Research Handbook on the EU’s Common Foreign and Security Policy 1785364073, 9781785364075

Table of contents :
Front Matter
Copyright
Contents
Contributors
Acknowledgements
Introduction
PART A THE LAW OF CFSP/CSDP
1. The position of CFSP/CSDP in the EU’s constitutional architecture
2. The institutional framework, legal instruments and decision-making procedures
3. Representing the EU in the area of CFSP: legal and political dynamics
4. ‘The Good, the Bad and the Ugly’: three levels of judicial control over the CFSP
PART B THE PRACTICE OF CSDP
5. Civilian CSDP missions: ‘the good, the bad and the ugly’
6. Military CSDP operations: strategy, financing, effectiveness
7. Responsibility and liability for CSDP operations
8. Capabilities and CSDP: resourcing political will or paper armies
9. Non-proliferation of weapons of mass destruction
PART C THE NEXUS BETWEEN CFSP/CSDP AND OTHER EXTERNAL POLICIES
10. The law and practice of EU sanctions
11. The nexus between the CCP and the CFSP: achieving foreign policy goals through trade restrictions and market access
12. EU development cooperation and the CFSP: mutual encroachment?
13. The CFSP–humanitarian aid nexus
14. The nexus between CFSP/CSDP and the Area of Freedom, Security and Justice
15. European Neighbourhood Policy: CFSP in disguise
16. The nexus between Common Foreign and Security Policy and energy policy
17. The European Union as a cybersecurity actor
18. EU external health security policy and law
PART D REFLECTION ON PERSPECTIVES
19. The European Union as post-national realist power
20. Too much, too little, too late? Reflections on law and ethics in the EU’s foreign policy
21. The European Union’s foreign policies: an external examination of the capabilities–expectations gap
Conclusions
Index

Citation preview

RESEARCH HANDBOOK ON THE EU’S COMMON FOREIGN AND SECURITY POLICY

RESEARCH HANDBOOKS IN EUROPEAN LAW This important series presents a comprehensive analysis of the latest thinking, research and practice across the field of European Law. Organised by theme, the series provides detailed coverage of major topics whilst also creating a focus on emerging areas deserving special attention. Each volume is edited by leading experts and includes specially commissioned chapters from distinguished academics as well as perspectives from practice, providing a rigorous and structured analysis of the area in question. With an international outlook, focus on current issues, and a substantive analysis of the law, these Research Handbooks are intended to contribute to current debate as well as providing authoritative and informative coverage. Forming a definitive reference work, each Research Handbook will be essential reading for both scholars in European law as well as for practitioners and policymakers who wish to engage with the latest thinking and ongoing debates in the field. Titles in this series include: Research Handbook on EU Agriculture Law Edited by Joseph A. McMahon and Michael Cardwell Research Handbook on EU Criminal Law Edited by Valsamis Mitsilegas, Maria Bergström and Theodore Konstadinides Research Handbook on EU Public Procurement Law Edited by Christopher Bovis Research Handbook on EU Consumer and Contract Law Edited by Christian Twigg-Flesner Research Handbook on EU Institutional Law Edited by Adam Łazowski and Steven Blockmans Research Handbook on EU Labour Law Edited by Alan Bogg, Cathryn Costello and A.C.L. Davies Research Handbook on the Law of the EU’s Internal Market Edited by Panos Koutrakos and Jukka Snell Research Handbook on EU Administrative Law Edited by Carol Harlow, Päivi Leino-Sandberg and Giacinto della Cananea Research Handbook on EU Health Law and Policy Edited by Tamara K. Hervey, Calum Alasdair Young and Louise E. Bishop Research Handbook on EU Law and Human Rights Edited by Sionaidh Douglas-Scott and Nicholas Hatzis Research Handbook on EU Tort Law Edited by Paula Giliker Research Handbook on EU Energy Law and Policy Edited by Rafael Leal-Arcas and Jan Wouters Research Handbook on Legal Pluralism and EU Law Edited by Gareth Davies and Matej Avbelj Research Handbook on EU Sports Law and Policy Edited by Jack Anderson, Richard Parrish and Borja García-García Research Handbook on the EU’s Common Foreign and Security Policy Edited by Steven Blockmans and Panos Koutrakos

Research Handbook on the EU’s Common Foreign and Security Policy

Edited by

Steven Blockmans Professor of EU External Relations Law and Governance, University of Amsterdam, the Netherlands and Head of EU Foreign Policy, CEPS, Belgium

Panos Koutrakos Professor of European Union Law and Jean Monnet Professor of EU Law, City, University of London, UK

RESEARCH HANDBOOKS IN EUROPEAN LAW

Cheltenham, UK + Northampton, MA, USA

© Steven Blockmans and Panos Koutrakos 2018 © David Aaron Smith image on cover ‘Martian Attack of 1885’ All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library Library of Congress Control Number: 2018946013 This book is available electronically in the Law subject collection DOI 10.4337/9781785364082

ISBN 978 1 78536 407 5 (cased) ISBN 978 1 78536 408 2 (eBook)

02

Typeset by Columns Design XML Ltd, Reading

To the memory of Professor Simon Duke, our friend and colleague.

Contents

List of contributors Acknowledgements

ix xi

Introduction Steven Blockmans and Panos Koutrakos

1

PART A

THE LAW OF CFSP/CSDP

1. The position of CFSP/CSDP in the EU’s constitutional architecture Marise Cremona 2. The institutional framework, legal instruments and decision-making procedures Stephan Marquardt 3. Representing the EU in the area of CFSP: legal and political dynamics Hylke Dijkstra and Peter Van Elsuwege 4. ‘The Good, the Bad and the Ugly’: three levels of judicial control over the CFSP Christophe Hillion and Ramses A. Wessel PART B

22 44

65

THE PRACTICE OF CSDP

5. Civilian CSDP missions: ‘the good, the bad and the ugly’ Ana E. Juncos 6. Military CSDP operations: strategy, financing, effectiveness Daniel Fiott 7. Responsibility and liability for CSDP operations Joni Heliskoski 8. Capabilities and CSDP: resourcing political will or paper armies Simon Duke 9. Non-proliferation of weapons of mass destruction Eileen Denza PART C

5

89 111 132 154 182

THE NEXUS BETWEEN CFSP/CSDP AND OTHER EXTERNAL POLICIES

10. The law and practice of EU sanctions Christina Eckes 11. The nexus between the CCP and the CFSP: achieving foreign policy goals through trade restrictions and market access Andrea Ott and Guillaume Van der Loo vii

206

230

viii Research handbook on the EU’s common foreign and security policy 12. EU development cooperation and the CFSP: mutual encroachment? Morten Broberg 13. The CFSP–humanitarian aid nexus Mireia Estrada Cañamares 14. The nexus between CFSP/CSDP and the Area of Freedom, Security and Justice Panos Koutrakos 15. European Neighbourhood Policy: CFSP in disguise Steven Blockmans 16. The nexus between Common Foreign and Security Policy and energy policy Theodore Konstadinides and Despoina Mantzari 17. The European Union as a cybersecurity actor Jed Odermatt 18. EU external health security policy and law Anniek de Ruijter PART D

254 276

296 312

331 354 374

REFLECTION ON PERSPECTIVES

19. The European Union as post-national realist power Achilles Skordas 20. Too much, too little, too late? Reflections on law and ethics in the EU’s foreign policy Jan Klabbers 21. The European Union’s foreign policies: an external examination of the capabilities–expectations gap Urfan Khaliq

