Market Abuse Directive seems to maintain such an integrated deontological and utilitarian perspective. Not all forms of market abuse behaviour are criminalised – only those which cause great damage and thus which severely impact the protected values. The fears of having purely regulatory criminal law can thus be temporarily suspended. Nonetheless, the fact that the Directive is arguably grounded in a presumption of effectiveness of the criminal law, rather than on an empirical understanding of it, raises different kinds of concerns. In other words, where there seemed to be a risk to have undesirable EU bureaucratic criminal law, the outcome is in fact the opposite, namely 94 Herlin-Karnell observes that already in the debates surrounding the adoption of the 2003 Market Abuse Directive, the question of whether market abuse actually distorts the market was raised: Herlin-Karnell (n 53) 484. 95 See ch 5, s III.B.

192  Legitimating EU Criminal Law in Practice EU symbolic criminal law. However, this is equally problematic, in that it raises the fear of an over-reliance on the criminal law to respond to public opinion concerns and thus of an event-driven criminalisation drift. The PIF Directive, which will be discussed in the next section and is equally grounded in the effectiveness-based Article 83(2) TFEU, conversely provides an example of a slightly more balanced, integrated legitimacy of EU criminal law. IV.  THE 2017 PIF DIRECTIVE: AN INTEGRATED LEGITIMACY FOR EU CRIMINAL LAW?

Protection of the financial interests (PIF) of the EU is one of the first areas where the European (at the time) Community adopted criminal sanctions. Back in 1995, Member States signed a Convention criminalising fraud against the financial interests of the EU, which entered into force in 2002.96 Two Protocols were annexed to the Convention in 1996 and 1997 – entering into force in 2002 and 2009 respectively – one criminalising corruption of EU officials to the extent that it affected the EU budget97 and the other criminalising money laundering.98 The Commission then tried to replace these Conventions in 2001 by proposing a first PIF Directive99 as part of a set of proposals aimed at introducing criminal law obligations in a number of community ‘first pillar’ policy areas, also including intellectual property,100 environment101 and ship-source pollution.102 While the last two proposals famously gave rise to the Environmental Crime cases before the Court of Justice103 and were then turned into directives,104 the PIF and the intellectual property proposals were not followed up. The PIF Convention and Protocols were finally replaced in 2017 by the Directive on the Protection of

96 Council Act of 26 July 1995 drawing up the Convention on the protection of the European Communities’ financial interests [1995] OJ C316/48. 97 Council Act of 27 September 1996 drawing up a Protocol to the Convention on the protection of the European Communities’ financial interests [1996] OJ C313/1. 98 Council Act of 19 June 1997 drawing up the Second Protocol of the Convention on the protection of the European Communities’ financial interests [1997] OJ C221/11. 99 ‘Proposal for a Directive of the European Parliament and of the Council on the criminal law protection of the Community’s financial interests, of 25 May 2001’ COM (2001) 272 (hereinafter PIF Proposal). 100 ‘Proposal for a European Parliament and Council Directive on criminal measures aimed at ensuring the enforcement of intellectual property rights, of 26 March 2006’ COM (2005) 276. 101 ‘Proposal for a Directive of the European Parliament and of the Council on the Protection of the Environment through Criminal Law, of 13 March 2001’ COM (2001) 139. 102 ‘Proposal for a Directive of the European Parliament and of the Council on ship-source ­pollution and on the introduction of sanctions, including criminal sanctions, for pollution offences, of 7 March 2003’ COM (2003) 92. 103 Case C-176/03 Commission v Council [2005] ECR I-7879; Case C-440/05 Commission v Council [2007] ECR I-9097. On both, see extensively ch 5, s III. 104 Directive 2008/99/EC of the European Parliament and of the Council of 19 December 2008 on the Protection of the Environment through Criminal Law [2008] OJ L328/28; Directive 2005/35/EC on Ship Source Pollution and the Introduction of Penalties for Infringements [2005] OJ L255/11.

The 2017 PIF Directive  193 the Financial Interests of the EU (hereinafter the PIF Directive), which is discussed in this section.105 Moreover, under the Treaty of Amsterdam, legislation was approved on counterfeiting the euro106 and to the question of the protection of EU financial interests, which has also been replaced by a directive.107 The importance of the protection of the financial interests of the EU was also explicitly recognised in the Treaties. As discussed in Chapter 4, the Treaty of Maastricht already included an ad hoc provision on the duty to counteract fraud against the EU finances (Article 209A TEC), which was then reproduced in the Treaty of Amsterdam (Article 280 TEC). However, both provisions ruled out the possibility of relying on criminal law to combat EU fraud. Conversely, the Lisbon version of this provision, Article 325 TFEU, does not explicitly exclude the possibility of using criminal law in this field. Moreover, the Treaty of Lisbon adds a further mention to the duty to protect the EU financial interests in Article 310(6) TFEU. Lastly, ad hoc actors have been entrusted with the task of fighting attacks against EU finances. In 1999, the European Anti-Fraud Office (commonly known as OLAF, from the French: Office Européen de Lutte Antifraude) was created to investigate fraud affecting the EU budget and corruption and serious misconduct within the EU Institutions.108 OLAF’s investigations are nonetheless of an administrative, non-criminal nature. Furthermore, as already mentioned in Chapter 4, the Lisbon Treaty re-introduces in Article 86 TFEU what had already been included in the failed Constitutional Treaty, namely a legal basis to establish an EPPO to be entrusted with the task of prosecuting cases of fraud against the financial interests of the EU. A regulation setting up the EPPO was adopted in 2017.109 Briefly, the EU has a long-standing commitment to the protection of its financial interests by means of administrative and criminal law, as demonstrated by the listed legislative and institutional reforms. Against this background, it is argued here that it is not surprising that the PIF Directive arguably benefits from a solid legitimacy. In more detail, the need for criminal sanctions in this 105 Directive (EU) 2017/1371 of the European Parliament and of the Council of 5 July 2017 on the fight against fraud to the Union’s financial interests by means of criminal law [2017] OJ L198, 28 July, p 29 (hereinafter PIF Directive). 106 Council Framework Decision 2000/383/JHA of 29 May 2000 on increasing protection by criminal penalties and other sanctions against counterfeiting in connection with the introduction of the euro [2014] OJ L151/1. 107 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 and replacing Council Framework Decision 2000/383/JHA [2014] OJ L151/1. 108 Regulation 883/2013/EU of the European Parliament and of the Council of 11 September 2013 concerning investigations conducted by the European Anti-Fraud Office (OLAF) and repealing Regulation 1073/1999/EC of the European Parliament and of the Council and Council Regulation (Euratom) 1074/1999 [2013] OJ L248/1. 109 Council Regulation 2017/1939/EU of 12 October 2017 implementing enhanced cooperation on the establishment of the European Public Prosecutor’s Office [2017] OJ L283/1 (hereinafter the EPPO Regulation).

194  Legitimating EU Criminal Law in Practice area is first advocated, especially by the Commission and the Parliament in their preparatory documents, on deontological grounds, that is, on the importance of the interests at stake. Second, there are equally frequent references in the preparatory documents and the final text to utilitarian rationales for having EU criminal law in this policy area, although admittedly these are not always thoroughly supported by empirical evidence. The PIF Directive is thus presented here as a positive example of EU criminal law arguably benefiting from a joint utilitarian and deontological legitimacy, albeit with the caveat mentioned below. The following discussion shows the Impact Assessment, the Explanatory Memorandum and the preamble in the proposed and the final text explicitly acknowledge both utilitarian and deontological justifications (see section IV.A), and how both rationales can also be inferred when looking at the features of the criminalised forms of conduct (see section IV.B) and at the level of sanctions (see section IV.C). A.  Explicit Rationales for Criminal Law in the Field of PIF Offences: An Apparent Focus on Utilitarian Grounds The Commission’s Impact Assessment, Explanatory Memorandum and legislative proposal provide a fairly detailed discussion of different utilitarian and deontological rationales for the use of criminal law in the area of PIF offences. Not all of the Commission’s arguments were picked up by the Parliament and the Council or are reproduced in the preamble of final text, which has a primarily utilitarian focus. Still, the actual criminal policy choices in the PIF Directive discussed below show that these are in fact grounded in both values-based and effectiveness rationales. It is thus worth reflecting on how these rationales were initially articulated in the Commission’s documents and what was retained the final text. i.  An Integrated Approach in the Commission’s Preparatory Documents The Commission’s preparatory documents provide quite an extensive discussion of the rationales for harmonisation in this area. These include a regulatory, effectiveness-based rationale and a cooperation rationale, both of which imply a utilitarian legitimacy for EU criminal law, and a criminalisation values-based rationale, which conversely envisages a deontological legitimacy. The Commission’s Impact Assessment begins by recalling the need to ensure the effectiveness of EU budgetary and financing norms, and that criminal law is chosen in light of its higher deterrence potential.110 Furthermore, the Explanatory 110 Commission Staff Working Paper Impact Assessment (Part I) Accompanying the document Proposal for a Directive of the European Parliament and of the Council on the protection of the financial interests of the European Union by criminal law (SWD (2012) 195 final) (hereinafter PIF Directive Impact Assessment) 4.

The 2017 PIF Directive  195 Memorandum lists the forum shopping argument – that is, the argument that criminals will choose the more lenient jurisdiction if legislation differs across states – to advocate the adoption of EU criminal law in this area, and it links it to the question of deterrence. It states that differences in legislation: [H]ave a negative impact on the effectiveness of the Union’s policies to protect its financial interests, as demonstrated in the Impact Assessment accompanying this proposal. Common offences in all Member States would reduce the risks of divergent practice, as they would ensure a uniform interpretation and a homogeneous way to meet all the necessary prosecution requirements. They would also strengthen the deterrent effect and enforcement potential of relevant provisions and reduce the incentive for potential perpetrators to move to more lenient jurisdictions within the Union to exercise their intentional illegal activities. (Emphasis added)111

Admittedly, the arguments of the Commission are not extensively substantiated by empirical evidence that illustrates why criminal law is more effective in this area and why administrative law is not sufficient. Relying on the deterrent character of criminal law seems more of a principled position than an empirically grounded one.112 Interestingly, the Commission reproduces the same utilitarian arguments in favour of criminal law, though again without much empirical support, when advocating Article 325(4) TFEU as a legal basis for the text. The provision grants the competence to the ‘European Parliament and the Council, acting in accordance with the ordinary legislative procedure … [to adopt] the necessary measures in the fields of the prevention of and fight against fraud affecting the financial interests of the Union with a view to affording effective and equivalent protection in the Member States and in all the Union’s institutions, bodies, offices and agencies’.113 The first argument of the Commission when advocating this legal basis is the inclusion of the term ‘deterrent’ in the first paragraph of Article 325. The provision states that: ‘The Union and the Member States shall counter fraud and any other illegal activities affecting the financial interests of the Union through measures … which shall act as a deterrent and be such as to 111 PIF Proposal (n 99) 3. 112 M Kaiafa-Gbandi, ‘The Commission’s Proposal for a Directive on the Fight against Fraud to the Union’s Financial Interests by Means of Criminal Law (COM (2012) 363 final): An Assessment Based on the Manifesto for a European Criminal Policy’ (2012) 2(3) European Criminal Law Review 319, 325. 113 It is worth mentioning that the provision speaks of equivalent protection, which could refer both to equivalent protection at the national and supranational levels, or that there should be the same protection within each Member States. The former option does not necessarily imply that each Member States should have the same legislation, but simply that Member State should treat the protection of the financial interests of the EU as seriously as the protection of its own financial interests. This option is coherent with the principle of assimilation, which is grounded exactly in this idea of equivalence between the national and supranational levels, and which has regulated the field of the protection of EU financial interests since the Greek Maize case. The principle is now enshrined in art 325(2) TFEU. On this, see the ch 4, s III.A. However, the Commission seems to favour the second option – equivalence between the various Member States – which justifies the presence of harmonised legislation across all Member States.

196  Legitimating EU Criminal Law in Practice afford effective protection in the Member States, and in all the Union’s institutions, bodies, offices and agencies.’ According to the Commission, ‘deterrent … comprises by nature, and historically (see the PIF Convention) a criminal law dimension, since criminal law is needed as a basis to create a risk for potential perpetrators to be caught under embarrassing circumstances, and thus disincentive to commit the illegal act in first place’.114 Furthermore, the Commission’s preparatory documents acknowledge the need to use harmonisation in this field to ensure smooth cooperation. The Impact Assessment insists on having equivalent sanctions to foster mutual trust between national authorities115 and similar substantive rules (for instance, on criminal liability of public officials) to enable cooperation in practice.116 The Commission also stresses the importance to ensure that the levels of sanctions in all Member States meet the threshold for issuing a European Arrest Warrant117 and has introduced the sanctions thresholds in the PIF proposal accordingly.118 Traditionally, EU instruments only included obligations to set minimum ‘maximum sanctions’.119 Finally, the Commission clearly spells out the interests that these norms protect, providing a complementary deontological legitimacy for criminal law in this area. Protecting public money is said to be ‘a solidarity interest at Union level, which is different from the sum of the Member States’ national financial interests … by nature and from the start, placed at Union level’.120 Furthermore, EU public money is said to be fundamental not only to ensure the functioning of the EU, but also its credibility as an institution,121 especially in times of austerity.122 It is in light of this special importance that particularly effective (and thus criminal law) measures are needed. Moreover, the Commission also highlights the scale of the problem, indicating in several points in the preparatory documents the estimates in terms of financial losses deriving from the misuse of EU funds.123 It thus not only deontologically underlines the importance of the interests at stake, but also insists on the degree of harmfulness the relevant forms of conduct cause to these interests.

114 PIF Directive Impact Assessment (n 110) 27; PIF Proposal (n 99) 6. 115 PIF Directive Impact Assessment (n 110) 9 and 11. 116 ibid 14. See also PIF Proposal (n 99) 3. 117 PIF Directive Impact Assessment (n 11), 24; PIF Proposal (n 99) 8. 118 PIF Directive Impact Assessment (n 110); PIF Proposal (n 99) 10. 119 This is only the second time that the Commission introduced such a requirement. The first time was in the ‘Proposal for a Directive of the European Parliament and the Council on the protection of the euro and other currencies against counterfeiting by criminal law, and replacing Council Framework Decision 2000/383/JHA’ COM (2013) 42 final. 120 PIF Directive Impact Assessment (n 110) 27–28; PIF Proposal (n 99) 6–7. 121 PIF Directive Impact Assessment (n 110) 27–28. 122 PIF Directive Impact Assessment (n 110) 1. 123 This can be read in the Impact Assessment both when generally discussing the impact of EU financial crimes (ibid 9 and 22), but also when specifically discussing the impact of each form of conduct that the Commission proposes to criminalise (ibid 18–22).

The 2017 PIF Directive  197 Finally, in the preamble of the proposed text, the Commission explicitly mentions public finances as the protected interest,124 acknowledging the ­deontological dimension to the instrument and the deterrent character of criminal law as a reason to rely on it,125 thus including a complementary mention to the utilitarian, effectiveness-based rationale. ii.  A Focus on the Utilitarian Rationale in the Preamble to the Final PIF Directive The Council’s and the Parliament’s preparatory documents are less elaborate on the question of the legitimacy of PIF crimes. For one thing, both Institutions rejected Article 325(4) TFEU as a legal basis mainly on text-based arguments without referring, as the Commission had done in its proposal, to the functions of criminal law. Both argued that Article 83(2) TFEU, which explicitly mentions the harmonisation of criminal law and which is subject to a number of special institutional arrangements, such as the emergency brake or, by being in the Area of Freedom, Security and Justice, the opt-outs,126 should be lex specialis.127 Admittedly, the Council Legal Service stresses that the arrangements are in place because of the special character of criminal law,128 but it does not specify what this ‘special character’ refers to and whether value-based or effectiveness considerations are included. The most likely explanation is that the Council is referring to the fact that criminal law is a sensitive field for Member States as it is considered close to their sovereignty. However, some insistence on the symbolic character of criminal law can be appreciated in other parts of the Parliament’s preparatory documents. For instance, like the Commission, the Parliament relies on the forum shopping argument to advocate harmonised criminal law in this field, but alongside the deterrence aspect, it also insists on the symbolic importance for the EU to speak with a ‘unique voice’.129 Moreover, the Parliament suggests having a reference to

124 PIF Proposal (n 99) Recital 1 of the preamble. 125 ibid Recitals 12 and 20. 126 On art 83(2) TFEU, see extensively ch 5. 127 Opinion of the Council Legal Service on ‘Proposal for a Directive of the European Parliament and of the Council on the fight against fraud to the Union’s financial interests by means of criminal law (12683/12 DROIPEN 107 JAI 535 GAF 15 FIN 547 CADREFIN 349 CODEC 1924) – Legal basis’, Council Doc 15309/12 (hereinafter Council Opinion on PIF Directive Legal Basis); European Parliament, Committee on Legal Affairs, Opinion on the legal basis for the proposal for a Directive of the European Parliament and of the Council on the fight against fraud to the Union’s financial interests by means of criminal law (COM (2012)0363 – C7-0192/2012 –2012/0193(COD)), of 29 November 2012. 128 Council Opinion on PIF Directive Legal Basis (n 127), especially pt 18. 129 European Parliament, ‘Report on the proposal for a directive of the European Parliament and of the Council on the fight against fraud to the Union’s financial interests by means of criminal law’ COM (2012) 0363 – C7-0192/2012–2012/0193(COD), n A7-0251/2014, of 23 March 2014 (hereinafter European Parliament Report), p 27.

198  Legitimating EU Criminal Law in Practice the credibility of the EU as a key justification for the Directive. In particular, it proposes having Article 1 reading as follows: ‘This Directive establishes necessary measures in the field of prevention of fight against fraud and other illegal activities affecting the Union’s financial interests by defining criminal offences and sanctions, with a view to affording effective and equivalent protection in the Member States and in Union institutions, bodies, offices and agencies and boosting the credibility of Union institutions and initiatives.’130 These observations seem to almost reflect a socialising rationale for the harmonisation of criminal law in this field. The Parliament arguably justifies the resort to EU criminal law on specific important interests (the deontological approach) in order to send a clear message (the utilitarian approach). However, the final text of the PIF Directive, having also incorporated the Council’s amendments,131 does not reproduce all the complexities of the arguments raised by the Institutions during the negotiations, except for a few mentions of the deterrent character of criminal law. In more detail, the preamble’s second recital recalls that a choice in favour of criminal law in this field was made in the 1990s when adopting the first PIF Convention. The third recital then states that continuing to approximate criminal law is essential in order to ensure the effectiveness of EU policies, somehow conjecturing rather than demonstrating that criminal law is the adequate instrument to secure this objective.132 The assumption is that once the choice in favour of criminal law as the most effective tool in this area has been made in the Convention, there is no need to justify it further. However, it should be noted that the 1990s Conventions included far fewer preparatory documents such as Impact Assessments, and thus research on the wisdom of policy choices in the area was much more limited.133 Finally, recital 28 of the preamble explicitly mentions the deterrent potential of criminal law as one of the strengths of this regulatory instrument, but it interestingly also warns against the dangers for fundamental rights of an overreliance on penal law. Conversely, there is no explicit mention in the text of the importance of protecting public finances, as there was in the Commission proposal. Moreover, Article 1 simply announces as an objective of the text that of protecting the EU financial interests, dropping the reference to the credibility of the EU suggested by the Parliament. Nonetheless, as mentioned above, the importance of the values-based rationale for the adoption of criminal law to protect EU financial interests can still be inferred from specific policy choices in the text, which will be discussed below. 130 ibid Amendment 11. 131 See the Council General Approach, Council Doc n 10729/13. 132 See Kaiafa-Gbandi (n 112). 133 The strategy of carrying out Impact Assessments of proposed legislation was introduced – and only for the first pillar – only in the 2000s. See the ‘Communication from the Commission to the Council and the European Parliament Better Regulation for Growth and Jobs in the European Union’ COM (2005) 97 final.