394

445

459

Conclusions Steven Blockmans and Panos Koutrakos

496

Index

517

Contributors

Steven Blockmans is Professor of EU External Relations Law and Governance at the University of Amsterdam and Head of EU Foreign Policy at CEPS, Brussels. Morten Broberg is Professor of Law at the University of Copenhagen. Marise Cremona is Professor Emeritus at the European University Institute, Florence. Anniek de Ruijter is Assistant Professor of European Law at the Amsterdam Centre for European Law and Governance (ACELG). Eileen Denza was Visiting Professor at University College London, and was Legal Counsellor, Foreign and Commonwealth Office and Counsel to the EC Committee of the House of Lords. Hylke Dijkstra is Assistant Professor in the Department of Political Science, Maastricht University. The late Simon Duke was Professor at the European Institute of Public Administration, Senior Research Fellow at Maastricht University and Visiting Professor at the Mediterranean Academy of Diplomatic Studies, University of Malta. Christina Eckes is Professor of European Law at the University of Amsterdam and Director of the Amsterdam Centre for European Law and Governance (ACELG). Mireia Estrada Cañamares is an Associate at Cuatrecasas, Madrid. Daniel Fiott is Security and Defence Editor at the EU Institute for Security Studies and a Visiting Lecturer at the Brussels School of International Studies, University of Kent. Joni Heliskoski is Director of EU Litigation at the Finnish Ministry for Foreign Affairs. Christophe Hillion is Professor of European Law at the Universities of Oslo and Leiden, and Researcher at the Norwegian Institute of International Affairs (NUPI) and at the Swedish Institute for European Policy Studies (SIEPS). Ana E. Juncos is Reader in European Politics at University of Bristol. Urfan Khaliq is Professor of Public International and European Laws at Cardiff University. ix

x Research handbook on the EU’s common foreign and security policy Jan Klabbers is Academy Professor (Martti Ahtisaari Chair) at the University of Helsinki and Visiting Research Professor at Erasmus School of Law, Rotterdam. Theodore Konstadinides is Professor of Law at the University of Essex. Panos Koutrakos is Professor of European Union Law and Jean Monnet Chair in European Union Law at City, University of London. Despoina Mantzari is Lecturer in Law at University College London. Stephan Marquardt is a Legal Adviser at the European External Action Service. Jed Odermatt is Lecturer in Law at City, University of London. Andrea Ott is Professor of EU External Relations Law and Jean Monnet Chair in EU law at Maastricht University. Achilles Skordas is Professor of International Law at the Max Planck Institute for Comparative Public Law and International Law, Heidelberg. Guillaume Van der Loo is a postdoctoral researcher at Ghent University and a researcher at CEPS, Brussels. Peter Van Elsuwege is Professor of European Union Law at Ghent University and Co-Director of the Ghent European Law Institute (GELI). Ramses A. Wessel is Professor of International and European Law & Governance at the University of Twente.

Acknowledgements

This book is dedicated to the memory of Professor Simon Duke. Simon was one of the pioneers and titans of the CFSP/CSDP scholarship whose work was a point of reference for scholars and practitioners in the field. We were delighted to have him contribute to this Research Handbook. More importantly, Simon was a kind and generous friend. We shall miss him very much. The Editors of this Research Handbook organised a meeting for the contributing authors that was held at City, University of London in April 2016. We are grateful for the funding provided by the Erasmus+ programme of the European Commission (Jean Monnet Chair in EU Law). Many thanks to Douglas Grant and Jackie West for their editorial assistance.

xi

Introduction Steven Blockmans and Panos Koutrakos

1. THE POLICY AND LEGAL CANVAS It has been nearly a decade since the entry into force of the Lisbon Treaty and the ensuing reorganisation of the legal rules and procedures governing the European Union’s external action. During this period, the Union has carried out its Common Foreign and Security Policy (CFSP) and the Common Security and Defence Policy (CSDP) at a steady pace while revamping its institutional framework. This has been a tumultuous period for the Union in general and its external action in particular. On the back of the eastern enlargement of the Union, the economic, financial and political crisis has challenged the most fundamental assumptions about the European integration process, namely the prosperity of the EU, the solidarity between creditors and debtors, and the viability of the institutional structures of the Eurozone. Furthermore, the decision of the United Kingdom to leave the Union following the referendum of 23 June 2016 has raised fundamental questions about the future of the EU and the cohesion and direction of the integration process. Against this evolving canvass in foro interno, a number of factors in foro externo have raised more acute questions about the design of the CFSP/CSDP. The first is regional: the Arab uprisings, the violent implosion of Libya and Syria, Russia’s annexation of Crimea and destabilisation of Donbas challenge the role of the EU as a security provider and foreign policy actor, and its effectiveness in the Union’s neighbourhood. The second factor is the increasingly broad scope of foreign policy: the prominence of cyber security as a foreign policy imperative and the consternation from the implications of the Ebola crisis have added further to the broad construction of the CFSP/CSDP. Pushed also by the other factors laid out here, this led to a reassessment of the Union’s strategic mainframe (i.e. the 2016 EU Global Strategy on Foreign and Security Policy).1 The third factor is the refugee crisis, which has caught headlines since 2014 and which has placed external border control back at the centre of the political and security agenda of Member States. The implications of this crisis straddle various aspects of the CFSP/CSDP, including the Union’s relations with Turkey, cooperation with NATO, and the effectiveness of the EU’s own operations in the Mediterranean. The fourth factor is terrorism, which continues to provide a thread that brings together different strands of the Union’s policies, including justice and home affairs, as borne out by the phenomenon of foreign fighters, the Paris attacks and the first-ever invocation of Article 42(7) Treaty on European Union (TEU). The fifth factor is the deepening and widening of the EU’s presence as an international economic actor 1 ‘Shared Vision, Common Action: A Stronger Europe’, June 2016, available at https:// eeas.europa.eu/archives/docs/top_stories/pdf/eugs_review_web.pdf.

1

2 Research handbook on the EU’s common foreign and security policy in an unstable global trading system. The conclusion of Deep and Comprehensive Free Trade Agreements with Ukraine, Georgia and Moldova, the trade and investment negotiations with Canada, China and other actors, and the response to protectionist tendencies by the United States have underlined the origins and core on which the Union’s foreign policy actorness continues to draw, namely trade and development. The sixth factor, the potential impact of which is gradually emerging, is the prospect of Brexit. Together with the other five dynamics, the need to put in place arrangements for continued cooperation between the EU and the UK on foreign, security and defence matters will shape the future of the CFSP/CSDP. Over the last nine years, the Union has been tweaking the rules, structures and procedures governing the conduct of its CFSP/CSDP. At the substantive level, the amalgamation of the objectives of external action aimed to not only introduce coherence in what the Union would do in the world, but also to reduce the scope for competence disputes. At the institutional level, the adjustment process has been particularly noticeable. The High Representative and the European External Action Service have been settling into place while the Commission and the General Secretariat of the Council have been trying to accommodate to the new modus operandi in ways to offset any loss or sharing of personnel and expertise. The European Council, on the other hand, has increasingly taken charge of crisis management in the realm of foreign and security policy. The European Parliament too has emerged as a confident, ambitious and active player, while new inter-institutional arrangements have been introduced in order to manage the interactions between the relevant actors. In this evolving context, the Court of Justice of the European Union has gradually become more prominent, even if the scope of its role and the implications of its case-law are still not clear. While there is no doubt as to its jurisdiction to patrol the border line between the CFSP/CSDP and the other strands of the Union’s external action, the question has been raised as to whether the integrated nature of the Union’s foreign policy requires that the Court exercise its powers on substantive issues with a CFSP policy dimension too.

2. OBJECTIVES OF THE RESEARCH HANDBOOK Having sketched out the policy and legal canvass against which this book examines the CFSP/CSDP, we would also like to set out the common objectives and threads of our chapters. These are as follows: + to establish and analyse the intrinsic links (institutional, procedural, substantive) between the EU’s legal rules and procedures and the deeply politicised context within which these are applied in the evolving external action of the Union. + to identify legal challenges to the implementation of an integrated approach to EU external action and to gauge their implications for both the legal and policy frameworks of the CFSP. + to examine the extent to which the legal framework and practice in the CFSP is governed by flexibility and contributes to the efficient and effective conduct of the Union’s external action. + to identify new trends emerging from the practice of the CFSP.

Introduction 3 In order to achieve these objectives in a manner that captures the multifarious nature and complexity of the policy, this book has brought together a range of legal scholars, from both EU and international law backgrounds, and international relations experts.

PART A THE LAW OF CFSP/CSDP

1. The position of CFSP/CSDP in the EU’s constitutional architecture Marise Cremona

1. THE INTEGRATION OF THE CFSP/CSDP INTO THE EU CONSTITUTIONAL FRAMEWORK Let us begin with the structure of the Treaties. The provisions on external action in the current TEU and the Treaty on the Functioning of the European Union (TFEU), which revised and replaced the EC Treaty, are now in three locations: a group of general provisions in the TEU which are applicable to all external action; substantive and procedural provisions on the CFSP and CSDP in the TEU; and provisions on all other external policy fields and some general procedural provisions in the TFEU. The CFSP/CSDP is thus, with one exception,1 the only substantive external policy field based in the TEU. This gives a sense of continuity with the position as it had been since the Maastricht Treaty, in which the CFSP was placed in the TEU as what was often referred to as the ‘second pillar’.2 However, this apparent continuity masks a significant change in the place of the CFSP within the EU’s constitutional structure, a change resulting from the Lisbon Treaty and which we are now starting to see take a more concrete shape. In 2008, in the first Kadi case, the Court of Justice referred to ‘the coexistence of the Union and the Community as integrated but separate legal orders, and the constitutional architecture of the pillars’.3 With the merger of the Union and Community into the European Union, which ‘replaced and succeeded’ the European Community,4 the constitutional architecture of the Treaties has been substantially altered. Instead of two Treaties (TEU and EC Treaty), each the foundation for a separate legal order, we have a single Union with a single legal personality, founded on two Treaties (TEU and TFEU) which are closely bound together. The TEU and TFEU are of equal legal value and in their texts are almost always referred to in combination as ‘the Treaties’. As already mentioned, the provisions on external action span the two Treaties and although the main provisions on the CFSP are found in the TEU, some procedural provisions – in particular the procedural rules for entering into external agreements – are found in the TFEU. More importantly, for the first time the Union is given a single external mandate in 1

The exception is Article 8 TEU, which invites the EU to establish a ‘special relationship’ with neighbouring countries. 2 The first pillar was the EC Treaty and the third pillar the provisions in the TEU on Justice and Home Affairs. 3 Joined Cases C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation EU:C:2008:461, para 202. 4 Article 1 TEU.

5

6 Research handbook on the EU’s common foreign and security policy Article 3(5) TEU encompassing all the EU’s ‘relations with the wider world’. The Union is to uphold its values and interests and protect its citizens by using all the different elements and instruments of its external action. The Union is bound to respect its values, including the rule of law and human rights, in all its external action, including the CFSP/CSDP.5 Not only do the general provisions governing the Union’s external principles, objectives and strategy (in particular Articles 21 and 22 TEU) apply equally to the CFSP and CSDP,6 so too do the general provisions contained in the TFEU, such as environmental protection,7 transparency,8 and the protection of personal data.9 Within this unified system, the CFSP/CSDP is a distinct Union competence and policy field. Its integration into the overall constitutional architecture has raised questions as to the nature of the competence granted to the Union, the specific rules applicable to the CFSP, its relationship with other policy fields, and the scope of the CFSP.10 It is with these questions that this chapter is concerned. 1.1 The Nature of CFSP Competence The categories of Union competence are set out in Articles 2–6 TFEU. Article 2 defines three types of competence (exclusive, shared, and ‘supporting, coordinating and supplementing’ competences) and two policy-specific competences which by implication do not fall within the three general types of competence. The first of the latter is the competence to provide arrangements for the coordination of the Member States’ economic and employment policies. The second is the CFSP: Article 2(4) TFEU provides that the Union shall have competence ‘to define and implement a common foreign and security policy, including the progressive framing of a common defence policy’. It is clear, therefore, that the CFSP is intended as a policy of the Union, distinct from (though in harmony with) the foreign policies of its Member States. It is not simply a coordination of Member State policy, nor is it confined to ‘supporting’ or ‘supplementing’ Member State policy; rather, the Member States are to support the Union’s policy.11 The Union itself is envisaged as a ‘cohesive force in international relations’12 with its own strategic interests.13 At the same time, the CFSP is not designed to replace or pre-empt Member State foreign or defence policy. It is not described as a shared competence and although the Treaties declare that competences designated as neither exclusive (Article 3 TFEU) nor supporting, coordinating and supplementary (Article 6 TFEU) are to be shared,14 it 5

C-263/14 European Parliament v Council EU:C:2016:435, para 47; C-72/15 PJSC Rosneft Oil Company EU:C:2017:236, para 72. 6 Article 23 TEU. 7 Article 11 TFEU. 8 Article 15 TFEU. 9 Article 16 TFEU. 10 See, e.g., RA Wessel, ‘Lex Imperfecta: Law and Integration in European Foreign and Security Policy’ (2016) European Papers 439. 11 Article 24(3) TEU. 12 Article 24(3) TEU. 13 Article 26(1) TEU. 14 Article 4(1) TFEU.

Position of CFSP/CSDP in the EU’s constitutional architecture 7 seems clear from the specific allocation of CFSP competence in Article 2(4) that this residual allocation does not apply. This is supported by the assurance that the CSDP ‘shall not prejudice the specific character of the security and defence policy of certain Member States’ and shall respect the obligations of Member States within NATO,15 as well as by the (non-binding) Declarations 13 and 14, which ‘underline’ that the CFSP will not affect the responsibilities and powers of the Member States in respect of the formulation and conduct of foreign policy. If the CFSP is not a shared competence subject to pre-emption, we should nonetheless consider whether it is possible for exclusive competence to arise on the basis of Article 3(2) TFEU.16 There is nothing in the Treaty structure to exclude its application a priori and in my view, there is no basis on which to do so. However, it is difficult to see how the conditions for exclusivity set out in Article 3(2) could arise in practice in the case of the CFSP: legislative acts are excluded, and a CFSP action could hardly be a necessary precondition for the exercise of an internal competence. Still, although unusual, it is possible to envisage that common rules could be affected within the meaning of Article 3(2),17 for example in the field of data protection.18 The Treaty provisions on competence in both the TEU and TFEU therefore identify the CFSP/CSDP as a Union policy designed to further the Union’s strategic interests and to operate alongside the continuing activity of the Member States, a policy which is singled out from the start as having a specific character which does not fit neatly into existing categories of Union competence. 1.2 ‘Specific Rules and Procedures’ The specificity of the CFSP, indicated by the individualized reference in Article 2(4) TFEU, is manifest in a number of specific institutional and decision-making rules, as well as in its relationship to other policies. The implementation of the CFSP, according to Article 40 TEU, is not to affect the procedures and institutional powers established for the exercise of TFEU-based competences; nor is the implementation of those policies to affect the procedures and the institutional balance of power within the

15

Article 42(2) and (7) TEU. According to Article 3(2) TFEU, ‘The Union shall also have exclusive competence for the conclusion of an international agreement when its conclusion is provided for in a legislative act of the Union or is necessary to enable the Union to exercise its internal competence, or in so far as its conclusion may affect common rules or alter their scope’. 17 For a recent application of this condition, although outside the scope of the CFSP, see Opinion 3/15 on the conclusion of the Marrakesh Treaty EU:C:2017:114. 18 See text at note 30 for data protection as it applies to the CFSP. Were the Council, acting on the basis of Article 39 TEU, to adopt a decision ‘laying down the rules relating to the protection of individuals with regard to the processing of personal data by the Member States when carrying out activities which fall within the scope of [the CFSP]’ then it is possible to envisage that the conclusion by a Member State of an international agreement with a third country which covered the processing of personal data in the context of CFSP-related activities (such as counter-terrorism) and its transfer to that third country, would ‘affect’ those common rules. 16

8 Research handbook on the EU’s common foreign and security policy CFSP/CSDP.19 No other policy field is subject to a similar rule. We should note first that this ‘non-affect’ clause is even-handed – applying to the interaction in both directions between the CFSP and other policy fields – and second that although it might appear designed to separate the CFSP from other policies, its focus in fact is not on policy content or the nature of the CFSP competence, but rather on procedure and institutional balance. Since the CFSP has a different institutional balance and instrumental toolkit, Article 40 is designed to ensure that each policy field respects its own boundaries and operates within its proper sphere. It therefore gives an explicit articulation to the general principle of choice of appropriate legal basis, which – as the Court has said – has a constitutional significance.20 In the next section we will look more closely at what this might mean. The CFSP/CSDP is subject to ‘specific rules and procedures’.21 These are summarized in Article 24(1) TEU and include the predominant roles of the European Council and Council; the European External Action Service, a ‘functionally autonomous body’ but not an institution, with a mandate to assist the High Representative;22 the importance of the Member States alongside the High Representative in ‘putting into effect’ the CFSP; limitations on the type of act that can be adopted; and the restricted jurisdiction of the Court of Justice.23 They are significant but it is important to emphasize that there is no CFSP ‘carve out’: Article 24(1) TEU does not mean that the CFSP/CSDP is completely exempted from the general rules. Rather it means that the general rules and principles will apply by default, subject to exceptions and specific provisions. We can illustrate this point by looking at two of the principles that are identified in the TFEU as ‘provisions having general application’: transparency and data protection. The principle of transparency found in Article 15(1) TFEU applies to the CFSP; there is no general exclusion but we do find certain exemptions and limits to its application. Thus, since the CFSP excludes the adoption of legislative acts, the requirement found in Article 15(2) TFEU that the Council will meet in public when deliberating and voting on a legislative act, will not be relevant to the CFSP.24 Further, Regulation 1049/2001 governs public access to European Parliament, Council and