The 2017 PIF Directive  199 B.  Deontological, Values-Based Rationales Guiding Definitions of Offences Alongside an analysis of the rationales explicitly mentioned in the preparatory documents, looking at the features of offences included in the Directive can also give indirect indications on the legitimacy of the offences in the text. As was argued when analysing the Framework Decision on Racism and Xenophobia, the assumption is that a text which criminalises forms of conduct not already present in a significant number of Member States is an expression of an independent EU normative criminal policy rather than a tool to support cooperation. This is even the more the case if it focuses on crimes which are not necessarily transnational. Conversely, a text focusing on cross-border crimes, only setting a common standard to enable cooperation while not excessively going beyond the acquis in a large number of Member States, is most likely justified on cooperation, utilitarian grounds. In practice, the two rationales – cooperation and normative – are naturally often intertwined. However, it is still interesting to look at which one played a bigger role in the context of the PIF Directive. With this aim in mind, the following sub-sections look first at the extent to which the focus is put on the cross-border nature of the crimes and, second, at the extent to which the offences designed in the Directive go beyond the existing legislation in Member States. i.  The Complementary Focus in the PIF Directive on Internal and Cross-border Crimes Historically, the EU fight against crimes affecting its financial interests has focused on both cross-border and national crime. For instance, the mandate of OLAF, the EU anti-fraud office, does not include any obligation to focus only on cross-border behaviour.134 Similarly, Article 86 TFEU, the legal basis for establishing the EPPO, generally speaks of crimes affecting the financial interests of the EU without insisting on their cross-border nature. Yet, naturally, the exact scope of action for the EPPO will be defined by the PIF Directive discussed here.135 That said, interestingly enough, the case studies mentioned in the Impact Assessment to illustrate the need for EU legislation in this field are for the most part non-cross-border in nature and do not mention the need for judicial cooperation.136 Moreover, the connection between harmonisation and cooperation is not reproduced in the final text of the Directive. The preamble does acknowledge that PIF crimes have a transnational nature, yet this is only

134 1999/352/EC, ECSC, Euratom: Commission Decision of 28 April 1999 establishing the European Anti-Fraud Office (OLAF) (notified under document number SEC (1999) 802) [1999] OJ L136/20. 135 Articles 4 and 22 of the EPPO Regulation (n 109). 136 PIF Directive Impact Assessment (n 110) 15, 18, 19, 21, 24.

200  Legitimating EU Criminal Law in Practice to stress that in order to ensure that the fight against them is effective, Member States should adapt their norms on jurisdiction and on ne bis in idem, as well as putting into place the necessary mechanisms for operational and technical assistance for cross-border information exchange.137 There is no mention of the need to harmonise definitions of crimes in this respect. Moreover, final text of the Directive does not include the obligation for Member States to adopt minimum ‘minimum sanctions’ for PIF crimes, which therefore defies the purpose of ensuring the possibility of resorting to European Arrest Warrants for these crimes. The question of sanctions is discussed further in section IV.D below. A specific restriction on cross-border crimes is made only in the context of VAT fraud in Article 2(2) of the Directive. The inclusion of this form of fraud had been an issue of contention during the negotiations. The Commission’s proposal only mentioned VAT fraud in the preamble to specify that this offence was included within the scope of the Directive, with no mention of any crossborder restrictions.138 The Council conversely opposed also extending EU action to VAT fraud in the course of the negotiations, arguing that it would have not been in compliance with the principle of proportionality as only a very small percentage of VAT revenue is to be considered an EU resource.139 In practice, the inclusion of VAT fraud within the scope of the Directive was a sensitive issue for Member States, since the text of the Directive defined the material scope of action for the EPPO. What Member States wished to prevent was having European prosecutors interfering with the national prosecution of VAT fraud cases. The Parliament opposed this view and actually conversely considered the inclusion of VAT fraud within the scope of the Directive to be a pre-condition for agreeing with the proposed Directive and the EPPO Regulation.140 The Institutions then decided to wait for the outcome of the Court of Justice’s decision in the Taricco141 case, which concerned the question of the impact of EU law on a VAT case. In the judgment, the Court further clarified what it had already suggested in previous case law,142 and it confirmed that there is a direct link between the collection of VAT levies and the contribution to the EU budget. This indirectly

137 ibid Recital 23. 138 PIF Proposal (n 99) Recital 4. The inclusion of VAT revenue among EU resources had actually already been clarified by the Court in Case C-539/09 European Commission v Federal Republic of Germany [2011] ECR I-11235; and Case C-617/10 Åklagaren v Hans Åkerberg Fransson [2013] ECR General Court Reports. 139 See art 2 of the Council General Approach (n 131), which explicitly rules out the inclusion of VAT fraud within the scope of the Directive. 140 European Parliament Legislative Resolution of 16 April 2014 on the Proposal for a Directive of the European Parliament and of the Council on the fight against fraud to the Union’s financial interests by means of Criminal law, A-0251/2014; see also the Council’s document to the presidency 12686/1/16REV. 141 Case C-105/14 Criminal Proceedings against Ivo Taricco and Others [2015] ECR Court General Reports. 142 See the cases cited in n 138.

The 2017 PIF Directive  201 legitimated EU regulation on VAT fraud. Scholars have argued that by clarifying that an obligation to introduce criminal law in this area flows directly from a joint reading of Article 325 TFEU and the PIF Convention, the Court actually wanted to send a message to national delegations negotiating in Brussels. It suggested that VAT fraud should be included in the Directive143 and arguably that Article 325 TFEU should be the legal basis for this.144 The compromise which was found was to include VAT fraud within the scope of the Directive, but with three important limitations. Article 2 of the Directive provides that it only applies to offences against the common VAT system (a legal limitation) which involve a total damage of €10,000,000 (a financial limitation) and which are connected to the territory of two or more Member States of the EU (a geographical limitation). To ensure the maximum clarity and legal certainty on the boundaries of the offence, and therefore of their obligations, Member States pushed for the incorporation of the definition of the crime into the core text of the Directive, and not in the preamble as in the Commission proposal, and upfront in Article 2 of the text.145 Incidentally, the EPPO Regulation also reproduces the limits to VAT fraud.146 Article 22 establishes the material scope of the EPPO by referring to the crimes listed in the PIF Directive. It then explicitly adds that for ‘offences referred to in point (d) of Article 3(2) of Directive (EU) 2017/1371 [VAT fraud], as implemented by national law, the EPPO shall only be competent when the intentional acts or omissions defined in that provision are connected with the territory of two or more Member States and involve a total damage of at least EUR 10 million’. The reason for having this definition explicitly reproduced in the text of the Regulation is similarly related to the need to have the maximum level of legal certainty as to what Member States’ obligations are in this area. Indeed, the PIF Directive can be amended by a qualified majority as per Article 83(2) TFEU, whereas the EPPO Regulation would require unanimity, as per Article 86 TFEU. As a consequence, directly incorporating the description of the VAT Fraud offence, as opposed to simply including a reference to the PIF Directive on this point, implies that in order to modify the scope of the EPPO in relation to VAT fraud, the approval of each Member State would be necessary. Conversely, modification of other definitions of crimes falling within the scope of the EPPO, which are only mentioned by reference to the text of the PIF Directive, can also be achieved through amendments to the Directive itself passed with a qualified majority. 143 R Sicurella, ‘A Blunt Weapon for the EPPO? Taking the Edge off the Proposed PIF Directive’ in W Geelhoed, L Erkelens and A Meij (eds), Shifting Perspectives on the European Public Prosecutor’s Office (The Hague, TMC Asser Press, 2018) 99, 112. 144 See N Parisi, ‘Chiari e scuri nella direttiva relativa alla lotta contro la frode che lede gli interessi finanziari dell’Unione’ (2017) 9 Giurisprudenza Penale, available at: http://giurisprudenzapenale. com/wp-content/uploads/2017/09/Scarica-il-contributo.pdf, 6. 145 Sicurella (n 143) 114. 146 EPPO Regulation (n 109).

202  Legitimating EU Criminal Law in Practice It is now interesting to reflect on what the theoretical implications are of insisting on the cross-border character of VAT fraud in terms of the functions of harmonisation and ultimately of the envisaged legitimacy of EU criminal law. It was mentioned earlier that a focus on transnational crimes can be interpreted as evidence of an instrumental role for harmonisation as a tool designed purely to create a level playing field for judicial cooperation. However, this reasoning does not provide a good fit with the criminalisation of VAT fraud in the PIF Directive. Admittedly, there is a close link between the definition of crimes and their enforcement in this context, since the PIF Directive is meant to define the scope of action of the EPPO. However, the EPPO is an EU agency, whose establishment is grounded in an earlier supranational normative decision to prosecute EU fraud. The EU has for a long time pursued the criminalisation of PIF crimes, the PIF Directive being the last step in this normative agenda. The EPPO is established to ensure the enforcement of such EU normative decisions on what forms of behaviour are to be considered intolerable through the use of adequate enforcement tools. The decision to only focus on cross-border crime should thus be considered against this background – as simply an extra limit to the expansion of EU criminal law set by Member States, not necessarily pointing to the link between harmonisation and judicial cooperation. Actually, the explicit mention of the cross-border nature of VAT fraud suggests that this requirement should not apply to the other offences included in the Directive, which will be discussed in more detail below. This confirms the interpretation suggested above that the Directive focuses on the relevant behaviour regardless of the presence of an inherent transnational nature.147 ii.  The Limited Harmonising Impact of the PIF Directive The second aspect that is relevant to the discussion on the respective weight of the values-based and cooperation rationales in the PIF Directive is the degree of the Directive’s harmonisation impact. The EU texts pre-dating the PIF Directive required Member States to criminalise fraud against the financial interests of the EU (Articles 1 and 2 of the 1995 PIF Convention), corruption (Articles 2–5 of the First Protocol) and money laundering (Article 2 of the Second Protocol). In the PIF Directive’s Impact Assessment, the Commission claims that this legal framework does not sufficiently protect the financial interests of the EU, also citing practitioners’ opinions to this aim. It points out that not only are there enforcement and implementation gaps in relation to these offences, but there are also a number of other harmful forms of behaviour which EU law should address.148 In fact, on the subject of enforcement gaps, the latest Commission Report on the implementation of the PIF Convention gave a relatively positive picture

147 Parisi 148 PIF

(n 144) 8. Directive Impact Assessment (n 110) 18–22.

The 2017 PIF Directive  203 of national legislation protecting the financial interests of the EU, especially in relation to corruption and money laundering.149 Still, the Commission’s Impact Assessment pointed to a few shortcomings concerning the legislation on fraud, such as ‘lacking coverage of all types of expenditure by the fraud offence in Italy’ or ‘too demanding intent requirements for fraud in Belgium’.150 On the issue of legislative gaps, the Commission lists a number of further offences, not included in the PIF Convention, which would deserve criminalisation. These are embezzlement/misappropriation, breach of professional secrecy, obstructing public tender procedures, conflict of interests and abuse of power.151 Interestingly, when justifying the need to address these problems, the Commission does not necessarily focus on differences in terms of legislation, which might be an obstacle to cooperation, but rather on the actual absence of criminalisation. It highlights that only five countries include the definition of embezzlement and misappropriation in their legislation and argues that this is a problem in itself as it ‘leads to unclear levels of protection for EU money which reduces the deterrent effect and enforcement potential of such provisions’.152 Only the offence of breach of professional secrecy is said to be present in a significant number of Member States, although the scope of the offence is not sufficiently wide to also encompass the general duty of non-disclosure in the EU Staff Regulations.153 In either case, there is no mention of difficulties in cooperation. Similarly, the Commission recalls that at least 13 Member States do not include the offence of ‘abusing public tender procedure’ and that general offences do not usually apply because some elements conditioning their application are not met, eg, fraudulent misrepresentation or a certain proven amount of damage to the EU budget.154 The Commission also recalls that the 2011 European Parliament resolution on the fight against fraud underlined the need for action in the field of public procurement.155 It then concludes again that the non-existence of the offence leads to a lack of deterrent effect and enforcement capabilities in the Member States concerned, with an estimated annual loss of €40 million to the EU budget.156 Lastly, the Commission mentions that favouritism and abuse of power are not to be found in at least 20 Member States. This legislative gap is considered a problem in itself, and not because divergence in legislation might hinder

149 Second Report from the Commission, Implementation of the Convention on the Protection of the European Communities’ financial interests and its protocols COM (2008) 77 final. 150 PIF Directive Impact Assessment (n 110) 18. 151 ibid 19–22. 152 ibid 19. 153 ibid 20. 154 ibid. 155 European Parliament Resolution of 6 April 2011, on the protection of the Communities’ ­financial interests and the fight against fraud, P7_TA(2011)0142, A7-0050/2011. 156 This entails an estimated €40 million of losses for the EU budget annually; see PIF Proposal (n 99) and the references mentioned therein.

204  Legitimating EU Criminal Law in Practice judicial cooperation.157 If anything, if very few states include the relevant offence in their legal system, there is also little appetite for cooperation. In the actual legislative proposal, the Commission reproduces definitions of fraud and of corruption that are similar to those found in the Convention, except for a few elements,158 and adds a definition of misappropriation and of abuse for public tender procedure,159 while no mention is made of favouritism and abuse of power. Moreover, as has already been mentioned, the Commission specifies in the preamble that VAT fraud is to be included in the broad definition of EU fraud in Article 3, since VAT levies constitute an important part of the EU budget.160 The final text, which incorporates the amendments proposed by the Council and the Parliament, retained the definition of fraud and corruption as well as the addition of misappropriation.161 However, it dropped that of abuse for public tender procedures, which is not present as a self-standing offence. Only Article 3(2)b, which defines ‘fraud affecting the Union’s financial interests’, specifies that fraud includes ‘fraud in respect of procurement related expenditure’, ‘at least’ when acts ‘were committed in order to make an unlawful gain for the perpetrator or another by causing a loss to the Union’s financial interests’. Lastly, the definition of VAT fraud was moved from the preamble to the core text in Article 2(2), adding the requirements in terms of financial damage and having a cross-border character. iii.  Interim Conclusions: Deontological/Integrated Rationales Justifying Harmonisation On the basis of the analysis above, it is possible to infer two aims behind the criminalisation choices in the PIF Directive. There is first the need to ‘Lisbonise’ the criminalisation obligations included in the Convention and the Protocols, that is, to ensure that the more stringent enforcement mechanisms available as a result of the Treaty of Lisbon can secure Member States’ greater compliance with pre-existing EU standards in this field. Admittedly, Giuffrida highlights how national legislation, at least in a selected number of countries (namely Italy, Germany and France) already covered and, in several cases, went beyond the EU standards enshrined in the Directive.162 Still, the ‘Lisbonisation’ ­rationale remains a valid one for those contexts where EU standards have not yet been met. 157 PIF Impact Assessment (n 110) 21. 158 Articles 3 and 4(3) of the PIF Proposal (n 99). See also Parisi (n 144) 10, who points out that in relation to fraud, the definition in the offence also extends to a number of offences not directly connected to public tenders. 159 Article 4(1) and 4(3) of the PIF Proposal (n 99). 160 ibid Recitals 4 and 5. 161 Articles 3, 4(2) and 4(3) of the PIF Directive (n 5). 162 F Giuffrida ‘The Protection of the Union’s Financial Interests after Lisbon’ in R Sicurella, V Mitsilegas, R Parizot and A Lucifora (eds), General Principles for a Common Criminal Law Framework in the EU (Milan, Giuffrè, 2018) 246.

The 2017 PIF Directive  205 Second, it is argued that the adoption of the PIF Directive is motivated by a need to expand the range of criminal law in all Member States regardless of and actually before any need for cooperation arises, but that this is due to valuesbased reasons – in other words, to effectively prevent the financial implications that these forms of behaviour have on EU public money, an estimate of which is also indicated for each type of conduct in the Impact Assessment.163 This normative rationale is fairly evident in the Commission proposal, which is more ambitious in terms of scope and therefore of harmonisation impact. Yet, such a rationale can also be appreciated in the final text of the Directive, which does include at least a new crime not covered by the pre-existing legislation. It is also confirmed by the fact that, but for the exception of VAT fraud (which is not a conclusive one), both the preparatory documents and the final text do not particularly insist on the cross-border nature of PIF crimes. Briefly, the adoption of the PIF Directive stems more from a supranational normative decision to have criminal law in this policy area than from a need to create a level playing field among all Member States forcing at least some of them to amend their criminal law agenda purely for cross-border enforcement purposes. Such a normative decision, as was illustrated above, is grounded in deterrence, utilitarian rationales, but also in values-based deontological reasons related to the importance of the interests that the EU wants to protect, for which only penal sanctions are adequate. Interestingly, a similar combination of deontological and utilitarian rationales can be appreciated in the choice of the level of sanctions for PIF crimes. Though the question of sanctioning is a different one from that of criminalisation,164 it is still interesting to look at how the choices in terms of penalties are justified in order to illustrate how an integrated agenda pervades different dimensions of the PIF Directive. C.  Deontological and Utilitarian Rationales in the Definitions of Sanctions As mentioned above, the preparatory documents and the proposed PIF Directive’s preamble justify the need for a level playing field for sanctions on the basis of the needs of effective judicial cooperation. The requirement for ‘minimum minimum sanctions’ included by the Commission in its proposal is especially linked

163 Each specific offence then of course has a more specific legal interest; for a discussion of the legal interest of corruption in the PIF Directive, see, for instance, MA Bartolucci, ‘Dei rapporti e delle interferenze tra la C.D. Direttiva PIF e le norme penali Spagnole e Italiane in materia di corruzione’ (2019) 6 Diritto Penale Contemporaneo 135. 164 The two processes can technically follow two different philosophies, with utilitarian reasons justifying resorting to criminal law in a certain policy area, while the choice of the sanctions is governed by a retributive, deontological logic. See more extensively the discussion on the principle of proportionality in ch 2, s II.B.