19

Article 40 TEU. ‘The choice of the appropriate legal basis of a European Union act has constitutional significance, since to proceed on an incorrect legal basis is liable to invalidate such an act, particularly where the appropriate legal basis lays down a procedure for adopting acts that is different from that which has in fact been followed …’ European Parliament v Council (n 5), para 42. 21 Article 24(1) TEU. 22 Article 13(1) and Article 27(3) TEU; Council Decision 2010/427/EU establishing the organisation and functioning of the European External Action Service [2010] OJ L201/30, Article 1. The EEAS is staffed by members of the Council General Secretariat, the Commission, and the diplomatic services of the Member States. 23 On the institutional structure, decision making and role of the Court of Justice, see further Chapters 2 and 4 in this volume. 24 Article 31(1) TEU. 20

Position of CFSP/CSDP in the EU’s constitutional architecture 9 Commission documents and there is no general exclusion for CFSP documents.25 Access is to be refused where disclosure would undermine the protection of the public interest as regards (inter alia) public security, defence and military matters, and international relations; while these grounds may of course apply to some CFSP documents, they are not exclusively directed at the CFSP.26 The rules on access to documents are made applicable to the EEAS by means of the decision establishing the EEAS,27 implemented by a decision of the High Representative.28 A right of protection of personal data is affirmed in Article 16(1) TFEU and this is applicable to the CFSP.29 Article 16(2) then establishes the legal basis for adopting rules on data protection, and these rules are to operate ‘without prejudice’ to the specific rules for the CFSP/CSDP laid down according to the terms of Article 39 TEU. Article 39 TEU does not exempt the CFSP from data protection; rather, it provides for a separate decision-making procedure to operate ‘in accordance with’ Article 16 TFEU and ‘by way of derogation from’ Article 16(2) where data is processed by Member States when acting within the scope of the CFSP. The principle of data protection is maintained but where Member States are concerned it is to be operationalized by way of separate decision-making procedures.30 As far as the Union institutions are concerned, no separate legal basis is created for CFSP-related data processing. Although the current Regulation, adopted before the entry into force of the Lisbon Treaty, applies only within the scope of Community law and therefore not to the CFSP, the decision establishing the EEAS provides that the EEAS will apply the Regulation.31 The 25 Regulation 1049/2001/EC of the European Parliament and of the Council [2001] OJ L145/43. The Preamble para 7 makes it clear that the right of access also applies to documents related to the CFSP. 26 Ibid, Article 4(1). See also Article 9 on the treatment of sensitive documents. See generally P Leino, ‘The Principle of Transparency in EU External Relations Law – Does Diplomatic Secrecy Stand a Chance of Surviving the Age of Twitter?’ in M Cremona (ed.), Structural Principles in EU External Relations Law (Hart Publishing 2018). 27 Article 11 of Council Decision 2010/427/EU establishing the organisation and functioning of the European External Action Service [2010] OJ L201/30. 28 Decision 2011/C 243/08 of the High Representative [2011] OJ C243/16, implements access to documents for the EEAS. The Decision provides that the right of access to EEAS documents will operate ‘according to the principles, conditions and limits’ laid down in Regulation 1049/2001. 29 See also Article 8, Charter of Fundamental Rights. 30 In conformity with this approach, and in contrast to that adopted in relation to access to documents, the General Data Protection Regulation ‘does not apply to the processing of personal data by the Member States when carrying out activities in relation to the common foreign and security policy of the Union’: Regulation 2016/679/EU [2016] OJ L119/1, Preamble para 16 and Article 2(2)(b). No measure has yet been adopted under Article 39 TEU. 31 Regulation 45/2001/EC of the European Parliament and of the Council of 18 December 2000 on the protection of individuals with regard to the processing of personal data by the Community institutions and bodies and on the free movement of such data [2001] OJ L8/1, Preamble para 15 and Article 3(1). Council Decision 2010/42/EU of 26 July 2010 establishing the organisation and functioning of the European External Action Service [2010] OJ L201/30, Article 11(3). Implementing rules were adopted by the High Representative in 2011: Decision 2012/C 308/07 of the High Representative of the Union for Foreign Affairs and Security Policy of 8 December 2011 on the rules regarding data protection [2012] OJ C308/8. See further

10 Research handbook on the EU’s common foreign and security policy proposed replacement for the 2001 Regulation would in any event apply to all activities falling within the scope of Union law.32 As these examples illustrate, the ‘specific rules and procedures’ applicable to the CFSP are concerned with variations in decision making and more broadly with the functions and roles of the different institutions, the ‘specific division of tasks among the institutions of the Union in that field’, as the Court of Justice has put it.33 Thus, within the CFSP the Council may adopt a limited range of legal acts: decisions, but not directives or regulations. The Council’s decisions in the field of the CFSP play a variety of roles; they may relate to operational action,34 define the Union’s position in relation to a third country or issue,35 or conclude an international agreement.36 These decisions are legal acts subject to the procedural requirements set out in the Treaty,37 including the obligation to state the reasons on which they are based.38 As they are not legislative acts they are not adopted according to the ordinary or special legislative procedure and the role of the European Parliament is limited.39 The most significant of these specific rules concerns the role of the Court of Justice. Its jurisdiction over ‘the provisions relating to’ the CFSP and acts adopted on the basis of those provisions is excluded, subject to two exceptions: first, the jurisdiction to monitor compliance with Article 40 TEU and second, jurisdiction to review the legality of decisions providing for restrictive measures against natural or legal persons adopted under CFSP powers.40 The already significant case law on the interpretation of this exclusion of jurisdiction and its exceptions is the focus of Chapter 4 and will not be discussed in detail here.41 We do need, however, to summarize the overall approach of the Court of Justice to the question of its jurisdiction over the CFSP as it demonstrates its view of the place of the CFSP in the EU legal order. First, the Court has stressed that the exclusion of jurisdiction is a derogation which should be interpreted strictly. The ‘rule of general jurisdiction’42 from which Article S Blockmans and C Hillion (eds), EEAS 2.0: A Legal Commentary on Council Decision 2010/427/EU Establishing the Organisation and Functioning of the European External Action Service (SIEPS 2013) 119–121. 32 Proposal for a Regulation of the European Parliament and of the Council on the protection of individuals with regard to the processing of personal data by the Union institutions, bodies, offices and agencies and on the free movement of such data, and repealing Regulation (EC) No 45/2001 and Decision No 1247/2002/EC COM (2017) 08 final. 33 Rosneft (n 5), para 92. 34 Article 28 TEU. 35 Article 29 TEU. 36 Article 37 TEU. 37 Articles 31 TEU and 218 TFEU. 38 Article 296 TFEU; e.g. Rosneft (n 5), paras 118–125. 39 Legislative acts are excluded by virtue of Article 31 TEU. The use of the so-called ‘flexibility clause’ is also excluded: Article 352(4) TFEU. For further discussion of CFSP instruments, see Chapter 2 in this volume. 40 Article 275 TFEU. 41 See also C Hillion, ‘A Powerless Court? The European Court of Justice and the EU Common Foreign and Security Policy’ in M Cremona and A Thies (eds), The European Court of Justice and External Relations Law – Constitutional Challenges (Hart Publishing 2014). 42 C-658/11 European Parliament v Council EU:C:2014:2025, para 70.

Position of CFSP/CSDP in the EU’s constitutional architecture 11 275 TFEU derogates is contained in Article 19 TEU, which requires the Court to ‘ensure that in the interpretation and application of the Treaties the law is observed’. Thus, general provisions or legislation which apply to, or operate in the context of, CFSP acts are not included among the ‘provisions relating to’ the CFSP which are excluded from the Court’s jurisdiction.43 Second, in addition to its reliance on Article 19 TEU, in particular where an individual challenge to the validity of CFSP acts is concerned, the Court has based its strict interpretation of the derogation on the fundamental principles of the rule of law and effective judicial protection. The rule of law is found among the Union’s values in Article 2 TEU and its application to the CFSP is made clear by Articles 21(1) and 23 TEU; the principle of effective judicial protection is contained in Article 47 of the Charter of Fundamental Rights. As the Court pointed out in a recent judgment, ‘the very existence of effective judicial review designed to ensure compliance with provisions of EU law is of the essence of the rule of law’44 and ‘[w]hile, admittedly, Article 47 of the Charter cannot confer jurisdiction on the Court, where the Treaties exclude it, the principle of effective judicial protection nonetheless implies that the exclusion of the Court’s jurisdiction in the field of the CFSP should be interpreted strictly’.45 This emphasis on the Court’s general jurisdiction and the use of general principles to guide the interpretation of the exclusionary rule provides us with a clear demonstration of what it means to say that the CFSP is an integral part of the general scheme of the Treaties and the Union’s legal order. The procedure for the conclusion of international agreements provides a good example of the application of general and specific procedural rules to the CFSP, and their interrelation. Article 218 TFEU provides a general set of rules for treaty making.46 The general framework also applies to the CFSP insofar as no specific rule is provided; thus, for example, the general rule that the Parliament should be kept informed throughout the treaty-making procedure also applies to CFSP agreements, despite its exclusion from the formal decision-making process.47 On the other hand, specific rules are included for initiating the opening of negotiations where the agreement ‘relates exclusively or principally’ to the CFSP, and for the conclusion of agreements which ‘relate exclusively’ to the CFSP. Negotiations are initiated by a recommendation of the High Representative (rather than the Commission) to the Council, which authorizes the opening of negotiations. Agreements are concluded by the Council acting unanimously and, unlike any other policy field, there is no requirement for the Parliament to consent 43

For example, Article 218(10) TFEU, applicable to CFSP agreements: European Parliament v Council (n 42); see text at note 47 below. See also C439/13 P Elitaliana SpA v Eulex Kosovo EU:C:2015:753, with respect to the operation of the Union’s financial regulations in the context of the CFSP. 44 Rosneft (n 5), para 73. 45 Ibid, para 74. 46 For a detailed analysis, see A Dashwood, ‘EU Acts and Member State Acts in the Negotiation, Conclusion and Implementation of International Agreements’ in M Cremona and C Kilpatrick (eds), EU Legal Acts – Challenges and Transformations (OUP 2018); M Cremona, ‘Who Can Make Treaties? The European Union’ in Duncan Hollis (ed.), The Oxford Guide to Treaties (OUP 2012). 47 European Parliament v Council (n 42), paras 81–86.

12 Research handbook on the EU’s common foreign and security policy or be consulted. Agreements ‘relate exclusively’ to the CFSP for the purposes of the choice of procedure where the CFSP provides the sole legal basis: thus, the procedural legal basis follows the substantive legal basis.48 How, then, is the substantive choice of legal basis made? 1.3 Choice of Legal Basis Before the Lisbon Treaty, the former Article 47 TEU which governed the relationship between the European Community and the CFSP pillar was interpreted to require that if a measure could be adopted under EC Treaty powers, then those powers, and not the CFSP, should be used.49 Under the current Treaties, the Court’s approach is very different and reflects the logic of the single legal order. Article 40 TEU has not been crafted into a CFSP-specific legal basis test. On the contrary, the Court uses its normal legal basis reasoning based on the aim and content of the measure in question, and Article 40 TEU simply makes it clear that the Court does have jurisdiction to determine CFSP legal basis questions. The Court insists that the choice of legal basis should follow the aim and content analysis of the measure rather than be determined by the nature of the CFSP’s specific procedures: ‘it is not procedures that define the legal basis of a measure but the legal basis of a measure that determines the procedures to be followed in adopting that measure’.50 Thus it rejected an argument proposed by the European Parliament that the restricted role for the Parliament in the adoption of CFSP measures should be a reason for favouring a non-CFSP legal basis, especially where individual rights are concerned, simply asserting that this outcome was the consequence of the choice made by the framers of the Lisbon Treaty.51 Similarly, where there appeared to be no institutional disagreement concerning the substantive legal basis of a measure the Court accepted this and did not take it upon itself to question that judgment.52 Most strikingly, where substantive legal basis is contested, the Court has not only applied its standard aim and content test, but has also adopted a ‘predominant purpose’ approach to identifying the aim of an international agreement, which favours a single legal basis where possible. As a result, an international agreement will be based on a CFSP legal basis (Article 37 TEU) where its predominant purpose falls within the scope of the CFSP even though some of its provisions relate to other policy fields.53 Given the breadth of the CFSP,54 this approach to legal basis has important consequences: it means that once a CFSP legal basis has been chosen, it will be hard to challenge. As a rough indication, at the time of writing 75 EU agreements in force in the field of the CFSP are listed in the EU’s database of international agreements.55 48

Ibid, paras 58–60. C-91/05 Commission v Council EU:C:2008:288. 50 C-130/10 European Parliament v Council EU:C:2012:472, para 80. 51 Ibid, para 82. 52 European Parliament v Council (n 42), paras 44–45. The contestation of institutional prerogatives in this case concerned the procedural legal basis. 53 European Parliament v Council (n 5). 54 See below. 55 For examples of the types of agreement concluded under CFSP powers, see text at note 96 below. 49

Position of CFSP/CSDP in the EU’s constitutional architecture 13 The specificities of CFSP decision-making might be expected to produce difficulties in combining a CFSP and non-CFSP legal basis. In practice this is more likely to be the case in the adoption of internal acts, especially where the non-CFSP legal basis requires the ordinary legislative procedure, since the Court seems to regard that procedure as incompatible with CFSP decision-making requiring a unanimous vote in the Council acting alone.56 In the case of international agreements, concluded in both cases by Council decision, the differences (for example the role of the Parliament) do not appear such as to render joint legal bases incompatible. A number of examples of such joint legal bases now exist,57 and the Court, while not expressing a direct view, has proceeded on the basis that there is no barrier in principle.58 This approach to legal basis is evidence of the assimilation of the CFSP into the Union legal order; the choice between the CFSP and other policies is treated in the same way as any other legal basis question. There are specific rules governing its operation which derogate from the norm – such as the limits to the Court’s jurisdiction – and these, as befits derogations, are interpreted strictly. But – and this is the crucial point – the CFSP itself is not a derogation, to be constrained within strict limits. This brings us to the second part of this chapter: a consideration of the proper scope and objectives of the CFSP. Put in terms of choice of legal basis: what are the defining features of the aim and content of a measure which determine that it should be adopted under CFSP powers, as opposed to other external competences? 2. The Objectives and Proper Scope of the CFSP/CSDP Article 24(1) TEU sets out a broad perspective for the Union’s CFSP/CSDP: ‘The Union’s competence in matters of common foreign and security policy shall cover all areas of foreign policy and all questions relating to the Union’s security, including the progressive framing of a common defence policy that might lead to a common defence.’ There is no indication here of how we might identify the scope of ‘foreign 56

European Parliament v Council (n 50), para 47. See, for example, Council Decision 2012/308/CFSP on the accession of the European Union to the Treaty of Amity and Cooperation in Southeast Asia [2012] OJ L154/1; Council Decision 2017/434/EU on the signing, on behalf of the Union, and provisional application of the Cooperation Agreement on Partnership and Development between the European Union and its Member States, of the one part, and the Islamic Republic of Afghanistan, of the other part [2017] OJ L67/1; Council Decision 2016/2118/EU on the signing, on behalf of the Union, and provisional application of the Strategic Partnership Agreement between the European Union and its Member States, of the one part, and Canada, of the other part [2016] OJ L329/43; Council Decision 2016/2079/EU on the signing, on behalf of the European Union, and provisional application of the Partnership Agreement on Relations and Cooperation between the European Union and its Member States, of the one part, and New Zealand, of the other part [2016] OJ L321/1. 58 European Parliament v Council (n 5), in which the Court applied the predominant purpose test without suggesting that the possibility of a joint legal basis should be excluded in principle. See also C-658/11 European Parliament v Council EU:C:2014:41, Opinion of AG Bot, para 39, in which the AG envisages the possibility of a joint legal basis, while ultimately finding it unnecessary; and to the same effect C-263/14 European Parliament v Council EU:C:2015:729, Opinion of AG Kokott, paras 52–53. 57

14 Research handbook on the EU’s common foreign and security policy policy’ or ‘the Union’s security’ and how they might be distinguished from other external policies, express or implied. That such an identification and distinction is necessary follows from the need to choose an appropriate legal basis for each measure and from the existence of Article 40 TEU, which emphasizes the constitutionality of this choice.59 2.1 Complementarity An important first step, which also signalled the degree to which the current legal architecture differs from the pre-Lisbon framework, is the clear position taken by the Court rejecting any suggestion that the CFSP should be seen as a general or residual competence, to be used (only) when other competences do not apply. Before the Lisbon Treaty, as has already been mentioned, Article 47 TEU was interpreted as establishing a ‘Community preference’; where a measure fell within the scope of a Community competence, that competence should be used to the exclusion of the CFSP.60 In the first case after the Lisbon Treaty in which the relationship between the CFSP and other competences fell to be considered, albeit indirectly, the legal basis under challenge was Article 215 TFEU, which provides for the adoption of restrictive measures against third countries or individuals following the adoption of a CFSP decision. The case concerned whether the restrictive measures in question were correctly adopted under this combined procedure, or whether the newly introduced Article 75 TFEU, which provides for asset-freezing measures for the purposes of combating terrorism, should have been used instead. In the view of the Advocate General, it was not ‘appropriate’ to view the relationship between Articles 75 and 215 TFEU as being that of lex specialis and lex generalis.61 The Court agreed, holding that the relationship was not that between specific and general competences, but rather one of complementarity.62 Complementarity and the absence of hierarchy or preference is also consistent with the approach of the Court to Article 40 TEU and its use of the standard legal basis tests, mentioned above, including its acceptance that non-predominant, non-CFSP purposes may be subsumed within a CFSP agreement, and that there is no barrier in principle to the possibility of joint CFSP and non-CFSP legal bases.63 While the procedural exceptions are being interpreted strictly, there is no evidence that the Court is trying to 59