206  Legitimating EU Criminal Law in Practice to the need to meet the European Arrest Warrant threshold.165 But it is also considered to be an undisputable indication of the seriousness of the offence, as it establishes a threshold below which judges cannot go.166 Alongside this, the Commission also explicitly refers to the deterrence potential of criminal sanctions reinforcing the utilitarian rationale for its sanctioning choices.167 The Explanatory Memorandum recalls how in relation to economic crime, ‘criminal sanctions can have a particularly deterrent effect, as potential perpetrators can be expected to make a certain calculation of risks before deciding to engage in such criminal activities’.168 The preamble to the proposed PIF Directive also underlines that the chosen levels of sanctions should remain proportionate to the seriousness of the offences.169 The proposed text correspondingly requires Member States to scale the degree of the sanctions according to the seriousness of the financial damage caused by each form of behaviour. In particular, if the damage is under €10,000, no criminal sanctions need to be imposed for any of the offences mentioned. Such a linking of the choice in favour of criminal law to the caused damage is evidence of a deontological understanding of the role for criminal law. Then, the directive requires that imprisonment sanctions to be imposed for conduct causing damages between €10,000 and €100,000 euros (or €30,000 for money laundering and corruption). For forms of conduct causing damages above the mentioned threshold, the imprisonment sanctions must have a minimum sanction of six months and a maximum of five years. As was clarified in Chapter 2 in the discussion on retrospective and prospective proportionality, linking the amount of the sanction to the caused damage is an expression of a deontological approach to sanctioning which looks at the values to be protected rather than the objectives to be pursued.170 Admittedly, Kaiafa-Gbandi strongly criticises the quality of the retributive proportionality exercise carried out in this context. She argues that the principle of proportionality, being a fundamental EU principle, deserves a more holistic discussion. The Commission only hints at a strictu sensu analysis relating every introduced minimum/maximum level of penalty to the offence to which it is attached. Kaiafa-Gbandi argues that. first, a horizontal assessment (comparing the proposed penalties with those imposed for different offences at a European level) should be included. Then, proportionality needs to be assessed vertically (comparing the proposed penalties with those imposed for similar offences in the Member States).171 She claims that in fact, the only real rationale for the Commission’s sanctioning choices is ensuring that the European Arrest Warrant

165 See

s IV.A.i. (n 143) 106. 167 Recital 12 of the PIF Proposal (n 99). 168 PIF Proposal (n 99) 8; see also PIF Directive Impact Assessment (n 110) 22. 169 Recital 12 of the PIF Proposal (n 99. 170 See ch 2, s II.B.ii. 171 Kaiafa-Gbandi (n 112) 330. 166 Sicurella

The 2017 PIF Directive  207 threshold is met.172 Be that as it may, regardless of the quality of the proportionality testing, it remains the case that the Commission insists on a broad need for coherence between the sanctions and the degree of harm caused to protected interests. The final text of the Directive reproduces a penalty threshold and a set of justifications, similar to those in the Commission’s proposal, except in relation to minimum sanctions, and a less explicit focus on retributive proportionality. Both the Council and the Parliament opposed the introduction of ‘minimum minimum sanctions’.173 The Parliament in particular argued against it because, among other things, such a system would not respect the diversity of legal systems and the need for judicial discretion, and it would not be consistent with the position that the Parliament held in other legislative files.174 Incidentally, the Parliament also suggested eliminating the categorisation of offences as ‘minor’ and/or ‘serious’, and thus accordingly the different penalty threshold of €100,000 and €30,000 respectively. It suggested that all PIF offences which are punished with criminal law sanctions should be taken seriously and thus punished seriously, showing an even stronger normative stance on the ontological undesirability of the forms of behaviour harming EU financial interests. The preamble to the PIF Directive conversely reproduces the explicit mention of deterrence-based, utilitarian rationales for criminal sanction in recital 15, though without a reference to the European Arrest Warrant threshold. Moreover, recital 28 refers to proportionality, but only in a general fashion, without necessarily linking it to the sanctioning choices. Nonetheless, in practice, the text of the Directive reproduces the Commission’s deontological scaling of sanctions according to the caused damage. Article 7(4) includes the €10,000 threshold below which no criminal law sanctions should be envisaged. The final text also leaves Member States free to choose their criminal law sanctions for offences causing damage up to the value of €100,000.175 Article 7(3) then requires Member States to envisage at least four years’ imprisonment as a maximum sanction for each crime, one year less than the initial proposal. The only exception to this scheme is VAT fraud, which in terms of damage is always presumed to be considerable, following what had been the Parliament proposed amendment for all crimes. VAT fraud should always deserve a maximum sanction of at least four years’ imprisonment. The final compromise on this difficult point during the negotiations is thus that internal VAT fraud should be left to Member

172 ibid. 173 Article 7(3) of the Council General Approach on the Proposal for a Directive of the European Parliament and of the Council on the criminal law protection of the Community’s financial interests, of 10 June 2012, Doc 10729/13. 174 European Parliament, Report of 25 March 2014, PE524.832v02-00, A7-0251/2014, amendments in nn 27 and 28. 175 This threshold applies to all offences, contrary to the PIF Proposal, which set a lower threshold of €30,000 for money laundering and corruption offences.

208  Legitimating EU Criminal Law in Practice States, but cross-border fraud should incur harsh sanctions for deontological reasons, namely that by definition, it causes considerable damage to important interests. D.  The PIF Directive as (Relatively) Soundly Legitimate EU Criminal Law The previous sections have shown how the preparatory documents and the actual text of the directive identify (albeit to different degrees) at least three rationales for the harmonisation of criminal law. These are easing judicial cooperation and enabling judicial authorities to resort to European Arrest Warrants, which implies a utilitarian legitimacy for EU criminal law; ensuring the highest effectiveness of EU policy on preserving the EU budget, which again justifies the restriction of personal liberty on utilitarian grounds; and ensuring that ­interests considered of high importance, notably EU financial interests, are protected throughout Europe, which implies a deontological legitimacy for EU criminal law. Admittedly, in the text there are a number of points which could have been argued better or policy choices which are not entirely convincing. For instance, in a similar fashion to what was said in relation to the Market Abuse Directive, the deterrence potential of criminal law in this field was not sufficiently substantiated with empirical evidence. The problem was admittedly more serious in the context of market abuse. In that case, the Commission relied on evidence arguing against the use of criminal sanctions to justify the introduction of penal sanctions for the first time in a field which up until then had been regulated by administrative law. This sensitively weakened the text’s regulatory rationale, the alleged main rationale for criminal law in this area. In the context of PIF crimes, the Commission, as well as the other Institutions, appears to be operating under the assumption that the pre-existing choice to introduce criminal sanctions in the 1990s to protect the EU budget does not need to be subject to further empirical questioning, but merely updates, possibly due to the importance of the interests at stake, which are supranational interests par excellence. Furthermore, the elimination of the requirement of minimum ‘minimum sanctions’ in the final text of the PIF Directive sits at odds with the goal of supporting cooperation. However, these aspects can be ascribed to the inherent ‘compromise nature’ of EU law that must accommodate the different agendas of three Institutions. That aside, it is argued here that the PIF Directive still represents a valid example of integrated, deontological and utilitarian legitimacy. V. CONCLUSIONS

This chapter has aimed to shed some light on how the rationales for harmonisation discussed from a doctrinal point of view and explored in Treaty and policy

Conclusions  209 language play out in practice. It has provided examples of the values-based criminalisation rationale (the Framework Decision on Racism and Xenophobia, the Market Abuse Directive and the PIF Directive); the effectiveness-based regulatory rationale (the Market Abuse Directive and the PIF Directive); the cooperation rationale (the Framework Decision on Racism and Xenophobia, and also the PIF Directive); and the socialising rationale (the Framework Decision on Racism and Xenophobia). Going beyond this descriptive exercise and drawing one overarching conclusion about the selected sources of legitimacy for EU criminalisation policy is an arduous task, due to the patchwork structure of EU criminal law. Still, on the basis of the three case studies analysed here, some general trends can be identified. First, all three instruments, which concern diverse policy areas and were adopted in different historical periods for EU criminal law, stem from the fact that the utilitarian legitimacy for EU criminal law is the weakest legitimating factor. The Framework Decision on Racism and Xenophobia and the Market Abuse Directive respectively recall the cooperation and the effectiveness-based rationales as the main justification for EU criminalisation in the relevant policy areas. This would imply a utilitarian justification for EU criminal law. Yet, a closer look at how these rationales were substantiated (or not) and at the actual policy choices in the text clarifies that the utilitarian argument was merely a façade. In both cases, the deontological rationale – that is, the importance of the values at stake – was to be the decisive factor. Both the Framework Decision and the Market Abuse Directive were thus qualified as ‘symbolic criminal law’, The PIF Directive was said to be the closest example to an integrated model for legitimacy. The values-based criminalisation rationales strongly supported the adoption of criminal law. The cooperation rationale is mentioned several times in the text. However, it was demonstrated that given the limited focus on the transnational nature of the crimes at stake and the harmonisation impact, it is unlikely to play a big role in practice. The effectiveness-based regulatory rationale is often recalled in the text and in the preparatory documents, but admittedly without much empirical evidence to support it. However, this case should be distinguished from the market abuse case, where the Commission actually mentions empirical evidence against the effectiveness of criminal law and then disregards it. Moreover, the use of criminal law in the context of PIF crimes had already occurred in the past and also found support in the case law of the Court of Justice in Taricco. Moreover, it is also interesting to note the tension between legal basis, policy documents and policy-making that these case studies uncover. Both in the case of the Framework Decision on Racism and Xenophobia and in the case of the Market Abuse Directive, but also in the case of the PIF Directive (though to a lesser extent), the symbolic character of EU criminal law legislation is at odds with the emphasis on the utilitarian legitimacy in the relevant Treaty norms. The Treaty of Amsterdam insisted on cooperation and the Treaty of Lisbon insisted on the effectiveness of criminal law (in Article 83(2) TFEU),

210  Legitimating EU Criminal Law in Practice regrettably neglecting the values-based dimension of criminal law. The policy documents adopted by three Institutions, especially post-Lisbon, progressively complemented the utilitarian focus in the Treaty, clarifying that values-based consideration would also be taken into account when legislating in practice. The relevant EU legislative instruments only acknowledged the latter dimension, using criminal law to protect important values, while ignoring the second, effectiveness-based dimension; in other words, they did not justify why criminal law or harmonised criminal law is the best tool to achieve this objective. Overreliance on the values-based rationale as opposed to the effectivenessbased rationale is possibly due to the fact that naturally principled claims are easier to make than empirically substantiated ones. However, this trend is problematic from an EU constitutional law on at least three levels. First, focusing too much on only one source of legitimacy for EU criminal law, be it deontological or utilitarian legitimacy, while neglecting the other is not consistent with EU constitutional values. As has been noted earlier, these require an integrated, restrictive approach to EU criminal law. Second, particularly neglecting the utilitarian dimension of EU criminal law causes tensions with the principle of subsidiarity. It was explained that subsidiarity requires the verification that criminal law is effective in achieving the desired result and that harmonised criminal law is more efficient than national criminal law. Failure to consider the effectiveness of criminal law in preventing market abuse, as the Institutions do, for instance, in the Market Abuse Directive, implies failing to comply with the first step of the subsidiarity principle. Failure to verify whether harmonised criminal law is actually necessary – for instance, to support judicial cooperation, as was arguably the case in the Framework Decision on Racism and Xenophobia – implies failing to comply with the second step of the subsidiarity principle. Among the three instruments, the PIF Directive is thus the closest example to the integrated model for legitimacy that would be coherent with the EU constitutional commitment to EU values and the principle of subsidiarity. Third, at least in the case of the Market Abuse Directive, failing to consider utilitarian arguments implies failing to also respect competence boundaries. Article 83(2) TFEU, the relevant legal basis, lists the utilitarian, effectivenessbased ground for criminal law as a requirement for the legality of EU law texts adopted on this legal basis, even before effectiveness as a requirement for the legitimacy of EU criminal law.

8 Conclusions

‘W

hy EU criminal law?’ This is seemingly a simple question, which, in fact, has a multitude of dimensions. These include a political dimension, a social dimension and a normative dimension. Exhaustively discussing such a query would thus require drawing from political and federal theory debates, from sociological and criminological analyses, and from criminal legal theories. Among these various dimensions, this book has focused on the latter normative, criminal legal theory one. It has investigated when the use of criminal law by the EU, as opposed to less intrusive forms of regulation, is legitimate, as it does not excessively encroach upon the liberty of individuals. This main aim of the book was articulated around three research questions: • first, what indications do EU constitutional values and principles provide on the legitimate use of criminal law by the EU?; • second, how does the EU justify resorting to criminal law in practice?; and • thirdly, are such EU criminalisation choices consistent with EU constitutional values and principles? The analysis has nonetheless shown that the legitimacy of EU criminal law from a normative perspective, which is read through the lens of criminal legal theory, is tightly intertwined with the legitimacy of EU criminal law from a political perspective in a federal sense. In other words, the question ‘Why EU criminal law?’ – why the EU should resort to criminal law rather than administrative law in a specific policy area – is thus inherently linked to the question ‘Why EU criminal law?’ – understood that criminal law is necessary, why should the EU be the one to enact it and not the Member States? The book does not extensively discuss federal or political theory and only focuses on EU constitutional law principles and strategies regulating power allocation across levels. However, the connection between the political and normative dimensions of the legitimacy of EU criminal law has been one of the recurring themes in the book. I.  THE LONG-STANDING DOCTRINAL DEBATE ON THE LEGITIMACY OF CRIMINAL LAW

The debate on the legitimacy of criminal law is a classic one of liberal criminal theory and has been ongoing since the nineteenth century. During the

212  Conclusions Enlightenment, criminal legal theorists refuted the pre-modern connection between crime and sin. Since then, criminal legal theorists have extensively reflected on what the new secular foundations of modern liberal criminal law should be. They have consistently argued that criminal sanctions, especially imprisonment, represent the ultimate public threat to the liberty and dignity of individuals. Embedded in an intellectual and political climate where the liberty of individuals was the most important value, scholars have accordingly argued that the use of criminal law should be an ultima ratio for the state. On these bases, scholars developed a wide range of theories on when penal law can be resorted to can be exceptionally considered legitimate. Among the various models, a conceptual opposition was traced between what Paul Roberts defines as deontological and utilitarian theories of criminal law. The deontological position includes Mill’s and Feinberg’s Anglo-Saxon harm principle theories and their continental counterpart, namely Roxin’s and Bricola’s theory of Rechtsgut or bene giuridico. It grounds the legitimacy of criminal law in the importance of the interests it seeks to protect. The second utilitarian position connects the legitimacy of criminal law to the desirability of the objectives it can achieve. Legitimate objectives can be collective utility as in Benthamite utilitarianism, or, in more recent law and economics analysis by Becker or Posner, even just the enforcement of prescriptive provisions. The deontological approach has traditionally benefited from more support from scholars and it is often considered to be conceptually contrasted with the utilitarian approach. However, this book has embraced the position explained by von Hirsh and Simester that criminal law is truly legitimate when it is based on both utilitarian and deontological grounds. This approach sets the highest quantitative and qualitative limits on the use of criminal law, and thus ultimately offers the highest level of protection of the liberty and dignity of individuals. II.  THE RELEVANCE AND LEGAL DIMENSIONS OF THE DEBATE ON THE LEGITIMACY OF EU CRIMINAL LAW

Whether such a fundamental criminal legal theory debate is also relevant to the EU legal order is not self-evident, considering that the EU is not a sovereign state and has a limited range of actions at its disposal. This book nonetheless argues that the question of the legitimacy of criminal law is not only relevant to the EU legal order, but also acquires a special constitutional law relevance at the supranational level. The debate on the legitimacy of criminal law has remained a doctrinal one. National constitutions naturally set fundamental rights that limit state action. This means that criminal law cannot unjustifiably restrict citizens’ fundamental rights, for instance, by prohibiting peaceful political demonstrations. Moreover, national fundamental texts include references to the conceptual premises of an integrated approach to criminalisation, namely the values of liberty and dignity

The EU Constitutional Values and Principles  213 of individuals. However, national constitutions hardly pronounce themselves on the values that criminal law should protect or the objectives it should pursue, and nor have constitutional courts. However, this position is not uncontroversial. In his famous dissenting opinion in the BVG incest case, Judge Hassemer argued in favour of a constitutional law acknowledgement of the Rechtsgut theory. Other authors see a clear advantage in leaving criminal legal theory outside constitutional codification. It arguably respects the freedom of democratically elected legislators. Moreover, it allows more flexibility and sophistication in the development of criminal legal theories models. This book has shown that the situation is different for EU constitutional norms, which contain precise indications for the functions of EU criminal law. As in federal states, the multi-level nature of the EU requires the EU Treaties to clearly identify the boundaries to the EU range of action, including in criminal law. Yet, unlike the Spanish or German Constitutions, which set static contentbased limitations for each level range of action in criminal law, EU competences are often drafted in a purposive way. They identify the objectives that the EU should pursue, including through criminal law. This arguably ensures more certainty as to what admissible criminalisation criteria are. But by crystallising them in legal provisions, it also reduces the flexibility that the development of a model for the legitimacy of EU criminal law might require. The legal dimension of the debate on the justifications for EU criminal law and the tensions this might create have been a second recurring theme throughout the book. III.  THE EU CONSTITUTIONAL VALUES AND PRINCIPLES AND THE LEGITIMACY OF EU CRIMINAL LAW

In 1992, in the Treaty of Maastricht, Member States made three specific choices for the constitutional structure of the EU that would determine the role that criminal law would and should play in the EU legal order. First, the EU competence to impose criminalisation obligations on the Member States by means of the harmonisation of criminal law was introduced. This competence was progressively expanded content-wise with each Treaty amendment. However, it admittedly remained an indirect one. Even in the Treaty of Lisbon, while the EU has acquired competence to adopt regulations in other areas of criminal justice, it still only has the power to set minimum standards in terms of definitions of offences by means of directives, which Member States then have to incorporate into national law. In practice, however, this book has shown that in a significant number of cases, EU definitions of offences are drafted very precisely, which does not leave a wide margin of discretion to Member States on how to implement them. EU law thus has a significant impact on national criminalisation policies. The liberty and dignity of EU citizens can be restricted as a result of supranational policy decisions. This book has argued that such supranational criminalisation policy choices thus beg for justification.