As expressed by AG Kokott in C-263/14 European Parliament v Council EU:C:2015:729, para 4, ‘the problem at issue here has considerable political and even constitutional implications because it is necessary to define more sharply the limits of the common foreign and security policy and to delimit it from other European Union policies’. 60 See note 49. 61 C-130/10 European Parliament v Council EU:C:2012:50, Opinion of AG Bot, para 69. The case concerned the choice between Article 75 TFEU (a competence in the field of the Area of Freedom, Security and Justice) and Article 215 TFEU (an external competence to adopt restrictive measures), the latter having a direct link with the CFSP via the prerequisite of a CFSP decision. Acts adopted on the basis of Article 215 are not themselves CFSP acts, but are designed to give effect to a CFSP decision, the latter being a prior condition for the use of Article 215. 62 European Parliament v Council (n 50), para 66. 63 European Parliament v Council (n 5).

Position of CFSP/CSDP in the EU’s constitutional architecture 15 minimize the scope of the CFSP; rather, its judgments suggest that the CFSP occupies a policy space with its own characteristics and legitimacy within the overall EU system. How then, against this background, should the boundaries of that policy space be defined? We can frame our attempt to answer that question in the terms in which the Court of Justice examines the issue of legal basis: on the one hand the aims and objectives of the CFSP, and on the other the content and policy instruments available.64 2.2 Aims and Objectives of the CFSP Prior to the Lisbon Treaty the CFSP possessed its own set of Treaty-defined objectives; as part of the integration of the CFSP into the general framework of the Treaties by the Lisbon Treaty, these objectives were incorporated into the set of general external objectives included in Article 21 TEU and applicable to all external action. The establishment of general external objectives is part of the Lisbon Treaty’s effort to encourage coherence in EU external action,65 and although some external policy fields have specific prioritised objectives,66 the CFSP does not. Advocate General Bot has argued that the first three of the objectives listed in Article 21(2) TEU are ‘among those traditionally assigned’ to the CFSP, and indeed had originally appeared among the CFSP’s objectives in Article 11(1) of the pre-Lisbon Treaty.67 While this is true, there is nothing in the text to set apart any of the objectives listed in Article 21(2) as being particularly concerned with the CFSP, and indeed Article 23 TEU provides that the EU’s CFSP actions shall be guided by each of these objectives without distinction. It is hard, therefore, to see how these Treaty-defined objectives could be used to help determine the scope of the CFSP in relation to other external policies. The Court of Justice has indeed refrained from taking up the Advocate General’s invitation to declare specific objectives as peculiarly linked to the CFSP. It has been able to avoid doing so due to the context of the specific cases coming before it. In these cases the determination has involved, alongside the CFSP, internally based implied external powers and in each case the Court has preferred the explicit external competence over an implied power derived from an internal competence.68 64 ‘[T]he choice of the legal basis of a European Union act, including one adopted in order to conclude an international agreement such as that at issue in this case, must rest on objective factors amenable to judicial review, which include the aim and content of that measure.’ European Parliament v Council (n 5), para 43. 65 Article 21(3) TEU. 66 For example, the eradication of poverty as the ‘primary objective’ of the EU’s development cooperation policy: Article 208(1) TFEU. 67 European Parliament v Council (n 61), Opinion of AG Bot, para 63. The first three objectives in Article 21(2) TEU are (a) to safeguard the values, fundamental interests, security, independence and integrity of the Union; (b) to consolidate and support democracy, the rule of law, human rights and the principles of international law; and (c) to preserve peace, prevent conflicts and strengthen international security, in accordance with the purposes and principles of the United Nations Charter, with the principles of the Helsinki Final Act and with the aims of the Charter of Paris, including those relating to external borders. 68 See in particular cases European Parliament v Council (n 50) and European Parliament v Council (n 5).

16 Research handbook on the EU’s common foreign and security policy In the first of these cases,69 the Court’s argument followed three steps. First, the CFSP-related power (Article 215 TFEU) is explicitly external in nature since it refers to relations with third countries, whereas the objectives of the implied power (Article 75 TFEU) are linked to the establishment of the Area of Freedom, Security and Justice and are thus internal in nature. Second, the measure in question is essentially external since it responds to a threat of international scope (terrorism) and is intended to give effect to a UN Security Council Resolution. Third, although combating terrorism is referred to expressly in Article 75, this can also be an objective of an external measure such as Article 215 since strengthening international peace and security is an explicit objective of the Union’s external action and terrorism is a threat to peace and international security.70 Thus the Court prefers a flexible use of express external powers over the extension of powers which are primarily internal to cover essentially external measures. In the second case, which involved an agreement with Tanzania on the transfer and trial of suspected pirates in the context of an EU naval mission off the coast of Somalia (Operation Atalanta),71 Advocate General Kokott took up an approach previously advocated by Advocate General Bot.72 This was to distinguish between international security, which – they argued – is the proper province of the CFSP, and internal security, as the proper concern of the Area of Freedom, Security and Justice (AFSJ).73 The distinction is at first sight attractive but on reflection it is clear that although it is certainly true that the AFSJ is concerned with the security of the Union, threats to international security will directly or indirectly impact the security of the Union, as exemplified by the phenomenon of ‘foreign fighters’. The Court did not, in any event, espouse this distinction between internal and international security, but preferred instead to link the objectives of the agreement at issue with those of the CSDP mission which it was designed to facilitate; the CSDP character of the agreement was thus made clear. The technique adopted by the Court is not to look in the Treaty for specific CFSP objectives but rather to identify the objectives of the measure and to see whether they can be aligned to the Treaties’ external objectives. Two points may be made here. The first is that this technique works as long as the alternative candidate legal bases are not also part of the Union’s external action: it might be more problematic, for example, were the issue to concern the CFSP versus development cooperation.74 At the time of writing such a case has not been litigated. Second, we can see here the beginnings of an approach which is more evident in the Tanzania Agreement case,75 and which may prove fruitful in the future. This is to address the issue of whether the objectives of a 69

European Parliament v Council (n 50). Ibid, para 61. Note that the Court simply points out that the preservation of international peace and security is an external objective; unlike AG Bot, the Court does not link this objective directly to the CFSP: see in comparison the Opinion of AG Bot (n 67). 71 European Parliament v Council (n 5). 72 In the earlier case, European Parliament v Council (n 42), the choice of legal basis was apparently not contested by the Parliament and the Court accepted the institutional choice of a CFSP legal basis without demur. 73 European Parliament v Council (n 59), Opinion of AG Kokott, paras 61–69. 74 Cf. Commission v Council (n 49). See further Chapter 14. 75 European Parliament v Council (n 5). 70

Position of CFSP/CSDP in the EU’s constitutional architecture 17 measure fall within the scope of the CFSP in two stages. First is the obvious step of ensuring that the objectives of the measure itself coincide with the EU’s general external objectives. Second is the examination of the legal and policy context of the measure in order to identify whether there is a link between the measure and the CFSP, taking into account its place within the EU’s overall policy as well as the international context. Let us examine this point a little more closely. The EU policy context may be derived from other legal acts as well as policy documents.76 The relationship between Article 215 TFEU and the CFSP decision that precedes it allows the Article 215 measure to form part of a broader policy response, identified in the act itself as one of its objectives.77 The Tanzania Agreement case is particularly clear on this point: the agreement was designed to facilitate (and was envisaged in the CFSP act which formed the basis for) the EU’s CSDP Operation Atalanta, and thus it had an organic link to the CFSP. As expressed by the Court: That agreement, concluded pursuant to Article 12 of Joint Action 2008/851, is intimately linked to Operation Atalanta, and consequently, were there to be no such operation, that agreement would be devoid of purpose. … [a]n examination of the aim of the EU-Tanzania Agreement confirms that the procedure … established by the Agreement constitutes an instrument whereby the European Union pursues the objectives of Operation Atalanta, namely to preserve international peace and security, in particular by making it possible to ensure that the perpetrators of acts of piracy do not go unpunished.78