214  Conclusions The Treaty of Maastricht also set the theoretical foundations as to what these justifications should be. With the aim of re-defining the EU, both internally and externally, as a project with an axiological core, the Treaty of Maastricht acknowledged respect of fundamental rights as a founding principle of the EU. Human rights would then be qualified as an EU value in the Treaty of Lisbon. The freedom and dignity of individuals would also be individually listed as founding values. This was a deliberate constitutional choice for the EU. It implies that EU legislation must respect fundamental rights as a matter of the rule of law. But this book argues that it also requires a more careful selection of the values or objectives to be protected by or pursued through EU criminal law, as a matter of coherence with the identity of the EU legal order. In short, the Treaty of Maastricht set the basis for a liberal minimalistic approach to EU criminal law, that is, criminal law grounded in an integrated deontological and utilitarian legitimacy. Both symbolic EU criminal law (potentially ineffective criminal law solely justified on the fact that it is meant to protect very important values) and bureaucratic EU criminal law (effective criminal law that conversely targets behaviour which does not affecting important interest) would thus be at odds with the EU principles and values. Lastly, the Treaty of Maastricht introduced the principle of subsidiarity, an efficiency-based criterion ordering the allocation of powers across national and supranational government levels. With subsidiarity, the drafters of the Treaty made a specific choice to ground the legitimacy of EU or national action in its relative effectiveness to achieve a certain policy objective balanced against the relevant costs. Alternative allocating criteria could have been the democratic credentials of each governance level or the values they protect. While subsidiarity is meant to govern the federal balance of power, it also has implications for its liberal dimension. This book has demonstrated how this principle requires the legitimacy of EU criminal law to be based on utilitarian grounds. This is because it legally requires factoring in effectiveness-based considerations when adopting EU norms, including EU criminal law norms. Therefore, even before being inconsistent with EU values, ineffective symbolic EU criminal law is constitutionally unlawful for not respecting subsidiarity. This is a first example of how the debate on the legitimacy of EU criminal law has a stronger legal dimension at the EU level. IV.  THE SPECIFICS OF THE EU CRIMINALISATION PROCESS

Against the background sketched above, this book has explored the following two research questions. It has investigated how the EU has justified its criminalisation choices (question 1) and to what extent the EU choices have been consistent with EU constitutional values and principles (question 2). In short, it has investigated to what extent the EU, in its criminalisation policies, has lived up to the constitutional standards it has set for itself. This exercise required

The Main Argument of the Book  215 reading criminalisation and harmonisation theories jointly. When harmonisation is pursued to ensure that all Member States punish violations of important values with the harshest instrument (the values-based criminalisation rationale) or to ensure that all Member States attach the most effective sanctions to violations of EU law (the regulatory rationale), it is fair to consider that EU criminal law is respectively legitimate on utilitarian and deontological grounds. However, there are other traditional rationales for harmonisation which have a less evident connection with the debate on the legitimacy of EU criminal law in the criminal legal theory sense that is intended here. They are more closely linked to the federal question of why Member States need to have similar norms in this field. For instance, one of the most common rationales for the harmonisation of criminal law is that of facilitating judicial cooperation and mutual recognition, for instance allowing a better functioning of European Arrest Warrants (the cooperation rationale). This book has interpreted this rationale as implying a utilitarian, enforcement-based rationale for EU criminal law. Harmonisation justified on this basis would require more liberal Member States to expand their criminal law – and thus further restrict the liberty and dignity of EU citizens – simply to enable Member States with more repressive agendas to enforce their own criminal laws transnationally and arrest fugitives abroad. This was illustrated in the book through the examples of the differences between Greek and Italian criminal law on the crime of looting, and the arrest of Greek protesters in the 2015 anti-expo demonstrations in Milan. V.  THE MAIN ARGUMENT OF THE BOOK: SYMBOLIC EU CRIMINAL LAW IN A BUREAUCRATIC CRIMINAL LAW INSTITUTIONAL FRAMEWORK

While it is not possible to give a general answer to the question of what the EU approach to the legitimacy of EU criminal law is, this book still identifies a number of important trends. The main argument is that both Treaty and policy language have traditionally insisted on a mainly utilitarian legitimacy for EU criminal law. Resort to such a harsh instrument was and, partially, still is justified on the basis of criminal law effectiveness in securing enforcement goals. Reference to the need for complementary deontological rationales for EU criminalisation was only introduced with the latest Treaty amendments and policy documents. The risk several scholars highlighted was that bureaucratic EU criminal law would be adopted on these constitutional bases. Criminal law that was only effectiveness-based would be at odds with the EU commitment to promote liberal values. In practice, interestingly, both under the Treaties of Amsterdam and Lisbon, EU criminal law legislation was adopted which only justified resorting to criminal law on deontological, values-based arguments. Briefly, if the shape of EU Competences in Lisbon, notably Article 83(2) TFEU, led scholars to fear a drift towards regulatory criminal law simply filling enforcement gaps,

216  Conclusions in fact ineffective symbolic EU criminal law was adopted on the aforementioned legal basis. This book concludes that the latter is equally problematic from an EU constitutional values perspective and it could even be in breach of subsidiarity and the relevant competence boundaries. In more detail, both in the Treaties of Maastricht and Amsterdam, resort to EU criminalisation was initially justified on a cooperation rationale and thus as a utilitarian tool to secure the cross-border enforcement of criminal law. This could be deduced, for example, from the subordinate position of the relevant legal basis in the Treaty and from the policy discourse in Tampere and The Hague Programmes, which clearly identifies a link between harmonisation and cooperation. Arguably, the introduction of the objective of the Area of Freedom, Security and Justice in the Treaty of Amsterdam also suggested social engineering goals, such as justice and free movement, as legitimate ones for EU criminalisation. These were equally expressions of the utilitarian legitimacy of criminal law. Despite having acknowledged fundamental rights as a founding, identityshaping principle of the EU, the Council of Ministers and the European Council had not yet drawn the relevant implications in terms of criminal policy-shaping. And yet it is under the Treaty of Amsterdam that one of the most striking examples of EU symbolic criminal law – only justified on the basis of the values it protects – was adopted. The 2009 Framework Decision on Racism and Xenophobia1 included a number of references to a cooperation rationale justifying the adoption of the text. However, it was arguably simply pursuing a deontological agenda, possibly including some EU identity-building purposes. This was said to be an illegitimate expansion of EU criminal law. Oddly, in contrast with the Treaty framework at the time, the failure of the Framework Decision to complement deontological rationales with utilitarian rationales was also inconsistent with the minimalist approach that a true commitment to liberal values would require. For the same reasons, it was also said to be in breach of the principle of subsidiarity. Since a common definition of hate speech and denialism was not necessary to ease judicial cooperation, it was not clear on what grounds harmonisation on these fields was necessary. The Treaty of Lisbon and the Stockholm Programme represent a turning point in that they loosen the link between harmonisation and judicial cooperation and admit a more autonomous role for the harmonisation of criminal law. In the Stockholm Programme, the European Council in particular expressly calls for questioning the connection between harmonisation and mutual recognition, exploring further roles for harmonisation. However, this evolution only partially implied a stronger, more integrated legitimacy for EU criminal law. 1 Council Framework Decision 2008/913/JHA of 28 November 2008 on combating certain forms and expressions of racism and xenophobia by means of criminal law [2008] OJ L328/55.

The Main Argument of the Book  217 The TFEU identifies two broad areas for EU criminalisation, which Mitsilegas defines as EU securitised and EU functional criminalisation. This book has argued that Article 83(1) TFEU legitimates securitised criminalisation – that is, criminalisation to respond to security threats for EU citizens – on both utilitarian and deontological grounds. The first requirement set by the provision is to focus only on cross-border crimes. This was interpreted as referring to the traditional use of harmonisation to ease judicial cooperation, thus supporting the cross-border enforcement of criminal law, as had been the case under the Treaties of Maastricht and Amsterdam. The second requirement, namely to focus only on serious cross-border crime, was interpreted as innovatively requiring a complementary deontological harm threshold for EU criminalisation. Article 83(1) TFEU was thus presented as a positive example of an integrated approach to criminalisation that was coherent with the EU commitment to protect and promote liberal values. Conversely, Article 83(2) TFEU legitimates functional criminalisation – that is, EU criminalisation to safeguard EU policies, objectives and interests – only on effectiveness-based, utilitarian grounds. EU criminal law is meant to be used wherever there is an enforcement gap in terms of EU law at a national level. However, there are no qualitative limits on the importance of the interests in the relevant policy area. While not closely linked to judicial cooperation, this new area of EU competence introduces a different sort of utilitarian legitimacy for EU criminal law – it legitimates resorting to bureaucratic EU criminal law, which is clearly inconsistent with the EU core values. Admittedly, after the Treaty of Lisbon, the EU Institutions partially modified their approach. In both the Stockholm Programme and in ad hoc EU criminalisation documents, the Council, the Commission and the Parliament declared their commitment to a more integrated approach to EU criminalisation, including to functional criminalisation. In this, they showed more awareness of what respect of the liberty and dignity of individuals, which in Lisbon were elevated to the status of EU values, imply in terms of criminal policy. Still, similarly to what had happened under the Treaty of Amsterdam, the drift towards bureaucratic criminal law that could be read in the Treaty was not reflected in the legislation; quite the contrary. The first text adopted on the basis of Article 83(2) TFEU, the 2014 Market Abuse Directive,2 is indeed another example of symbolic EU criminal law. In this case, the failure to properly justify the adoption of criminal law on utilitarian grounds does not only imply inconsistency with EU values, but also arguably a breach of EU competence and of subsidiarity. However, there are some exceptions to the argument that in practice EU criminal law follows a purely symbolic agenda. The 2017 Directive on the Protection

2 Directive 2014/57/EU of the European Parliament and of the Council of 16 April 2014 on criminal sanctions for market abuse (market abuse directive) [2014] OJ L173/17.

218  Conclusions of the Financial Interests of the EU3 was included here as a third case study of a positive example of a partially more integrated approach to EU criminal law, which was more consistent with EU liberal values. Still, also in this case, the utilitarian, effectiveness-based rationale, if present, is nonetheless the weakest rationale. VI.  THE DANGERS OF AN EXPANSION OF NON-LEGITIMATE EU CRIMINAL LAW

This book has illustrated the shortcomings of the EU approach to criminalisation both within Treaty norms and in the adopted legislation. The central argument has been an ‘internal coherence’ argument. EU Treaty norms on criminal law present some inconsistencies with the founding values that the EU selected for itself. They disproportionately insist only on a utilitarian legitimacy of EU criminal law. A number of EU legislative instruments are similarly inconsistent with EU values, but this time for conversely pursuing a purely deontological agenda. This further implies tension with EU competences and subsidiarity. Admittedly, it is sometimes the case that the EU legislator pursues a normative, values-based agenda, pushing the boundaries of EU competences and subsidiarity. The EU legislation on fair trial rights in criminal procedure was recently adopted on the basis of Article 82(2) TFEU, which grants to the EU the competence to harmonise procedural rights to the extent that these can facilitate mutual recognition. Legislation adopted on this basis, such as the Directive on victims’ rights4 or on the right of interpretation and translation,5 embraced a broad approach that arguably extended the listed rights both to transnational and national cases.6 In light of this, it can be7 (and was) argued8 that the EU is pursuing a normative human rights agenda, rather than an instrumental agenda, aimed at supporting mutual recognition. This creates tension with the respect of competence boundaries and subsidiarity.

3 Directive (EU) 2017/1371 of the European Parliament and of the Council of 5 July 2017 on the fight against fraud to the Union’s financial interests by means of criminal law [2017] OJ L198/29. 4 Directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012 establishing minimum standards on the rights, support and protection of victims of crime, and replacing Council Framework Decision 2001/220/JHA [2012] OJ L315/57. 5 Directive 2012/13/EU of the European Parliament and of the Council of 22 May 2012 on the right to information in criminal proceedings [2012] OJ L142/1. 6 See art 1 in both Directive 2012/29/EU (n 4) and 2012/13/EU (n 5). 7 I Wieczorek, ‘The Principle of Subsidiarity in EU Criminal Law’ in C Brière and A Weyembergh (eds), The Needed Balances in EU Criminal Law: Past, Present and Future (Oxford, Hart Publishing, 2017) 71, 101–02. 8 J Öberg, ‘Subsidiarity and EU Procedural Criminal Law’ (2015) 5 European Criminal Law Review 19.

The Dangers of an Expansion of Non-legitimate EU Criminal Law  219 Going beyond the criminal justice sphere, Article  114 TFEU provides an almost open-ended competence for the harmonisation of national legislation. EU legislation is justified provided that it contributes to the establishment of the internal market, namely that it eases cross-border economic transactions. Nonetheless, this legal basis has been famously and unlawfully relied upon to promote public health policies that have no connection with market-making.9 At the same time, the spread of non-legitimate, symbolic or bureaucratic criminal law is also a well-known phenomenon at the national level. As was discussed in Chapter 1, leading criminal legal theorist like Ashworth, Delmas-Marty, Silva Sanchez, Baratta and Husak have increasingly warned against the worrying spread of the criminal law in the UK, France, Spain, Italy and the US respectively. Yet, this book concludes that the combination of both phenomena – nonlegitimate criminal law adopted at the EU level that in some cases stretches EU constitutional limits – is even more problematic than the two phenomena taken separately. On the one hand, poorly legitimate and possibly constitutionally unlawful internal market legislation undoubtedly unduly restricts Member States sovereignty. Yet, it has a disparate impact on individuals’ rights, in some cases enhancing them and in others restricting them. Legislation on preventing tobacco advertising might work in the interests of individuals, while of course limiting the economic freedoms of producers or traders.10 Other internal market instruments, such as in the field of data protection, might imply more coercive measures for individuals’ rights.11 Conversely, as has been extensively argued in this book, an expansion of EU criminal law beyond EU competences is not only in tension with Member States’ sovereignty but also by definition implies restrictions on the personal liberty and dignity of individuals. On the other hand, symbolic or bureaucratic national criminal law is an undesirable policy choice, which can be remedied by means of

9 See Case C-376/98 Germany v Parliament and Council [2000] ECR I-8419. For a comment on this case and the subsequent legislation, see S Weatherill, ‘The Limits of Legislative Harmonization Ten Years after Tobacco Advertising: How the Court’s Case Law Has Become a Drafting Guide’ (2011) 12 German Law Journal 827. On values and the harmonisation of legislation in internal market policies, see A Alemanno and A Garde, ‘The Emergence of an EU Lifestyle Policy: The Case of Alcohol, Tobacco and Unhealthy Diets’ (2013) 50 Common Market Law Review 1745; and B van Leeuwen, ‘Euthanasia and the Ethics of Free Movement Law: The Principle of Recognition in the Internal Market’ (2018) 19(6) German Law Journal 1417. 10 Directive 2003/33/EC of the European Parliament and of the Council of 26 May 2003 on the approximation of the laws, regulations and administrative provisions of the Member States relating to the advertising and sponsorship of tobacco products [2003] OJ L152/16. 11 It can clearly be read in the preamble to the General Data Protection Regulation (GDPR) that the aim of EU regulation is both to enhance data protection while at the same time to actually enhance the circulation of data. See Recital 9 and 10s in Regulation 2016/679/EU 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.

220  Conclusions de-criminalisation. However, this safety-valve for the system does not extend to criminal law norms adopted in observance of EU law obligations. Indeed, failure to respect them might trigger infringement proceedings. Moreover, it is unclear whether the EU has any de-criminalisation competences itself. In practice, the EU has never re-examined its criminalisation choices and actually has consistently increased the level of repressiveness of EU criminal law, as the case study on PIF crimes and market abuse, as well other examples on trafficking in human beings, child pornography, attacks against information systems and terrorism, have shown.

Bibliography Albrecht H-J, ‘Imprisonment and Alternatives to Prisons: Changes and Prospects in a Comparative Perspective’ (2006) 3(6) Revista Académica – Facultad de Derecho de la Universidad la Salle 27 Alemanno A and Garde A, ‘The Emergence of an EU Lifestyle Policy: The Case of Alcohol, Tobacco and Unhealthy Diets’ (2013) 50(6) Common Market Law Review 1745 Alemika E et al, The Theory and Practice of Criminal Justice in Africa (Tswhane, Institutes for Security Studies Monographs Series, 2016) Alix J, ‘Les Frontières de l’harmonisation autonome’ in Giudicielli-Delage G and Lazerges C (eds), Le droit pénal de l’Union Européenne au Lendemain du Traité de Lisbonne (Paris, Société de Législation Comparée, 2012) 147 Alogna R and Riondato S (eds), Studies on the Criminal Law Reform in the Post-Soviet Countries (Padua, Padua University Press, 2013) Ambos K, Treatise on International Criminal Law. Volume 1: Foundations and General Part (Oxford, Oxford University Press, 2013) Arroyo Zapatero L and Munoz de Morales Romero M, ‘Droit pénal européen et Traité de Lisbonne : le cas de l’harmonisation autonome (article 83.1 TFEU)’ in Giudicielli-Delage G and Lazerger C (eds), Le Droit Pénal de l’Union Européenne au Lendemain du Traité de Lisbonne (Paris, Société de Législation Comparée, 2012) 113 Ashworth A, ‘Towards a Theory of Criminal Law’ (1989) 1 Criminal Law Forum 41 ——. ‘Is the Criminal Law a Lost Cause?’ (2000) 116 Law Quarterly Review 225 ——. ‘Conceptions of Overcriminalisation’ (2007–08) 5 Ohio State Journal of Criminal Law 407 ——. Principles of Criminal Law (Oxford, Oxford University Press, 2009) Ashworth A and Zedner L, ‘Defending the Criminal Law: Reflection on the Changing Character of Crime, Procedure, and Sanctions’ (2008) 2 Criminal Law and Philosophy 21 ——. ‘Preventative Orders: A Problem of Undercriminalisation?’ in Duff RA et al (eds), The Boundaries of the Criminal Law (Oxford, Oxford University Press, 2010) 59 ——, ‘Prevention and Criminalisation: Justification and Limits’ (2012) 15(4) New Criminal Law Review 542 Asp P, The Substantive Criminal Law Competence of the EU (Stockholm, Stiftelsen Skrifter utgivna av Juridiska fakulteten vid Stockholms universitet, 2012) Ayres I and Braithwaite J, Responsive Regulation: Transcending the Deregulation Debate (Oxford, Oxford University Press, 1992) Azoulai L, ‘The Case of Fundamental Rights: A State of Ambivalence’ in Micklitz H-W and de Witte B (eds), The European Court of Justice and the Autonomy of the Member States (Antwerp, Intersentia, 2012) 1 ——. ‘The Complex Weave of Harmonisation’ in Arnull A and Chalmers D (eds), Oxford Handbook of EU Law (Oxford, Oxford University Press, 2015) 589 Baker D, ‘Constitutionalising the Harm Principle’ (2008) 27 Criminal Justice Ethics 3 ——. The Right Not to Be Criminalized: Demarcating Criminal Law’s Authority (Farnham, Ashgate, 2011) Backer E and Harding C, ‘A Longitudinal Study of the Third Pillar’ (2009) 34(1) European Law Review 25 Baldwin DA, ‘The Concept of Security’ (1997) 23 Review of International Studies 5 Baratta A, ‘Les fonctions instrumentales et les fonctions symboliques du droit pénal’ (1991) 15(1) Deviance et société 1 Bartolucci MA, ‘Dei rapporti e delle interferenze tra la C.D. Direttiva PIF e le norme penali Spagnole e Italiane in materia di corruzione’ (2019) 6 Diritto Penale Contemporaneo 135

222  Bibliography Bassiouni MC, ‘The Philosophy and Policy of International Criminal Justice’ in Vohrah LC, Pocar F, Featherstone Y, Fourmy O, Graham MF, Hocking J and Robson N (eds), Man’s Inhumanity to Man: Essays on International Law in Honour of Antonio Cassese (Leiden, Brill, 2003) 65 Becker GS, ‘Crime and Punishment: An Economic Approach’ in Becker GS and Landes WM (eds), Essays in the Economics of Crime and Punishment (New York, National Bureau of Economic Research, Distributed by Columbia University Press, 1974) 1 Bengoetxea J, Jung H and Nuotio K, ‘Ultima Ratio, a Principle at Risk: European Perspectives’ (2013) 3(1) Onati Socio-legal Series. Available at: http://opo.iisj.net/index.php/osls/issue/archive Bentham J, A Fragment on Government, Burns JH and Hart HLA (eds) (Cambridge, Cambridge University Press, 1988 [1776]) ——. Introduction to the Principles of Morals and Legislation, Hart HLA and Burns JH (eds) (York, Methuen, 1982 [1789]) Bergeron H and Colson R (eds), European Drug Policies: The Ways of Reform (Abingdon, Routledge, 2016) Bergstrom M, ‘The Dynamic Evolution of EU Criminal Law’ in M Bergstrom and AJ Cornell (eds), European Police and Criminal Law Co-operation (Oxford, Oxford University Press, 2014) 185 Berkman ET, ‘Sacrificed Sovereignty?: Dutch Soft Drug Policy in the Spectre of Europe without Borders’ (1996) 11(1) Boston College International and Comparative Law Review 173 Bermann GA, ‘Taking Subsidiarity Seriously’ (1994) 94 Columbia Law Review 332 Bernardi A, ‘Europeizzazione del diritto penale e progetto di costituzione europea’ (2004) Diritto Penale e Processo 5 ——. ‘La competenza penale accessoria dell’Unione Europea. Problemi e Prospettive’ (2012) 1 Diritto Penale Contemporaneo 43 Berthier A, ‘Transparency in EU Law-Making’ (2016) 7(4) Era Forum. DOI: 10.1007/s12027-0160445-0 Birnbaum JMF, ‘Uber das Erordernis einer Rechtsverletzung zum Begriff des Verbrechens’ (1834) 15 Archiv des Criminalrechts 149 Boister N, ‘Transnational Criminal Law?’ (2003) 14 European Journal of International Law 953 Boodman M, ‘The Myth of Harmonisation of Laws’ (1991) 39 American Journal of Comparative Law 699 Bowels R, Faure M and Garoupa N, ‘The Scope of Criminal Law and Criminal Sanctions: An Economic View and Policy Implications’ (2008) 35(3) Journal of Law and Society 389 Bricola F, ‘Teoria Generale del Reato’ (1973) XIX Novissimo Digesto Italiano 3 Brown D, ‘History’s Challenge to Criminal Legal Theory’ (2009) 3 Criminal Law and Philosophy 271 Caeiro P, ‘The Relationship between European and International Criminal Law (and the Absent(?) Third)’ in Mitsilegas V, Bergstrom M and Konstantidines T (eds), Research Handbook on EU Criminal Law (Cheltenham, Edward Elgar, 2016) 580 Calaza Lopez S, ‘Principio Rectores del Proceso Judicial Espagnol’ (2011) 8 Revista de Derecho UNED 49 Calderoni F, Organized Crime Legislation in the European Union, Harmonization and Approximation of Criminal Law, National Legislations and the EU Framework Decision on the Fight against Organized Crime (Dordrecht, Springer, 2010) Carrera S and Guild E, ‘The European Council’s Guidelines for the Area of Freedom, Security and Justice 2020 Subverting the “Lisbonisation” of Justice and Home Affairs?’ (2014) 13/14 CEPS Policy Paper 3 Cervini R and Tavarez J, Principios de Cooperaçao Judicial Penal Internacional no Protocolo do Mercosul (Sao Paulo, Editora Revista dos Tribunais, 2000) Chalmers D, Davies G and Monti G, European Union Law (Cambridge, Cambridge University Press, 2010) Cooper I, ‘The Watchdogs of Subsidiarity: National Parliaments and the Logic of Arguing in the EU’ (2006) 44(2) Journal of Common Market Studies 281 Craig P, ‘Subsidiarity: A Political and Legal Analysis’ (2012) 50(1) Journal of Common Market Studies 72

Bibliography  223 Craig P and de Búrca G, EU Law: Text, Cases and Materials (Oxford, Oxford University Press, 2015) Crawford A and Hucklesby A (eds), Legitimacy and Compliance in Criminal Justice (Abingdon, Routledge, 2013) Cronin-Furman K and Taub A, ‘Lions and Tigers and Deterrence, Oh My: Evaluating Expectations of International Criminal Justice’ in Schabas WA, McDermott Y and Hayes N (eds), The Ashgate Research Companion to International Criminal Law (Farnham, Ashgate, 2013) 435 Damaska MR, What is the Point of International Criminal Justice? (2008) 83(1) Chicago-Kent Law Review 329 Davies G, ‘Subsidiarity: The Wrong Idea, in the Wrong Place, at the Wrong Time’ (2006) 43(1) Common Market Law Review 63 ——. ‘Democracy and Legitimacy in the Shadow of Purposive Competences’ (2015) 21(1) European Law Journal 2 De Búrca G, ‘The Principle of Proportionality and its Application in EC Law’ (1993) 13(1) Yearbook of European Law 105 ——. ‘The Language of Rights and European Integration’ in Show J and More G (eds), New Legal Dynamics of European Union (Oxford, Clarendon Press, 1995) 29 ——. ‘The Principle of Subsidiarity and the Court of Justice as an Institutional Actor’ (1998) 36(2) Journal of Common Market Studies 217 ——. ‘Reappraising Subsidiarity’s Significance after Amsterdam’ (1999) 7/99 Harvard Jean Monnet Working Paper 4, www.jeanmonnetprogram.org/archive/papers/99/990701.rtf ——. ‘The Evolution of EU Human Rights Law’ in Craig P and de Búrca G (eds), The Evolution of EU Law (Oxford, Oxford University Press, 2011) 465 De Hert P, ‘Division of Competences between National and European Levels with Regard to Justice and Home Affairs’ in Anderson M and Apap J (eds), Police and Justice Cooperation in the New European Borders (The Hague, Kluwer Law International, 2002) 65 ——. ‘Cybercrime and Jurisdiction in Belgium and the Netherlands: Lotus in Cyberspace – Whose Sovereignty is at Stake?’ in Koops B and Brenner SW (eds), Cybercrime and Jurisdiction: A Global Survey (The Hague, TCM Asser Press, 2006) 88 ——. ‘The EU Data Protection Reform and the (Forgotten) Use of Criminal Sanctions’ (2014) 4(4) International Data Privacy Law 262 ——. ‘EU Criminal Law and Fundamental Rights’ in Mitsilegas V, Bergstrom M and Konstantidines T (eds), Research Handbook on EU Criminal Law (Cheltenham, Edward Elgar, 2016) 105 De Hert P and Boulet G, ‘The Co-existence of Administrative and Criminal Law Approaches to Data Protection Wrongs’ in De Hert P and Wright D (eds), Enforcing Privacy. Legal and Technological Approaches (Berne, Springer International Publishing, 2016) De Hert P, Boulet G and Wieczorek I, ‘Fondements et objectifs des incriminations au niveau Européen: le cas de la criminalité informatique’ in Bernard D, Cartuyvels Y, Guillain C, Scalia D and van de Kerchove M (eds), Fondements et Objectifs des incriminations et des peines en droit Européen et International (Brussels, Anthémis, 2013) 267 De Hert P, Gutwirth S, Snacken S and Dumortier E, ‘La montée de l’Etat pénal: que peuvent les droits de l’homme ?’ in Cartuyvels Y, Dumont H, Ost F, van de Kerchove M and van Drooghenbroeck S (eds), Les droits de l’homme: bouclier ou épée du droit pénal ? (Brussels, Publications des FUSL/ Bruylant, 2007) 235 De Hert P and Kloza D, ‘WikiLeaks, Privacy, Territoriality and Pluralism: What Does Matter and What Does Not?’, TILT Weblog Law and Technology. Available at: http://www.vub.ac.be/LSTS/ pub/Dehert/Dehert_361.pdf De Hert P and Wieczorek I, ‘Testing the Principle of Subsidiarity in EU Criminal Policy: The Omitted Exercise in the Recent EU Documents on Principles for Substantive European Criminal Law’ (2012) 3–4 New Journal of European Criminal Law 394 De Jong C, ‘The European Parliament Resolution of 22 May 2012 on an EU Approach to Criminal Law’ in Weyembergh A and Galli F (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Edition de l’Université de Bruxelles, 2013) 37

224  Bibliography De Witte F, ‘Sex, Drugs & EU Law: The Recognition of Ethical and Moral Diversity in Europe’ (2013) 50(6) Common Market Law Review 1545 De Wree E, vander Beken T and Vermeulen G, ‘The Transfer of Sentenced Persons in Europe’, 11(1) Punishment and Society (2009) 111 Delmas-Marty M, Pour un droit commun (Paris, Le Seuil, 1994) ——. ‘The European Union and Penal Law’ (1998) 4(1) European Law Journal 87 ——. Le Flou du droit, 2nd edn (Paris, PUF, 2004) ——. Les forces imaginantes du droit. Le relatif et l’universel (Paris, Le Seuil, 2004) ——. ‘Le processus d’interaction’ in Delmas-Marty M, Pieth M and Sieber U (eds), Les Chemins de l’harmonisation pénale (Paris, Société de Legislation Comparé, 2008) 419 Delmas-Marty M and Spencer JR (eds), European Criminal Procedures (Cambridge, Cambridge University Press, 2005) Delmas-Marty M et al, Corpus Juris: Introducing Penal Provisions for the Purpose of the Financial Interests of the European Union (Paris, Economica, 1997) Delors J, ‘The Principle of Subsidiarity: Contribution to the Debate’ in European Institute of Public Administration, Subsidiarity: The Challenge of Change. Proceedings of the Jacques Delors Colloquium (Maastricht, EIPA, 1991) Devlin P, The Legal Enforcement of Morality (Oxford, Oxford University Press, 1965) Dorinda O, The Enlightenment (Cambridge, Cambridge University Press, 2013) Dougan M, ‘Minimum Harmonisation and the Internal Market’ (2000) 37(4) Common Market Law Review 853 ——. ‘The Vicissitudes of Life at the Coalface: Remedies and Procedures for Enforcing EU Law before the National Courts’ in Craig P and de Búrca G (eds), The Evolution of EU Law (Oxford, Oxford University Press, 2011) 407 Douglas H, Overcriminalisation (Oxford, Oxford University Press, 2008) Douglas-Scott S, ‘The Rule of Law in the European Union: Putting the Security into the Area of Freedom, Security and Justice’ (2004) 29(2) European Law Review 219 ——, ‘The Problem of Justice in the European Union’ in Dickson J and Eleftheriadis P (eds), Philosophical Foundations of European Union Law (Oxford, Oxford University Press, 2012) 413 Dripps D, ‘The Liberal Critique of the Harm Principle’ (1998) 17(3) Criminal Justice Ethics 3 Du Bois-Pedain A, ‘The Place of Criminal Law Theory in the Constitutional State’ in Simester AP, du Bois-Pedain A and Neumann U (eds), Liberal Criminal Theory (Oxford, Hart Publishing, 2018) 305 Dubber D, ‘Theories of Crime and Punishment in German Criminal Law’ (2005) Buffalo Legal Studies Research Paper 2005/02 1 Duff RA, ‘Towards a Modest Legal Moralism’ (2012) 8(1) Criminal Law and Philosophy 217 Duff RA and Marshall SE, ‘Remote Harms and the Two Harm Principles’ in Simester A P, du  Bois-Pedain A and Neumann U (eds), Liberal Criminal Theory (Oxford, Hart Publishing, 2018) 205 Duff RA et al, ‘Introduction’ in Duff RA et al (eds), The Structure of the Criminal Law (Oxford, Oxford University Press, 2011) 1 Durnescu I, Montero Perez de la Tudela E and Ravagnani L, ‘Prisoner Transfer and the Importance of the “Release Effect”’ (2016) 17(4) Criminology and Criminal Justice 450 Eady D and Smith ATH, Arlidge, Eady and Smith on Contempt (London, Sweet & Maxwell, 2015) Elholm T, ‘Does EU Criminal Law Cooperation Always Mean Increased Repression?’ (2009) 17(3) European Journal of Crime Criminal Law and Criminal Justice 191 Elholm T and Colson R, ‘The Symbolic Purpose of EU Criminal Law’ in Colson R and Field S (eds), EU Criminal Justice and the Challenges of Diversity: Legal Cultures in the Area of Freedom, Security and Justice (Cambridge, Cambridge University Press, 2016) 48 Ellis E (ed), The Principle of Proportionality in the Laws of Europe (Oxford, Hart Publishing, 1999) Emiliou N, The Principle of Proportionality in European Law: A Comparative Study (Dordrecht, Kluwer, 1996)

Bibliography  225 Engle E, ‘The History of the General Principle of Proportionality: An Overview’ (2012) 10 Dartmouth Law Journal 1 Estella A, The Principle of Subsidiarity and its Critique (Oxford, Oxford University Press, 2002) European Criminal Policy Initiative, ‘The Manifesto on European Criminal Policy’ (2011) 1(1) European Criminal Law Review 86 Faure M, ‘European Environmental Criminal Law: Do We Really Need it?’ (2004) 13 European Environmental Law Review 18 Feinberg J, ‘The Expressive Function of Punishment’ in Doing and Deserving (Princeton, Princeton University Press, 1970) 95 ——. Harm to Others: The Moral Limits of the Criminal law (Oxford, Oxford University Press, 1984) ——. Offence to Others: The Moral Limits of the Criminal Law (Oxford, Oxford University Press, 1985) ——. Harm to Self: The Moral Limits of the Criminal Law (Oxford, Oxford University Press, 1986) ——. Harmless Wrongdoing: The Moral Limits of the Criminal Law (Oxford, Oxford University Press, 1988) Fiandaca G and Musco E, Diritto Penale – Parte Generale (Bologna, Zanichelli, 2014) Fichera M, ‘Criminal Law beyond the State: The European Model’ (2013) 19(2) European Law Journal 174 ——. ‘Sketches of a Theory of Europe as an Area of Freedom, Security and Justice’ in Fletcher M, Herlin-Karnell E and Matera C (eds), The European Union as an Area of Freedom, Security and Justice (Abingdon, Routledge, 2017) 34 Findlay M, The Globalization of Crime: Understanding Transitional Relationships in Context (Cambridge, Cambridge University Press, 1999) Fletcher G, ‘Criminal Theory in the Twentieth Century’ (2001) 2 Theoretical Inquiries in Law 267 ——. ‘Parochial versus Universal Criminal Law’ (2005) 3 Journal of International Criminal Justice 20 Fletcher M, Lööf R and Gilmore B, EU Criminal Law and Justice (Cheltenham, Edward Elgar, 2008) Flore D, ‘Fondements et objectifs de la coopération internationale en matière pénale’ in Bernard D, Cartuyvels Y, Guillain C, Scalia D and van de Kerchove M (eds), Fondements et Objectifs des incriminations et des peines en droit Européen et International (Brussels, Anthémis, 2013) 427 Forti G, ‘Principio del danno e legittimazione “personalistica” della tutela penale’ in Fiandaca G and Francolini G (eds), Sulla Legittimazione del Diritto Penale. Culture Europee-continentali e AngloAmericane a confronto (Turin, Giapichelli, 2008) 43 Frase RS, ‘What’s ‘Different’ (Enough) in Eighth Amendment Law?’ (2013) Ohio State Journal of Criminal Law 9 Galbiati R and Garoupa N, ‘Keeping Stigma out of Administrative Law: An Explanation of Consistent Beliefs’ (2007) 15 Supreme Court Economic Review 273 Garfinkel P, Criminal Law in Liberal and Fascist Italy (Cambridge, Cambridge University Press, 2016) Geyer F and Guild E (eds), Security vs Justice? Police and Judicial Cooperation in the European Union (Abingdon, Routledge, 2016) Gibbs AH, ‘Reasoned “Balance” in Europe’s Area of Freedom, Security and Justice’ (2011) 17(1) European Law Journal 121 Gindre E, ‘L’harmonisation pénale accessoire. Eléments de réflexion sur la place du droit pénale au sein de l’Union Européenne’ in Giudicielli-Delage G and Lazerges C (eds), Le droit pénal de l’Union Européenne au Lendemain du Traité de Lisbonne (Paris, Société de Législation Comparée, 2012) 201 Giudicielli-Delage G and Manacorda S (eds), L’intégration Pénale Indirecte (Paris, Société de législation comparée, 2005)