The objectives of the agreement were tied to the objectives of the CSDP mission, and thus to the CFSP/CSDP; the CSDP mission was designed to further one of the EU’s general external objectives: the mission’s CFSP/CSDP basis (while not contested) can be established via the nature of the instrument – a point we will turn to in the next section. In this case, the link was to a CFSP instrument (a Joint Action79) launching a specific operational initiative. In other cases, the link may be to a decision establishing a strategy or policy framework. In addition to CFSP decisions determining operational action (under Article 28 TEU) and defining specific Union positions (under Article 29 TEU), Article 26 TEU envisages that the European Council will adopt decisions identifying the Union’s strategic interests and defining general guidelines for the CFSP, and that the Council will adopt decisions which frame and define the CFSP. These policy-setting decisions adopted on the basis of Article 26 TEU may both indicate specific CFSP objectives and serve as an indication that future implementing decisions would also be adopted as CFSP instruments.80 EU strategy documents may also help to identify the EU’s CFSP objectives on a particular issue, which can then be linked to the 76

For an example, outside the CFSP, of the Court referring to a policy document to establish the link between a Union act and its legal basis, see C-377/12 Commission v Council EU:C:2014:1903, para 42. 77 A point that was alluded to by AG Bot in European Parliament v Council (n 61), para 68. 78 European Parliament v Council (n 5), paras 51 and 54. 79 Joint Actions were a pre-Lisbon CFSP instrument; they are now adopted as decisions under Article 28 TEU. 80 An example of a decision adopted on the basis of Article 26 TEU is Council Decision 2015/2309/CFSP on the promotion of effective arms export controls [2015] OJ L326/56.

18 Research handbook on the EU’s common foreign and security policy objectives of a specific measure to justify using CFSP powers. To take just one example, the EU’s Strategy against Proliferation of Weapons of Mass Destruction,81 while non-binding and therefore not itself providing a legal basis, establishes a set of concrete objectives which have been implemented via a variety of measures; the identification of such an objective within an instrument would clearly indicate that a measure is CFSP related.82 However, care is needed here due to the Union-wide strategic function that such documents may possess. The Union’s Global Strategy, for example, although primarily CFSP in orientation, is designed to ensure coherence in foreign policy overall and contains many references to other external policies such as trade and development cooperation.83 A reference to the Global Strategy would not therefore necessarily suggest that a measure should be identified with the CFSP. As far as the international context is concerned, in both the cases we have discussed here the EU was acting in support of and in accordance with a UN Security Council Resolution. Certainly, the EU is also committed to compliance with the United Nations Charter when acting outside the CFSP and such a link could not be conclusive;84 however, given the security focus shared by the EU’s CFSP/CSDP and the UN Security Council and the explicit reference to the latter in Article 34(2) TEU, such a Security Council context would indicate that the objectives of the measure have a CFSP orientation. In summary, the general external objectives established in the Treaties, precisely because they are general, do not offer much help in assessing whether a measure should be based on CFSP powers. That does not mean that the objective of a measure is irrelevant. The objective of a measure, read in the light of the overall policy context, both the EU context (derived from other legal acts and strategic policy documents) and the international context, will be relevant in determining whether a CFSP legal basis is necessary and/or sufficient. 2.3 Policy Content and Policy Instruments We have seen that, while the aims of a particular instrument may be helpful in assessing whether a CFSP legal basis is appropriate, the CFSP basis thus established will be founded on the EU institutions’ own expression of CFSP objectives in other legal acts or policy documents. The identification of objectives, other than the very broad parameters of Article 21 TEU, can thus give us little guidance in establishing the proper scope of the CFSP from the perspective of the Treaties. In practice the second limb of the classic legal basis test, the content of the measure, is likely to be more helpful in determining the scope of CFSP powers, alongside the type of instrument used. As we have already seen, the type of instrument available is relatively limited: the 81 EU Strategy against proliferation of Weapons of Mass Destruction, adopted by the European Council 12–13 December 2003, Council Document 15708/03. 82 See, for example, Council Decision 2016/51/CFSP in support of the Biological and Toxin Weapons Convention (BTWC) in the framework of the EU Strategy against Proliferation of Weapons of Mass Destruction [2016] OJ L12/50. 83 ‘Shared Vision, Common Action: A Stronger Europe: A Global Strategy for the European Union’s Foreign and Security Policy’ (European External Action Service 2016). 84 Articles 3(5) and 21(1) TEU.

Position of CFSP/CSDP in the EU’s constitutional architecture 19 Treaties provide for the Council and European Council to adopt decisions, and for the conclusion (by Council decision) of international agreements. There is no possibility of adopting legislative instruments. These limits will in themselves constrain the Union’s action under the CFSP, although it should be said that the ‘decision’ may encompass a wide variety of types of action. As far as the substantive content of the CFSP is concerned, on the one hand the Treaties provide some rather general powers: the taking of unilateral positions,85 operational action,86 and the conclusion of international agreements.87 It is difficult to see how the use of these instruments may be substantively constrained under the Treaties other than by the broad principles and objectives applicable to external action in general. Any CFSP-specific constraints result from institutional policy framing, such as the Global Strategy on Foreign and Security Policy,88 rather than the Treaties. On the other hand, we also find some more specific powers: the adoption of data protection rules applicable to the Member States when acting within the scope of the CFSP,89 and the so-called ‘Petersberg tasks’ – tasks deploying civilian and military means including joint disarmament operations, humanitarian and rescue tasks, military advice and assistance tasks, conflict prevention and peace-keeping tasks, tasks of combat forces in crisis management, including peace making and post-conflict stabilization.90 Civilian and military missions may be deployed outside,91 or even within,92 the Union. These provisions give an idea of the types of action in which the Union may engage and are explored elsewhere in this volume.93 In practice, and despite the absence of a formal legal lex generalis/specialis rule,94 CFSP instruments are used for a variety of measure and types of action which, unless they fall within the ‘Petersberg tasks’, relate to general foreign policy rather than sectoral issues. These include disarmament and non-proliferation, counter-terrorism, and restrictive measures against third countries and individuals including visa bans, arms embargoes and asset freezing. International agreements include the foreign policy cooperation and political dialogue provisions in broader international agreements,95 as well as agreements which support CSDP missions such as status of forces and status of mission agreements,96 agreements on the participation of third countries in EU 85

Article 29 TEU. Article 28 TEU. 87 Article 37 TEU. 88 ‘Shared Vision, Common Action’ (n 83). 89 Article 39 TEU. 90 Articles 42(1) and 43 TEU. 91 Article 42(1) TEU. 92 Article 222 TFEU. 93 See for example Chapters 5 and 6. 94 See text at note 61 above. 95 For examples, see note 57. 96 For example, Agreement between the European Union and the Islamic Republic of Afghanistan on the Status of the European Union Police Mission in Afghanistan, EUPOL AFGHANISTAN [2010] OJ L294/2; Agreement between the European Union and the Republic of Mali on the status in the Republic of Mali of the European Union military mission to contribute to the training of the Malian Armed Forces (EUTM Mali) [2013] OJ L106/2. 86

20 Research handbook on the EU’s common foreign and security policy missions,97 agreements on the exchange of confidential information,98 and agreements on the transfer and trial of persons suspected of piracy.99 One further influence and practical constraint on the scope and strategic direction of the CFSP/CSDP should be mentioned: the important role given to the Member States. This role includes both policy definition and capabilities. Article 24(2) TEU provides for the definition and implementation of the CFSP ‘within the framework of the principles and objectives of [the Union’s] external action’, but also involving the Member States’ mutual political solidarity, identification of questions of general interest and convergence of Member States’ actions. The importance of the European Council in defining overall strategic objectives is relevant here too; policy framing and priority setting will be strongly influenced by the Member States. And the implementation of policy, especially in the form of CSDP missions, is dependent on Member States’ capabilities: ‘[t]he performance of these tasks shall be undertaken using capabilities provided by the Member States.’100 The Member States undertake to make civilian and military capabilities available to the Union and to increase their capacity to do so,101 but in practice this is an important constraint on EU action.102

3. CONCLUSIONS This chapter has sought to place the CFSP/CSDP within the overall constitutional architecture of the European Union. It makes three main arguments. The first is the assimilation of the CFSP/CSDP into the general Treaty framework and the overall constitutional structure of the Union, as evidenced by the application to the CFSP/CSDP of the general Treaty principles and objectives. The second is that the specific rules and procedures applied to the CFSP relate to decision-making procedures and the role of the institutions. They are significant but do not take the CFSP/CSDP out of the single legal order of the European Union; as exceptions, they are to be interpreted strictly and they apply only within the strict confines of the CFSP/CSDP. The third is that although the CFSP/CSDP is subject to exceptional procedural and institutional rules, the CFSP/CSDP as a policy field is not treated as an exception. There are no signs that the Court is attempting to limit its scope. It possesses its own 97 For example, Participation Agreement between the European Union and the Swiss Confederation on the participation of the Swiss Confederation in the European Union CSDP mission in Mali (EUCAP Sahel Mali) [2016] OJ L105/3; Agreement between the European Union and Australia establishing a framework for the participation of Australia in European Union crisis management operations [2015] OJ L149/3. 98 For example, Agreement between Georgia and the European Union on security procedures for exchanging and protecting classified information [2016] OJ L300/3. 99 For example, Agreement between the European Union and the United Republic of Tanzania on the conditions of transfer of suspected pirates and associated seized property from the European Union-led Naval Force to the United Republic of Tanzania [2014] OJ L108/3. 100 Article 42(1) TEU. 101 Article 42(3) TEU. 102 See further Chapter 8 in this volume.