226  Bibliography Giuffrida F, ‘The Protection of the Union’s Financial Interests after Lisbon’ in Sicurella R, Mitsilegas V, Parizot R and Lucifora A (eds), General Principles for a Common Criminal Law Framework in the EU (Milan, Giuffrè, 2018) 246 Grasso G, ‘Il Trattato di Lisbona e le nuove competenze penali dell’Unione europea’ in Bertolino M, Eusebi M and Forti, G. (eds), Studi in onore di Mario Romano (Naples, Jovene, 2011) 2344 ——. ‘Il Trattato di Lisbona e le nuove competenze penali dell’Unione europea’ in Studi in onore di Mario Romano, 4th edn (Naples, Jovene, 2011) 2308 Green SP, Lying, Cheating, and Stealing: A Moral Theory of White-Collar Crime (Oxford, Oxford University Press, 2005) Grogger J, ‘Arrests, Persistent Youth Joblessness, and Black/White Employment Differentials’ (1992) 74(1) Review of Economics and Statistics 100 ——. ‘The Effect of Arrests on the Employment and Earnings of Young Men’ (1995) 110(1) Quarterly Journal of Economics 51 Guild E (ed), Constitutional Challenges to the European Arrest Warrant (Tilburg, Wolf Legal Publishers, 2006) Gur-Arye M and Weigend T, ‘Constitutional Review of Criminal Prohibitions Affecting Human Dignity and Liberty: German and Israeli Perspectives’ (2011) 44 Israel Law Review 63 Harcourt BE, ‘The Collapse of the Harm Principle’ (1999) 1 Journal of Criminal Law and Criminology 109 Harding C and Banach-Gutierrez JB, ‘The Emergent EU Criminal Policy: Identifying the Species’ (2012) 37 European Law Review 758 Hart HLA, Punishment and Responsibility: Essays in the Philosophy of Law (Oxford, Clarendon Press, 1968) Hassemer W, ‘The Harm Principle and the Protection of “Legal Goods” (Rechtsgüterschutz): A German Perspective’ in Simester AP, du Bois-Pedain A and Neumann U (eds), Liberal Criminal Theory (Oxford, Hart Publishing, 2018) 107 Hecker DB, Europaisches Stratfrech, 3rd edn (Dordrecht, Springer, 2010) Herlin-Karnell E, ‘Commission v. Council: Some Reflection on Criminal Law in the First Pillar’ (2007) 13 European Public Law 69 ——. The Constitutional Dimension of EU Criminal Law (Oxford, Hart Publishing, 2012) ——. ‘White-Collar Crime and the European Financial Crisis: Getting Tough on EU Market Abuse’ (2012) 37(4) European Law Review 481 ——. ‘All Roads Lead to Rome: The New AFSJ Package and the Trajectory to Europe 2020’ (2014) 1 EuCrim 27 ——. ‘Is Administrative Law Still Relevant? How the Battle of Sanctions has Shaped EU Criminal Law’ in Bergstroom M, Mitsilegas V and Konstatinides T (eds), Research Handbook on EU Criminal Law (Cheltenham, Edward Elgar, 2015) 233 ——. ‘The EU as a Promoter of Preventive Criminal Justice and the Internal Security Context’ (2016) 17(2) European Politics and Society 215 ——. ‘The Domination of Security and the Promise of Justice: On Justification and Proportionality in Europe’s “Area of Freedom, Security and Justice”’ (2017) Transnational Legal Theory 79 ——. The Constitutional Structure of Europe’s Area of ‘Freedom, Security and Justice’ and the Right to Justification (Oxford, Hart Publishing, 2019) Hillebrandt M, Curtin D and Meijer A, ‘Transparency in the EU Council of Ministers: An Institutional Analysis’ (2014) 20(1) European Law Journal 1 Hillyard P and Tombs S, ‘From “Crime” to Social Harm?’ (2007) 48 Law and Social Change 9 Hirsch HJ (ed), Krise des Strafrechts und der Kriminalwissenschaften?Tagungsbeiträge eines Symposiums der Alexander von Humboldt-Stiftung, Bonn-Bad Godesberg, veranstaltet vom 1. bis 5. Oktober 2000 in Bamberg (Berlin, Duncker & Humblot, 2001) Hoefer F, ‘The Nazi Penal System – I’ (1944–45) 35 Journal of Criminal Law & Criminology 385 Hoffe O, ‘Moral Reasons for an Intercultural Criminal Law: A Philosophical Attempt’ (1998) 11(3) Ratio Juris 206

Bibliography  227 Horder J, ‘Bureaucratic “Criminal” Law: Too Much of a Bad Thing?’ (2014) LSE Working Papers 1/2014 Hornle T, ‘Criminalising Behaviour to Protect Human Dignity’ (2012) 6 Criminal Law and Philosophy 307 Horsley T, ‘Subsidiarity and the European Court of Justice: Missing Pieces in the Subsidiarity Jigsaw’ (2012) 50(2) Journal of Common Market Studies 267 Husak D, ‘Justification and the Criminal Liability of Accessories’ (1988) 80 Journal of Criminal Law and Criminology 201. ——. ‘The Criminal Law as a Last Resort’ (2004) 24(2) Oxford Journal of Legal Studies 207 Husak D, Overcriminalisation (Oxford, Oxford University Press, 2008) Jachtenfuchs M and Krisch N, ‘Subsidiarity in Global Governance’ (2016) 79 Law and Contemporary Problems 1 Jareborg N, ‘Criminalisation as a Last Resort (Ultima Ratio)’ (2005) 2 Ohio State Journal of Criminal Law 53. Jehle J-M and Harrendorf S (eds), Defining and Registering Criminal Offences and Measures (Göttingen, Universitätsverlag Göttingen, 2010) Johnson D, ‘Why the Possession of Computer-Generated Child Pornography Can Be Constitutionally Prohibited’ (1994) 4 Albany Law Journal of Science & Technology 311 Kadish H, ‘Why Substantive Criminal Law: A Dialogue’ (1980) 29 Cleveland State Law Review 1 Kahan D and Nussbaum M, ‘Two Conceptions of Emotion in Criminal Law’ (1996) 96 Columbia Law Review 296 Kaiafa-Gbandi M, ‘The Importance of Core Principles of Substantive Criminal Law for a European Criminal Policy Respecting Fundamental Rights and the Rule of Law’ (2011) 1(1) European Criminal Law Review 11 ——. ‘The Commission’s Proposal for a Directive on the Fight against Fraud to the Union’s Financial Interests by Means of Criminal Law (COM (2012) 363 final): An Assessment Based on the Manifesto for a European Criminal Policy’ (2012) 2(3) European Criminal Law Review 319 ——. ‘Approximation of Substantive Criminal Law Provisions in the EU and Fundamental Principles of Criminal Law’ in Weyembergh A and Galli F (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Edition de l’Université de Bruxelles, 2013) 27 Klip A, ‘Definitions of Harmonisation’ in Klip A and van der Wilt H (eds), Harmonisation and Harmonising Measures in Criminal Law (Amsterdam, Koninklijke Nederlandse Akad van Wetenschappen, 2002) 23 ——. ‘EU Criminal Policy’ (2012) 20(1) European Journal of Crime, Criminal Law and Criminal Justice 3 ——. European Criminal Law, 3rd edn (Antwerp, Intersentia, 2016) Kochenov D, de Búrca G and Williams A (eds), Europe’s Justice Deficit (Oxford, Hart Publishing, 2015) Kotiswaran P and Palmer N, ‘Rethinking the “International Law of Crime”: Provocations from Transnational Legal Studies’, King’s College London Dickson Poon School of Law Legal Studies Research Paper Series, 2015-34 Kouroutakis AE, ‘The Italian European Arrest Warrants for the Five Greeks Taking Part in Riots and Their Rejection by the Greek Authorities’ (2016) 7(3) New Journal of European Criminal Law 295 Kumm M, ‘Constitutionalising Subsidiarity in Integrated Markets: The Case of Tobacco Regulation in the European Union’ (2006) 4(12) European Law Journal 503 Labayle H, ‘L’ouverture de la jarre de Pandore : réflexions sur la compétence de la Communauté en matière pénale’ (2006) 3 Cahiers de Droit Européen 382 ——. ‘Le Traité de Lisbonne et l’entraide repressive dans l’Union européenne’ (2007/2008) 2 Revue des Affairs Européens 209 Lacey N, ‘Historicising Criminalisation: Conceptual and Empirical Issues’ (2009) 72(6) Modern Law Review 936

228  Bibliography ——. ‘What Constitute Criminal Law?’ in Duff RA et al (eds), The Constitution of the Criminal Law (Oxford, Oxford University Press, 2013) 13 Lauterwein C, The Limits of Criminal Law: A Comparative Analysis of Approaches to Legal Theorizing (Farnham, Ashgate, 2010) Lemonde M, ‘Justice and the Media’ in Delmas-Marty M and Spencer JR (eds), European Criminal Procedures (Cambridge, Cambridge University Press, 2005) 688 Lenaerts K and van Nuffel P, Constitutional Law of the European Union, 2nd edn (London, Sweet & Maxwell, 2005) Ligeti K, ‘Approximation of Substantive Criminal Law and the Establishment of the European Public Prosecutor’s Office’ in Weyembergh A and Galli F (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Edition de l’Université de Bruxelles, 2013) 73 Lippman M, Contemporary Criminal Law: Concepts, Cases and Controversies (London, SAGE Publishing, 2013) Lobba P, ‘Punishing Denialism beyond Holocaust Denial: EU Framework Decision 2008/913/JHA and Other Expansive Trends’ (2014) 5 New Journal of European Criminal Law 58 Long N, ‘Towards a European Criminal Law Code?’ (2011) 01(49) EipaScope Bulletin 49 ——. ‘Harmonisation of Criminal Law in the EU’, Study for the European ParliamentLIBE Committee, available at: https://www.europarl.europa.eu/RegData/etudes/etudes/ join/2010/419631/IPOL-LIBE_ET%282010%29419631_EN.pdf Lott JR, ‘Do We Punish High Income Criminals Too Heavily?’ (1992) 30 Economic Inquiry 583 Luchtman M and Vervaele JA, ‘Enforcing the Market Abuse Regime: Towards an Integrated Model of Criminal Law and Administrative Law Enforcement in the European Union’ (2014) 5(2) New Journal of European Criminal Law 193 Malamuth N and Huppin M, ‘Drawing the Line on Virtual Child Pornography: Bringing the Law in Line with the Research Evidence’ (2007) NYU Review of Law & Social Change 773 Manacorda S, ‘La consolidation de l’Espace de Liberté Securité et Justice : vers une mise à l’écart du rapprochement pénal?’ (2007) Revue de science criminelle et de droit comparé 890 ——. ‘La mutation “à droit constant” du Troisième Pilier : renforcement et élargissement de la cooperation’ (2008) Revue de science criminelle et de droit comparé 995 ——. ‘Un bilan des dynamiques d’intégration pénale à l’aube du traité de Lisbonne’ (2009) Revue de Science Criminelle, 927 ——. ‘Le programme pour une politique pénale de l’Union entre mythe et réalité’ (2011) 4 Revue de science criminelle 900 ——. ‘L’age de la Maturité : stabilisation et traits conservateurs dans la politique pénale de l’Union Européenne’ (2012) Revue de Science Criminelle 931 Manes V, Il principio di offensività. Canone di politica criminale, criterio ermeneutico, parametro di ragionevolezza (Turin, Giappichelli, 2005) Mégret F, ‘International Criminal Justice: A Critical Research Agenda’ in Schwöbel C (ed), Critical Approaches to International Criminal Law: An Introduction (Abingdon, Routledge, 2014) 26 Mendes de Sousa S, Cooperaçao Juridica Penal no Mercosul: novas possibilidades (Renovar, Rio de Janeiro, 2011) Miettinen S, Criminal Law and Policy in the European Union, 2nd edn (Abingdon, Routledge, 2014) Miglietti M, ‘The First Exercise of Article 83(2) TFEU under Review: An Assessment of the Essential Need of Introducing Criminal Sanctions’ (2014) 5(1) New Journal of European Criminal Law 5 Mikkinen P, ‘“If Taken in Earnest”: Criminal Law Doctrine and the Last Resort’ (2006) 45(5) Howard Journal 521 ——. ‘The “Last Resort”: A Moral and/or Legal Principle?’ (2013) 3(1) Onati Socio-legal Series 21 Mill JS, On Liberty and Other Essays (Oxford, Oxford University Press, 1991 [1859]) Mitsilegas V, EU Criminal Law (Oxford, Hart Publishing, 2009) ——. ‘From Overcriminalisation to Decriminalisation: The Many Faces of Effectiveness in European Criminal Law’ (2014) 5(3) New Journal of European Criminal Law 416

Bibliography  229 ——. ‘Regional Organization and the Suppression of Transnational Crime’ in Boister N and Currie R (eds), Routledge Handbook of Transnational Criminal Law (Abingdon, Routledge, 2015) 57 ——. EU Criminal Law after Lisbon (Oxford, Hart Publishing, 2016) Moccia S, La Perenne emergenza. Tendenze autoritarie nel sistema penale, 2nd edn (Naples, Edizioni Scientifiche Italiane, 2000) Monar J, ‘The Dynamics of Justice and Home Affairs: Laboratories, Driving Factors and Costs’ (2001) 39(4) Journal of Common Market Studies 747 Moohr GS, ‘The Crime of Copyright Infringement: An Inquiry Based on Morality, Harm, and Criminal Theory’, available at: www.law.uh.edu/faculty/gmoohr/Criminal.pdf Moore KE, Stuewig JB and Tangney JP, ‘The Effect of Stigma on Criminal Offenders’ Functioning: A Longitudinal Mediational Model’ (2016) 37(2) Deviant Behavior 196 Moore M, Placing Blame: A Theory of the Criminal Law (Oxford, Oxford University Press, 2010) Muñoz de Morales Romero M, ‘In Pursuit of Basics for a New Principle of Legal Reserve in Supranational Criminal Law’ (2012) 3(2) EU Criminal Law Review 252 Muraszkiewicz J and De Hert P, ‘Gary Becker and the Economics of Trafficking in Human Beings’ (2014) 5(2) New Journal of European Criminal Law 116 Murswiek D, ‘The Stealthy Development of the Treaty on European Union into the Supreme European Constitution’, available at: www.jura.uni-freiburg.de/institute/ioeffr3/forschung/ papers.php Nederveen PJ, ‘Leaking Superpower: WikiLeaks and the Contradictions of Democracy’ (2012) 33(10) Third World Quarterly 1909 Nelles U, ‘Definition of Harmonization’ in Klip A and van der Wilt H (eds), Harmonisation and Harmonising Measures in Criminal Law (Amsterdam, Koninklijke Nederlandse Akad van Wetenschappen, 2002) Neumann U, Liberal Criminal Theory (Oxford, Hart Publishing, 2018) 326. Nieto Martin A, ‘Americanisation or Europeanisation of Corporate Crime?’ in Delmas- Marty M, Pieth M and Sieber U (eds), Les Chemins de l’harmonisation pénale (Paris, Société de Legislation Comparé, 2008) 327 Nilsson HG, ‘How to Combine Minimum Rules with Maximum Legal Certainty?’ (2011) 4 Europarättslig tidskrift 665 ——. ‘Where Should the European Union Go in Developing its Criminal Policy in the Future?’ (2014) 1 EuCrim 19 Öberg J, ‘EU Criminal Law, Democratic Legitimacy and Judicial Review of Union Criminal Law Legislation in the Wake of the Lisbon Treaty’ (2011) 6 Tilburg Law Review 60 ——. ‘Union Regulator Criminal Law Competence after the Lisbon Treaty’ (2011) 19 European Journal of Crime, Criminal Law and Criminal Justice 289 ——. ‘Subsidiarity and EU Procedural Criminal Law’ (2015) 5 European Criminal Law Review 19 Packer H, The Limits of the Criminal Sanctions (Stanford, Stanford University Press, 1968) Parisi N, ‘Chiari e scuri nella direttiva relativa alla lotta contro la frode che lede gli interessi finanziari dell’Unione’ (2017) 9 Giurisprudenza Penale 1 Peers S, ‘The Community’s Criminal Law Competence: The Plot Thickens’ (2008) 33 European Law Review 399 ——. ‘EU Criminal Law and the Treaty of Lisbon’ (2008) 33 European Law Review 507 ——. EU Justice and Home Affairs Law (Oxford, Oxford University Press, 2011) Pelkmans J, ‘The New Approach to Technical Harmonisation and Standardization’ (1987) 25(3) Journal of Common Market Studies 249 Peristeridou C, The Principle of Legality in EU Criminal Law (Antwerp, Intersentia, 2015) Perron W, ‘Perspective of the Harmonization of Criminal Law and Criminal Procedure in the European Union’ in Husabø EJ and Strandbakken A (eds), Harmonisation of Criminal Law in Europe (Antwerp, Intersentia, 2005) 5 Persak N, The Harm Principle and its Continental Counterpart (New York, Springer, 2012) Pettit P, ‘Republican Theory and Criminal Punishment’ (1997) 9 Utilitas 59

230  Bibliography Piccotti L, ‘Il Corpus juris 2000. Profili di diritto penale sostanziale e prospettive di attuazione alla luce del progetto di Costituzione per l’Europa’ in Piccotti L (ed), Il Corpus juris 2000. Nuova formulazione e prospettive di attuazione (Padua, Cedam, 2004) 85 Posner R, ‘An Economic Theory of the Criminal Law’ (1985) 85 Columbia Law Review 1193 Prechal S, EC Directives in EC Law, 2nd edn (Oxford, Oxford University Press, 2005) Rakar I and Ticar B, ‘Normative Principles in the System of Criminal Justice’ (2016) 24 European Journal of Crime, Criminal Law and Criminal Justice 135 Rasmusen E, ‘Stigma and Self-Fulfilling Expectations of Criminality’ (1996) 39 Journal of Law and Economics 519 Renzo M, ‘A Criticism of the International Harm Principle’ (2010) 4(3) Criminal Law and Philosophy 267 Roberts P, ‘Criminal Law Theory and the Limits of Liberalism’ in Simester AP, du Bois-Pedain A and Neumann U (eds), Liberal Criminal Theory (Oxford, Hart Publishing, 2018) 326 Robinson G and McNeil F (eds), Community Punishment, European Perspectives (Abingdon, Routledge, 2016) Robinson P, Criminal Law Defenses (St Paul, MN, West Publishing Co, 1984) Roxin C, Allgemeiner Teil: Grundlagen, Aufbau der Verbrechensehre, vol 1, 4th edn (Munich, CH Beck, 2006) Rubi-Cavagna E, ‘Le domaine et les methods de l’harmonisation autonome’ in Giudicielli-Delage G and Lazerges C (eds), Le Droit Pénal de l’Union Européenne au Lendemain du Traité de Lisbonne (Paris, Société de Législation Comparée, 2012) 141 Ruotolo M, ‘Gli itinerari della giurisprudenza costituzionale in tema di libertà personale’ (2004) 2–3 Questione Giustizia 235 Salazar L, ‘EU’s Criminal Policy and the Possible Contents of the New Multi-annual Program’ (2014) 1 EuCrim 22 Satzger H, International and European Criminal Law (Oxford, Hart Publishing, 2012) Saul B, Defining Terrorism in International Law (Oxford, Oxford University Press, 2008) Saydé A, ‘One Law, Two Competitions: An Inquiry into the Contradiction of Free Movement Law’ (2010–11) 13 Cambridge Yearbook of Legal Studies 365 Schütze R, From Dual to Cooperative Federalism (Oxford, Oxford University Press, 2009) ——. ‘Limits to the Union’s “Internal Market” Competences: Constitutional Comparisons’ in Azoulai L (ed), The Question of Competence in the European Union (Oxford, Oxford University Press, 2014) 215 Scoccia D, ‘In Defence of “Pure” Legal Moralism’ (2013) 7(3) Criminal Law and Philosophy 513 Sicurella R, ‘“Prove tecniche” per una metodologia dell’esercizio delle nuove competenze concorrenti dell’Unione Europea in materia penale’ in Grasso G, Sicurella R and Piccotti L (ed.), L’evoluzione del diritto penale nei settori d’interesse europeo alla luce del Trattato di Lisbona (Milan, Giuffrè, 2011) 896 ——. ‘Setting up a European Criminal Policy for the Protection of EU Financial Interests: Guidelines for a Coherent Definition of the Material Scope of the European Public Prosecutor’s Office’ in Ligeti K (ed), Toward a Prosecutor for the European Union: A Comparative Analysis – Vol 1 (Oxford, Hart Publishing, 2012) 870 ——. ‘EU Competence in Criminal Matters’ in Mitsilegas V, Bergstrom M and Konstantinides T (eds), Research Handbook in EU Criminal Law (Cheltenham, Edward Elgar, 2015) 49 ——. ‘A Blunt Weapon for the EPPO? Taking the Edge off the Proposed PIF Directive’ in Geelhoed  W,  Erkelens L and Meij A (eds), Shifting Perspectives on the European Public Prosecutor’s Office (The Hague, TMC Asser Press, 2018) 99 Sieber U, ‘The Forces behind the Harmonisation of Criminal Law’ in Delmas-Marty M et al, Les chemins de l’harmonisation pénale (Paris, Société de lègislation comparée, 2008) 385 Siems M and Nelemans M, ‘The Reform of the EU Market Abuse Law: Revolution of Evolution?’ (2012) 19 Maastricht Journal of European and Comparative Law 195 Silva Sanchez JM, La expansion del derecho pénal. Aspectos de la politica criminal en las sociedades post-industriales (Madrid, Civitas, 2001)

Bibliography  231 Simester AP and von Hirsch A, Crimes, Harms, and Wrongs: On the Principles of Criminalisation (Oxford, Oxford University Press, 2011) Skinner S, ‘Tainted Law? The Italian Penal Code, Fascism and Democracy’ (2011) 7(4) International Journal of Law in Context 423 Slawomir R, ‘The United Nations Criminal Justice System in the Suppression of Transnational Crime’ in Boister N and Currie R (eds), Routledge Handbook of Transnational Criminal Law (Abingdon, Routledge, 2015) 57 Sotis C, Il diritto senza codice (Milan, Giuffrè, 2007) ——. ‘Il Trattato di Lisbona e le competenze penali dell’Unione Europea’ (2010) 3 Cassazione Penale 326 ——. ‘“Criminaliser sans punir”: Reflexions sur le pouvoir d’incrimination (directe et indirecte) de l’Union européenne prévu par le traité de Lisbonne’ (2011) 4 Revue de Science criminelle et de droit pénal comparé 773 ——. ‘Diritto penale della rete e prospettiva europea: il caso della pedopornografia virtuale’ (2011) 4 Diritto dell’informazione e dell’informatica 561 Spencer JR, ‘Why is the Harmonisation of Penal Law Necessary?’ in Klip A and van der Wilt H (eds), Harmonisation and Harmonising Measures in Criminal Law (Amsterdam, Koninklijke Nederlandse Akad van Wetenschappen, 2002) 43 Stephen JF, Liberty, Equality, Fraternity, White RJ (ed) (Cambridge, Cambridge University Press, 1967 [1873]) Stuckenberg CF, ‘The Constitutional Deficiencies of the German Rechtsgutslehre’ (2013) 3(1) Onati Socio-legal Series 31 Svatikova K, ‘Economic Criteria for Criminalisation: Why Do We Need the Criminal Law?’ (2008) Rotterdam Institute of Law and Economics Working Paper Series 2008/12 Tadic F, ‘How Harmonious Can Harmonization Be? A Theoretical Approach towards Harmonization of (Criminal) Law’ in Klip A and van der Wilt H (eds), Harmonisation and Harmonising Measures in Criminal Law (Amsterdam, Koninklijke Nederlandse Akad van Wetenschappen, 2002) 1 Tadors V, ‘Harm Sovereignty and Prohibition’ (2011) 17 Legal Theory 35 Teichmann D, ‘The Economics of Crime Control’ in Robinson, PH, Garvey S and Kessler Ferzan K (eds), Criminal Law Conversations (Oxford, Oxford University Press, 2015) 87 Torny M, ‘Can Deserts Be Just in an Unjust World?’ in Simester AP, du Bois-Pedain A and Neumann U (eds), Liberal Criminal Theory (Oxford, Hart Publishing, 2018) 142 Trauner F and Ripoll Serventi A, Policy Change in the Area of Freedom, Security and Justice: How EU Institutions Matter (Abingdon, Routledge, 2015) Tricot J, ‘L’harmonisation pénale accessoire : question(s) de method : observations sur l’art et la manière de légiférer pénalement selon l’Union Européenne’ in Giudicielli-Delage G and Lagerzer C (eds), Le Droit Pénal de l’Union Européenne au Lendemain du Traité de Lisbonne (Paris, Société de Législation Comparée, 2012) 185 Tridimas T, The General Principles of EU Law, 2nd edn (Oxford, Oxford University Press, 2006) Triffterer O, ‘Preamble’ in Triffterer O (ed), Commentary on the Rome Statute of the International Criminal Court, 2nd edn (Oxford, Hart Publishing, 2008) 1 Turner JI, ‘The Expressive Dimension of EU Criminal Law’ (2012) 60 American Journal of Comparative Law 555 Tyler T, ‘Procedural Justice, Legitimacy, and the Rule of Law’ (2003) 30 Crime and Justice 283 Van Leeuwen B, ‘Euthanasia and the Ethics of Free Movement Law: The Principle of Recognition in the Internal Market’ (2018) 19(6) German Law Journal 1417 Vander Beken T, ‘Freedom, Security and Justice in the European Union: A Plea for Alternative Views on Harmonisation’ in Klip A and van der Wilt H (eds), Harmonisation and Harmonising Measures in Criminal Law (Amsterdam, Koninklijke Nederlandse Akad van Wetenschappen, 2002) 95 Vassalli G, La potestà punitiva statuale (Turin, UTET, 1942) Vavoula N, ‘Prevention, Surveillance, and the Transformation of Citizenship in the “Security Union”: The Case of Foreign Terrorist Fighters’ in Sieber U et al (eds), Alternative, Informal,

232  Bibliography and Transitional Types of Criminal Justice and the Legitimacy of New Sanction Models in the Global Risk Society (Berlin, Max Planck Institute for Foreign and International Criminal Law in collaboration with Duncker & Humblot, 2018) 307 Vermeulen G, ‘Where Do We Currently Stand with Harmonisation in Europe?’ in Klip A and van der Wilt H (eds), Harmonisation and Harmonising Measures in Criminal Law (Amsterdam, Koninklijke Nederlandse Akad van Wetenschappen, 2002) 73 Vervaele JAE, Fraud against the Community: The Need for European Fraud Legislation (Deventer, Kluwer Law and Taxation Publishers, 1992) ——. ‘Harmonised Union Policies and the Harmonisation of Substantive Criminal Law’ in Weyembergh A and Galli F (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Edition de l’Université de Bruxelles, 2013) 43 Victor JL, ‘National Parliaments and the Principle of Subsidiarity: Legal Options and Practical Limits’ (2008) 4(3) European Constitutional Law Review 429 Vogel J, ‘Why is Harmonisation of Penal Law Necessary? A Comment’ in Klip A and van der Wilt H (eds), Harmonisation and Harmonising Measures in Criminal Law (Amsterdam, Koninklijke Nederlandse Akad. van Wetenschappen, 2002) 55 Vogliotti M, Tra Fatto e Diritto. Oltre la modernità giuridica (Turin, Giappichelli, 2008) Wade M, ‘Developing a Criminal Justice Area in the European Union’, European Parliament Study 2014, Study PE 493.043, available at: www.europarl.europa.eu/RegData/etudes/etudes/ join/2014/493043/IPOL-LIBE_ET(2014)493043_EN.pdf Walker N, In Search of the Area of Freedom, Security and Justice: A Constitutional Odyssey’ in Walker N (ed), Europe’s Area of Freedom, Security and Justice (Oxford, Oxford University Press, 2004) 3 Wasmeier M and Thwaites N, ‘The “Battle of the Pillars”: Does European Community Have the Power to Approximate National Criminal Laws?’ (2004) 29(5) European Law Review 613 Waxman M C, ‘Administrative Detention of Terrorists: Why Detain, and Detain Whom?’ (2009) 3(1) Journal of National Security Law and Policy 1 Weatherill S, Law and Integration in the European Union (Oxford, Clarendon Press, 1995) ——. ‘Why Harmonise?’ in Tridimas T and Nebbia P (eds), European Union Law for the TwentyFirst Century: Rethinking the New Legal Order (Oxford, Hart Publishing, 2004) 11 ——. ‘The Limits of Legislative Harmonisation Ten Years after Tobacco Advertising: How the Court’s Case Law Has Become a “Drafting Guide”’ (2011) 12(3) German Law Journal 827 Weiler J ‘European Models: Polity, People and System’ in Craig P and Harlow C (eds), Lawmaking in the European Union (Boston, Kluwer, 1998) Wendt R, ‘The Principle of ‘Ultima Ratio” and/or the Principle of Proportionality’ (2013) 3(1) Onati Socio-legal Series 81 Werle G and Jessberger F, Principles of International Criminal Law, 3rd edn (Oxford, Oxford University Press, 2014) Weyembergh A, L’harmonisation des législations : condition de l’espace pénal européen et révélateur de ses tensions (Brussels, Editions de l’Université de Bruxelles, 2004) ——. ‘The Functions of Approximation of Penal Legislation within the European Union’ (2005) 12 Maastricht Journal of European and Comparative Law 149 ——. ‘Introduction’ in A Weyembergh and F Galli (eds), Approximation of Substantive Criminal Law in the EU: The Way Forward (Brussels, Edition de l’Université de Bruxelles, 2013) 9 Weyembergh A and Wieczorek I, ‘Is There an EU Criminal Policy?’ in Colson R and Field S (eds), EU Criminal Justice and the Challenges of Legal Diversity (Cambridge, Cambridge University Press, 2016) 29 White S, ‘Harmonisation of Criminal Law under the First Pillar’ (2006) 31 European Law Review 81 Wieczorek I, The EPPO Draft Regulation Passes the First Subsidiarity Test: An Analysis and Interpretation of the Commission’s Hasty Approach to National Parliaments’ Subsidiarity Arguments’ (2015) 16(5) German Law Journal 1247

Bibliography  233 ——. ‘The Principle of Subsidiarity in EU Criminal Law’ in Brière C and Weyembergh A (eds), The Needed Balances in EU Criminal Law: Past, Present and Future (Oxford, Hart Publishing, 2017) 71 ——. ‘EU Constitutional Limits to the Europeanization of Punishment: A Case Study on Offenders’ Rehabilitation’ (2018) 25(6) Maastricht Journal of European and Comparative Law 1 Williams A, EU Human Rights Policies: A Study in Irony (Oxford, Oxford University Press, 2004) Williams G, ‘The Definition of Crime’ (1955) 8(1) Current Legal Problems 107 Wilsher D, ‘The Administrative Detention of Non-nationals Pursuant to Immigration Control: International and Constitutional Law Perspectives’ (2004) 53(4) International and Comparative Law Quarterly 897 Wilson JQ and Herrnstein RJ, Crime and Human Nature (New York, Simon & Schuster, 1985) Yeung K, Securing Compliance (Oxford, Hart Publishing, 2004) Zedner L and Ashworth A, ‘Defending the Criminal Law: Reflections on the Changing Character of Crime, Procedure, and Sanctions’ (2008) 2 Journal of Criminal Law and Philosophy 21 Zumbansen P, Theorising as Activity: Transnational Legal Theory in Context’ in McCrudden C, Baxi U and Paliwala A (eds), Law’s Ethical, Global and Theoretical Contexts: Essays in Honour of William Twining (Cambridge, Cambridge University Press, 2015) 280

234

Index abuse of power, 29, 203–4 adultery, 80 American Law Institute, 37 Aquinas, T, 33 arms trafficking, 45–6, 96, 99, 108 Arroyo Zapatero, L, 117, 118 ASEAN, 1 Ashworth, A, 26, 36, 37, 219 Asp, P, 50 assimilation principle, 124 asylum seekers, 24 Ayres, I, 34 Baker, D, 28 Banach Gutierrez, JB, 153n65, 157–8 Baratta, A, 36, 219 Becker, GS, 34, 38 Belgium: criminal law and constitution, 41 Dutroux scandal, 170 fraud offence, 203 benchmarks: policy documents, 10–11, 140, 141–67 Bentham, J, 33, 34, 35 Berlusconi, 59 Bernardi, A, 119 Birnbaum, JMF, 32 Bowels, R, 34, 38 Braithwaite, J, 34 Bricola, 32, 41n139, 212 Brown, D, 37 canon law, 1 certainty see legal certainty child pornography, 48, 49, 56, 60–1, 220 citizenship: expulsion of foreign EU citizens, 115–16 collective action rationale, 70–1, 117, 139 colonial criminal law, 2, 79 competences see also specific crimes classification, 96–8 decriminalisation, 220 development, 95–6, 213–14 draft Constitutional Treaty, 144

EU limits, 44 functional criminalisation see functional criminalisation Lisbon Treaty, 45–7, 106–19 protection of financial interests, 109, 193 purposive drafting, 9, 213 racism and xenophobia, 173 relevant crimes, 45–6 securitised criminalisation, 95–120 Lisbon Treaty, 106–19 Maastricht/Amsterdam Treaties, 98–106 utilitarianism, 98–106 tensions, 218–19 computers see cybercrime conflicts of interests, 203 conflicts of jurisdiction, 88 cooperation: 2012 Parliament Resolution, 164 abuse of power and, 204 collective action rationale, 117, 139 competence, 102, 181 double criminality and, 87, 88, 90–1, 178 Framework Decision on Racism and Xenophobia, 177–9, 216 functional criminalisation and, 138–9 Hague Programme, 144, 181 harmonisation rationale, 87–8, 89, 90, 118, 215 Market Abuse Directive, 187–8 policy documents, 149 securitised criminalisation and, 100–1, 116, 117 Stockholm Programme, 145, 216 Tampere Programme, 143, 181 Vienna Action Plan, 143, 181 Corpus Juris, 5 corruption: cross-border criterion, 116 definition, 204 EU competence, 46, 99 PIF Convention, 202 sanctions, 61 securitised criminalisation, 96 Tampere Programme, 143

236  Index Council of Europe, 1, 176 counterfeiting, 46, 48, 57, 96, 143, 158 Court of Justice of the European Union: development of harmonisation competences, 95 functional criminalisation early case law, 123–6 environmental crimes, 126–35 integrated approach, 126–35 fundamental rights and, 11 Lisbon Treaty and, 106 criminal law: consequentialism, 22–8 definition of crimes, 22–3 dignity and, 25–7 European Union see EU criminal law higher procedural safeguards, 27–8 historical debate, 28–30 last resort, 41 law and economics, 7, 26, 27, 34–5, 38, 212 legitimacy, 13, 21–43 constitutional character, 40–3 harm principle, 30–1 joint deontological-utilitarianism, 35–40 liberal deontological approach, 29–33, 81, 212 liberal utilitarianism, 33–5, 212 relevance, 212–13 liberty and, 23–5 need to legitimate choices, 28–30 regulatory criminal law, 36, 39, 121, 128 social disapproval, 186 stigma, 25–7 customs rules, 121, 159 cybercrime: assessment, 220 definition, 48, 114 enforcement of Directive, 56 EU competence, 46 European Convention, 176 sanctions, 61 securitised criminalisation, 96 Tampere Programme, 143 Cyprus: hate speech and, 179 data protection, 159, 172, 219 De Búrca, G, 71 De Hert, P, 91, 170, 171–2 Delmas-Marty, M, 7, 36, 77n6, 219 democracy: EU criminal law and, 112, 136, 213 EU value, 19, 62, 180

Framework Decision on Racism, 174–7 international criminal law and, 2, 3, 4, 14 liberal democracies, 43 subsidiarity and, 71, 214 denialism, 49, 174, 175, 176, 177, 179 Denmark: opt-outs, 14 deontological approach: criminal law, 29–33, 81, 212 EU criminal law, 84–5 EU legislation and, 169 EU values, 62–4 functional criminalisation and, 125, 128–31 normative legitimacy, 2–3, 5 PIF Directive, 196, 199–208 policy documents, 151–4 2012 EU Parliament Resolution, 163 growing importance, 165 Hague Programme, 144 Stockholm Programme, 146 retrospective proportionality, 68–9 seriousness criterion, 113–16 subsidiarity and, 74 utilitarian/deontological approach criminal law, 35–40 EU proportionality, 64–9 deterrence: 2011 Commission Communication, 160 criminal law rationale, 186 international criminal law and, 3 Market Abuse Directive, 186–7, 208 PIF Directive, 195–6, 206, 208 Devlin, P, 31 digital crimes see cybercrime dignity: constitutions, 41 criminal law and, 25–7, 81 EU criminal law and, 215 EU value, 62, 63, 212–13 direct effect: criminal legislation: case law, 47 enforcement obligations and, 53–5 MAS and MB, 54–5 prohibitions, 51–3 statutes of limitation and, 53–5 substitution direct effect, 52–3 Taricco, 53–5 double criminality, 87, 88, 90–1, 178 Douglas-Scott, S, 105 draft Constitutional Treaty, 128–9, 144, 162 Dripps, D, 37 drug trafficking, 45–6, 48, 72, 84, 96, 99, 143

Index  237 Du Bois-Pedain, A, 42 Duff, RA, 28, 31 easing transactions rationale, 85–91 embezzlement, 203 enforcement obligations: direct effect see direct effect infringement proceedings, 51, 55, 56, 60 sanctions see sanctions state liability, 51 Enlightenment, 22, 29, 212 environmental crimes: 2011 Commission Communication, 159 cross-border criterion, 116 events responsiveness, 170 functional criminalisation, 123, 126–35 Environmental Crime, 127–32, 149, 161, 166 Ship Source Pollution, 133–5, 149 legislative rationale, 171 Tampere Programme, 143 equality: EU value, 12, 42, 62 justice and, 86, 106 Estonia: hate speech and, 179 estoppel, 52 EU: competences see competences criminal law see EU criminal law direct legal effect see direct effect directives: member states’ discretion, 48–51 effectiveness of law, 134 flags, 92 harmonisation see harmonisation identity, 17–20, 180 legitimacy crisis, 14–15, 162 values and principles see values and principles EU Charter of Fundamental Rights: binding nature, 20 EU criminal law legitimacy and, 11 proportionality, 67–8, 69, 157 EU criminal law see also harmonisation; specific crimes benchmarks, 10–11, 140, 141–67 broader outreach, 15–20 case studies, 168–210 competences see competences consistence with EU legal order, 64, 218 declarations, 9–10 development, 3 direct effect see direct effect

emerging debate, 4–5 enforcement obligations, 51–3 EU constitutional identity and, 17–20 event-driven legislation, 170–1 harmonisation process see harmonisation impact of policy choices, 56–8 jurisdiction, 3–4 legality, 11–12 legitimacy debate constitutional values, 213–14 dangers of non-legitimate expansion, 218–20 doctrinal debate, 211–12 relevance, 45–62, 212–13 members’ discretion, 48–51 non-legitimate expansion: dangers, 218–20 normative legitimacy, 5 patchwork structure, 169–72 practical importance of theory, 13–15 repressive potential, 58–62 research methodology, 5–13 analytical framework, 7 book structure, 13 case studies, 10 constitutional values, 7–8, 10–13 practice, 8–10 questions, 6, 214–15 sources, 11–12 symbolic law, 180–1, 191–2, 197, 215–18 values and see values and principles Eurojust, 16, 105 European Anti-Fraud Office (OLAF), 193, 199 European arrest warrants: better functioning, 215 CJEU case law, 14 double criminality and, 88, 90–1, 178 Expo Exhibition protests (2015), 90 legality principle, 88 PIF crimes and, 200, 206 thresholds, 200, 206 European Convention on Human Rights (ECHR), 55, 79 European Criminal Policy Initiative (ECPI), 5, 63, 64 European Public Prosecutor Office (EPPO), 5, 14, 16, 46, 87, 111–12, 193, 199, 201–2 European Year against Racism, 172 Europol, 16, 105 favouritism, 203–4 Feinberg, J, 26, 30–1, 212

238  Index Fichera, M, 102, 104 financial interests see PIF Directive; protection of financial interests fisheries policy, 159 flag burning, 92 Fletcher, M, 102, 104, 105 Flore, D, 100 forum shopping, 156, 195, 197 France: Constitutional Treaty and, 162 criminal law, 219 Douvres case, 170 illegal immigration, 170 PIF Directive and, 204 Prestige oil spill, 170 fraud see also protection of financial interests 2011 Commission Communication, 158 definition, 48, 204 EU competence Lisbon Treaty, 109, 110 origins, 99 functional criminalisation, 122–3 early case law, 123–4 PIF Directive see PIF Directive VAT fraud, 53–5, 200–2, 204, 205, 207–8 free movement: harmonisation rationale, 46–7, 60, 83, 86, 88–9, 94, 103–5, 153, 156, 216 Stockholm Programme, 107, 146 Tampere Programme, 143 Vienna Action Plan, 142 functional criminalisation: 20th century case law, 123–6 21st century case law deontological approach, 128–31 Environmental Crime, 126–32, 149, 161, 166 integrated approach, 126–35 Ship Source Pollution, 127, 149 assessment, 217 assimilation principle, 124 classification, 96–8 collective action rationale, 117, 139 competences, 121–40 cooperation and, 138–9 dissuasive measures, 124–5, 135 effectiveness, 124–5, 128, 132, 134, 135–6, 138, 158, 164, 215–16, 217 Lisbon Treaty, 135–9 Article 83(2), 97, 135–9, 150 EU values, 136–8

norms, 166 subsidiarity and, 138–9 supranationalisation, 137 meaning, 96–7 necessity test, 132, 157 proportionate measures, 124–5, 157–8 regulatory rationale, 120, 128, 132, 136 scope, 122–3 securitised criminalisation and, 96–8, 120 utilitarianism, 121–40 fundamental rights see also specific rights 2012 EU Parliament Resolution, 163 Amsterdam Treaty, 19 criminal law and, 23–5 Declaration on European Identity, 19 early EU treaties and, 18 EU criminal law legitimacy, 5, 11, 17, 214, 218 EU hypocrisy, 20 EU respect for, 62, 63 European common heritage, 20 harmonisation, 79, 81 international criminal law and, 2–3 Lisbon Treaty, 20 proportionality see proportionality Stockholm Programme, 144 Germany: criminal law, 40, 41, 213 denialism, 176–7 dignity, 25 incest, 42, 213 Lisbon case, 14 Nazi criminal law, 16 PIF Directive and, 204 proportionality, 64 Rechtsgut, 31–2, 36, 42, 213 Gindre, E, 136n76, 138–9 Giuffrida, F, 204 globalisation, 1 Greece: criminal law, 215 Greek Maize, 123–5 Hague Programme, 100, 143–4, 165, 181 Harcourt, BF, 31 Harding, C, 153n65, 157–8 harmonisation see also specific subjects coherence rationale, 85–91, 163 competences see competences

Index  239 concept, 76–8 cooperation rationale see cooperation cross-border criterion, 116 deontological EU criminal law, 84–5 dirty rationales, 84, 170 double criminality and, 87 easing transactions rationale, 85–91 EU theories, 7 free movement rationale, 86, 89, 103–5 functional criminalisation see functional criminalisation goals, 79–80 internal market and, 82–4 legal certainty rationale, 85–91 legal integration strategy, 82–4 neutral legal process, 78–82 normative v instrumental types, 78–80 rationales, 83–4 legitimacy and, 80–2 regulatory rationale, 92–3, 120, 149, 215 market abuse, 186–9 market abuse empirical data, 188–9 securitised criminalisation see securitised criminalisation socialising rationale, 91–2 utilitarian rationales, 85–93, 98–106, 116–17 Hart, HLA, 34, 35 hate speech, 17, 80, 92, 173, 175, 176, 177, 179, 216 Herlin-Karnell, E, 105, 112, 160 Holocaust denial, 49, 174, 175, 176, 177 homosexuality, 22, 31 Horder, J, 23, 37 human trafficking: assessment, 220 definition, 48 enforcement of Directive, 56 EU competence, 45–6 Lisbon Treaty, 108, 109, 112 origins, 99 events responsiveness, 170 harmonisation rationale, 84 sanctions, 60–1 securitised criminalisation, 96, 97 Tampere Programme, 143 Husak, D, 28, 36, 219 immigration: detention, 24 EU policy, 97 expulsion of foreign EU citizens, 115–16 imprisonment see prison

incest, 42, 213 indirect effect: prohibitions, 51–3 insider trading, 22–3, 50, 181, 184–5 intellectual property, 192 internal market see also free movement 2011 Commission Communication, 159 harmonisation and, 82–4, 109, 219 main features, 134 market abuse and, 183–4 International Criminal Court, 1, 3 international criminal law: legitimacy, 1–4 Iontcheva, J, 112 Ireland: opt-outs, 14 Israel: criminal law and liberty, 25 Italy: criminal law, 32, 41, 42, 219 Expo Exhibition protests (2015), 89–90, 215 false accounting, 59–60 fraud offence, 203 liberty, 63 moral liberty, 25 PIF Directive and, 204 Rechtsgut theory, 7–8, 41 VAT fraud, 53–5 Jewish Moore communities, 1 Jong, Cornelis de, 14–15, 161–2 justice: concept, 105, 106 securitised criminalisation and, 105–6 Kadish, S, 17 Kaiafas-Gbandis, M, 63, 91, 206 Kant, I, 30, 33, 35 Khommeini, A, 33 Klip, A, 7, 50, 156 Kostiswaran, P, 2 Kumm, M, 70, 71 law and economics, 7, 26, 27, 34–5, 38, 212 legal certainty, 37, 50, 59, 79–80, 85–91, 105, 201, 213 legal moralism, 31, 39, 66 legality: EU principle, 52, 55 European arrest warrants, 88 statutes of limitation and, 53–5 liberalism: criminal law legitimacy deontological liberalism, 29–33 joint utilitarian-deontologicalism, 35–40

240  Index ongoing debate, 211–12 utilitarian liberalism, 29, 33–5 Enlightenment, 22, 29, 212 EU criminal law, 214 harm principle, 30–1 values, 62–3 liberty: constitutions, 41 criminal law and, 23–5, 81 EU criminal law and, 58, 212–13, 215 EU value, 62, 63 Framework Decision on Racism and Xenophobia, 174–7 limitation statutes, 53–5 Lippman, M, 22 Listz, Franz von, 34 Lithuania: criminal law, 41 Manacorda, S, 160, 168n1 Market Abuse Directive: assessment, 220 choice of case study, 10, 172 cooperation rationale, 187–8 definition of market abuse, 48, 181 deterrence rationale, 186–7, 208 effectiveness rationale, 186–8 enforcement, 57 Framework Decision, 209–10 Impact Assessment, 188 legislative rationale, 171 legitimacy, 181–92 negotiations, 15 normative rationale, 189–91 offences, 184–5 origin of legislation, 171 origins, 183–4 regulatory rationale, 186–9 weakness of empirical data, 188–9 sanctions, 61 seriousness threshold, 189–91 symbolic law, 191–2, 217 utilitarianism, 191–2 market manipulation, 50, 181, 184–5, 189 see also Market Abuse Directive Marx, K, 33 MERCOSUR, 1 methodology: EU criminal law legitimacy, 5–13 Miglietti, M, 189 Mikkinen, P, 25 Mill, JS, 30, 212 misappropriation, 203, 204

Mitsilegas, V, 47, 95–6, 113, 217 Molan, M, 22 money laundering: debate, 181–2 definition, 48, 168 Directives, 56, 126, 182–3 EU competence, 46, 98, 108, 182 PIF Directive, 192, 202–3 sanctions, 61, 181–2, 206 securitised criminalisation, 96 Tampere Programme, 143 moral agency, 26 moralism: legal moralism, 31, 39, 66 Muñoz de Morales Romero, M, 4, 117, 118 mutual recognition see cooperation mutual trust see cooperation nationality discrimination, 82 ne bis in idem, 185 Nelles, U, 79–80 Netherlands: Constitutional Treaty and, 162 prison escapes, 91 Nilsson, HG, 50 Nuremberg Tribunal (1945), 174 Öberg, J, 4 OLAF (European Anti-Fraud Office), 193, 199 organised crime, 46, 48, 49, 84, 96, 99, 108, 114, 138, 168 Paker, H, 34 Palmer, S, 2 paternalism, 32, 66 Peristeridou, C, 4 Persak, N, 66 PIF Directive: assessment, 204–5, 208–10, 220 choice of case study, 10, 172 Commission preparatory documents, 194–7 Commission proposal, 155 competences, 103, 193, 195 complementary focus, 199–202 definitions of crimes, 199–208 deontological legitimacy, 196, 199–208 deterrence rationale, 195–6, 206, 208 empirical evidence, 195 Explanatory Memorandum, 194–6, 206 Impact Assessment, 194, 196, 203 implementation deadline, 56 integrated legitimacy, 192–210 preparatory documents, 194–7

Index  241 Italy and, 54, 55 limited harmonising impact, 202–4 origins, 192–4 Preamble, 197–8, 199–200, 204, 206, 207 regulatory rationale, 208 sanctions, 196, 205–8 minimums, 207, 208 proportionality, 206, 207 socialising rationale, 198 solidarity, 196 symbolic law, 197, 217–18 utilitarian rationales, 194–8, 204–8 VAT fraud, 200–2, 204, 205, 207–8 Plato, 33 policy documents: 2009 Council, 151–4 2011 Commission Communication approaches, 154–61 integration, 158–61 selection of policy areas, 158–60 utilitarianism, 155–8, 159, 160–1 2012 Parliament Resolution, 161–4 2014 Guidelines, 147–8 core discussion, 148–64 deontological approach development, 165 harm-based, 151–4 EU development, 141–67 Treaty norms, 166 utilitarianism to integration, 164–6 Hague Programme, 143–4, 165 PIF Directive, 194–7 post-Lisbon integration, 144–8 2011 Commission Communication, 154–61 deontology, 151–4 Treaty norms, 166 pre-Lisbon utilitarianism, 142–4 Stockholm Programme, 144–7, 165, 216, 217 Tampere Programme, 143, 165 Vienna Action Plan, 142–3, 181 Portugal: criminal law, 41 Posner, R, 34, 38 prison: criminal law and, 24–5, 212 drug trafficking and, 72, 73 escapes, 91 EU criminal law and, 58–62 market abuse and, 186–7 PIF Directive and, 206, 207 prisoners’ transfers, 4 violence, 25

professional secrecy, 203 property rights, 17, 24, 192 proportionality: 2012 EU Parliament Resolution, 163 deontological and utilitarian considerations, 64–9 deontological approach, 68–9 EU Charter of Fundamental Rights, 67–8, 69, 157 EU criminal law legitimacy and, 8, 11 functional criminalisation, 124–5 incest and, 42 PIF crimes, 206, 207 policy documents, 157 prospective proportionality, 65–8 retrospective proportionality, 65, 68–9 ultima ratio principle, 66, 157–8 prostitution, 31 protection of financial interests: 2011 Commission Communication, 158 competences, 109, 193, 195 functional criminalisation, 96–8, 122–3 legislative rationale, 171 PIF Directive see PIF Directive public health policies, 219 public tendering, 203, 204 racism and xenophobia: Corfu Council (1994), 173 definition, 48 denialism, 49, 174, 175–7, 179 EU competence, 173 Framework Decision, 10, 101 assessment, 209–10, 216 case study, 172–81 choice of case study, 172 cooperation rationale, 177–9, 216 origins, 173–4 social engineering, 179–80 symbolic legislation, 180–1 utilitarianism, 180–1 values, 174–7 harmonisation rationale, 92 hate speech, 17, 80, 92, 173, 175, 176, 177, 179, 216 historical context, 172–3 Joint Action, 101, 173, 178 online, 177–8 Tampere Programme, 173 rape, 23, 115 Rawls, J, 30

242  Index Rechtsgut theory, 7–8, 31–2, 36, 41–2, 130, 137, 156, 190, 212, 213 Reding, V, 150 road transport, 159 Roberts, P, 29, 33, 212 Roxin, C, 32, 176, 212 rule of law: EU value, 19, 62, 214 Framework Decision on Racism, 174–7 justice and, 106 Stockholm Programme, 144 safe havens, 85, 156 sanctions see also prison alternatives, 59 effectiveness, 73 money laundering, 61, 181–2, 206 penal v administrative, 40, 60, 160 penalty thresholds, 60–1 PIF Directive, 196, 205–8 potential repression, 58–62 retrospective proportionality, 68–9 SARPCCO, 1 securitised criminalisation: AFSJ, 101–107, 117 assessment, 217 classification, 96–8 competences, 95–120 cooperation rationale, 100–1, 116, 117 free movement rationale, 103–5 functional criminalisation and, 96–8, 120 justice rationale, 105–6 Lisbon Treaty, 106–19 Article 83(1), 97, 113–19 cross-border criterion, 113–14, 116–17 integrated legitimacy, 11–319 norms, 166 seriousness criterion, 113–16 Maastricht/Amsterdam Treaties, 98–106 utilitarianism, 98–106 security: Framework Decision on Racism, 174–7 seriousness criterion, 113–16, 159, 189–91 ship pollution, 133–5 Sieber, U, 7, 79 Silva Sanchez, JM, 36, 219 Simester, AP, 26, 38–9, 212 Slovenia: hate speech and, 179 social disapproval, 160, 186

social engineering, 33, 81, 89, 103, 146, 153, 156, 179–80, 216 social security, 82 socialising rationale, 80, 84, 91–2, 94, 101, 117, 172, 174, 179, 180, 198, 209 Spain: criminal law, 40, 213, 219 Prestige oil spill, 170 state liability, 51 statutes of limitation, 53–5 Stockholm Programme, 106, 107, 144–7, 165, 216, 217 subsidiarity: 2012 EU Parliament Resolution, 163 efficiency criterion, 70–4 EU criminal law legitimacy and, 8, 11, 214 tensions, 218 utilitarianism, 64, 69–74 functional criminalisation, 138–9 subsidies, 121, 122 Svatikova, K, 34, 38 Sweden: EU presidency, 151 symbolic criminal law, 40, 73, 75, 180–1, 191–2, 197, 215–18 Tampere Programme, 9, 100, 102–3, 105, 106, 107, 143, 165, 173, 181 terrorism: 9/11 effect, 170 cross-border criterion, 116 definition, 48 Directive, 56, 220 EU competence, 45–6, 99 event driven legislation, 170, 171 harmonisation rationale, 84 legislative rationale, 171 sanctions, 61 securitised criminalisation, 96, 104 theocracy, 80 tolerance, 62, 80, 175, 180 tourism, 83 transparency, 13–14 Transparency International, 1 Tridimas, T, 65–6 Turner, JI, 178–9 ultima ratio principle, 66, 157–8, 160 United Kingdom: animal welfare, 37 EU criminal law and, 219

Index  243 opt-outs, 14 prison escapes, 91 United Nations: Human Rights Committee: denialism and, 176 transnational crime and, 100 United States: 9/11 effect, 170 criminal law, 84, 219 insider trading, 22–3 legal moralism, 31 Native American jurisdictions, 1 utilitarianism: criminal law legitimacy, 29, 33–40, 81, 212 cross-border criterion, 116–17 EU legislation and, 169 Framework Decision on Racism, 180–1 functional criminalisation, 121–40, 138–9 harmonisation rationales, 85–93 instrumental harmonisation, 81, 85–91 Market Abuse Directive and, 191–2 PIF Directive, 194–8, 204–8 policy documents 2011 Commission Communication, 155–8, 159, 160–1 evolution, 164–6 pre-Lisbon, 142–4 regulatory rationale, 92–3, 121 securitised criminalisation, 98–106 socialising rationale, 91–2 subsidiarity, 64, 69–74, 138–9 utilitarian/deontological approach criminal law, 35–40 EU proportionality, 64–9

Amsterdam Treaty, 19 deontological approach foundation, 62–4, 81 EU criminal law and, 7–8, 10–13, 44–5, 62–74 harmonisation rationale, 84–5 integrated approach, 212–13 legitimacy, 213–14 EU identity, 17–20, 180 EU values, 19, 62, 136–8 Framework Decision on Racism, 174–7 functional criminalisation, 136–8 harmonisation process and, 78–82 liberal values, 62–3 Market Abuse Directive, 189–91 PIF Directive, 199–208 policy documents, 157 sanctions, 19 Stockholm Programme, 144–5, 217 Van der Wilt, H, 7 Vander Baken, T, 105–6 VAT fraud, 53–5, 200–2, 204, 205, 207–8 Vervaele, J, 121, 149, 171 victims’ rights, 3, 175–6, 218 Vienna Action Plan, 142–3, 181 Von Hirsch, A, 26, 38–9, 212 WACI, 1 Weatherill, S, 170 Weyembergh, A, 7, 170 Wikileaks, 1 Williams, G, 22, 23 xenophobia see racism and xenophobia

values and principles see also specific values

Zumbansen, P, 5–6

244