Position of CFSP/CSDP in the EU’s constitutional architecture 21 policy space and there is no longer a legally mandated preference against recourse to the CFSP/CSDP to achieve Union objectives. Defining the scope of this policy space is not, however, straightforward and as a result a legal challenge to the choice of a CFSP legal basis, once made, is not easy. CFSP objectives as defined by the Treaties are coterminous with the general external objectives of Union external action, and the definition of specific CFSP objectives is thus a matter for the political institutions, in particular the European Council, the EEAS and the Member States. The objectives of a specific measure may be used in their context to identify the appropriate use of a CFSP legal basis but cannot provide real limits to the CFSP as a policy field. These limits are to be found in the instruments at the EU’s disposal, as well as Member State capabilities. It is thus a policy with great potential scope and power, but is heavily dependent on the engagement of the Member States and the forging of political consensus. To these conclusions we may add a final reflection. The somewhat inchoate nature of the CFSP as a policy field, despite its undoubted centrality to EU external action, is in part the result of the fact that the Treaties – even more clearly than before the Lisbon Treaty – give the CFSP two different functions. It is simultaneously a policy field in its own right with its own specific actions, such as arms control, counter-terrorism, restrictive measures and civilian and military missions – actions which need to be coherent with other elements of EU external action – and the means by which the EU defines its overall foreign policy strategies, bringing (in theory) the different sectoral policies such as trade, development, environment and energy under a coherent umbrella. The overall strategic function was made possible by the revision of Article 40 TEU (the former Article 47 TEU), recast in terms of institutional balance and procedural specificity rather than CFSP ‘encroachment’.103 This dual role for the CFSP is encapsulated in the office of High Representative for Foreign Affairs and Security Policy (chair of the Foreign Affairs Council, a Vice-President of the Commission and chair of the Commissioners’ Group on External Action) and the External Action Service, and has most recently found expression in the Global Strategy for the EU’s foreign and security policy, which takes an overall foreign policy perspective while making frequent reference to sectoral fields of action.104 It poses considerable challenges, not least resulting from different institutional cultures and policy orientations.105 The expectation that the CFSP would offer leadership and policy direction despite relatively weaker institutional structures was one of the imbalances in the Lisbon Treaty architecture, and its ability to do so, while improving, is still in question.

103 104 105

See Commission v Council (n 49). ‘Shared Vision, Common Action’ (n 83). See further the chapters in Part III of this volume.

2. The institutional framework, legal instruments and decision-making procedures Stephan Marquardt

1. INSTITUTIONAL ARRANGEMENTS IN CFSP AND CHANGES INTRODUCED BY THE TREATY OF LISBON Since the creation of the CFSP by the Treaty on European Union in 1993, the CFSP has already undergone a number of significant changes through the Treaties of Amsterdam and Nice. Among the significant developments were the creation of the new function of ‘Secretary-General of the Council/High Representative of the Union for the Common Foreign and Security Policy’ (SG/HR), mandated to assist the Presidency of the Council in its tasks, and the development of new CSDP structures, in particular a permanent Political and Security Committee (PSC) with the possibility for the Council to delegate to it decision-making powers for the conduct of EU crisis management operations. These changes reflected arrangements that had previously been agreed to by the European Council (EUCO) and decisions taken by the Council. In addition, a procedure for the conclusion by the EU of international agreements in the domain of the CFSP was introduced.1 The essential modifications introduced by the Treaty of Lisbon (ToL) in the field of external relations and foreign policy, however, stem from the Treaty establishing a Constitution for Europe (TCE), which represented the most ambitious reform project for the EU until now.2 1.1 Institutional Actors The ToL brought a number of very fundamental changes to the institutional framework of the CFSP, in particular the creation of the function of High Representative of the Union for Foreign Affairs and Security Policy and of the European External Action Service. It also strengthened the role of the EUCO by endowing it with a permanent President, conferring on it a decision-making power and elevating it to the rank of an EU institution.

1

This development could be seen as evidence of the EU’s international legal personality; see S Marquardt, ‘The Conclusion of International Agreements under Article 24 of the Treaty on European Union’ in V Kronenberger (ed.), The European Union and the International Legal Order: Discord or Harmony? (TMC Asser Press 2001). 2 For a comprehensive description of its drafting history and detailed commentary of its provisions, see G Amato, H Bribosia and B De Witte (eds), Genesis and Destiny of the European Constitution (Bruylant 2007).

22

Institutional framework, legal instruments and decision-making procedures 23 1.1.1 The European Council For the first time the EUCO is ranked among the institutions of the EU (Article 13(1) TEU). Its powers are set out in detail in Article 15 TEU. The new feature is a permanent President, elected (by qualified majority) for a period of two-and-a half years (Article 15(5) TEU). Of note in relation to the CFSP is the role of the President of the EUCO to ensure the external representation of the Union ‘in this capacity and at his level’ in this domain, however ‘without prejudice to the powers of the High Representative’. The role of the EUCO in the CFSP is further described in Article 26(1) TEU: as before, it is responsible for defining the ‘strategic interests’ and general guidelines of the Union’s CFSP (subparagraph 1).3 The EUCO normally meets every three months but its President may also convene extraordinary meetings if required by the international situation (subparagraph 2), a provision that did not exist before. This possibility has been used in practice quite frequently, notably in the context of the situation in Ukraine and the migration crisis.4 As before the ToL, the EUCO thus maintains a dominant role, while the Council is responsible for the implementation of the political decisions taken at that level (Article 26(2) TEU). What is new is that the EUCO is now endowed with an express decision-making power (Articles 15(4) and 26(1), 1st subparagraph, last sentence TEU).5 This potentially reinforces its role, although it has not availed itself of this possibility until now (except for the adoption of its rules of procedure).6 1.1.2 The Council of the European Union Notwithstanding the role of the EUCO, the Council remains the central decisionmaking body in the CFSP. In accordance with Article 26(2) TEU, it takes the necessary decisions to define and implement the CFSP.7 The Council further has the mandate, together with the High Representative, to ensure the unity, coherence and efficiency of the Union’s action in the CFSP (Article 26(2) TEU). One important innovation introduced by the ToL directly affects the Council and its working methods in the CFSP, namely the responsibility of the HR to chair the Council in its formation of Ministers for Foreign Affairs (FAC) (Article 18(3) TEU).8 As a consequence, the Council decided that the main advisory body in CFSP matters, the 3

As illustrated by the various ‘EU strategies’ adopted by the EUCO (e.g. on Weapons of Mass Destruction, Small Arms and Light Weapons, Counter-Terrorism); the EUCO also ‘welcomed’ the EU ‘Global Strategy’ presented by the HR in 2016 (cf. EUCO conclusions of 28 June 2016, doc. EUCO 26/16, paragraph 20). 4 Since 2010, ten informal or extraordinary meetings of the EUCO have been convened (and two meetings with Turkey). 5 Including, where provided in the Treaties, the power to decide on treaty amendments. The EUCO cannot, however, adopt legislative acts (Article 15(1), last sentence TEU). 6 Decision of 1 December 2009, OJ 2009 L 315/51. 7 In accordance with Article 24(1), 1st subparagraph TEU, the adoption of legislative acts is excluded in the CFSP. 8 The FAC also regularly meets in the composition of defence ministers; this is, however, not an official formation of the Council (cf. the list of Council formations in Annex I of the Council’s Rules of Procedure (Council Decision 2009/937/EU of 1 December 2009, OJ 2009 L 325/35, as amended)).

24 Research handbook on the EU’s common foreign and security policy PSC, as well as most of the Council Working Parties dealing with CFSP matters, would be chaired by representatives of the EEAS appointed by the HR and no longer by representatives of the rotating Presidency of the Council.9 In practice, this allows the EEAS to play a substantive role in ensuring follow-up to initiatives and proposals of the HR at the stage of deliberations in the Council. Particularly in the domain of the CSDP, the EEAS chairmanship of the PSC has created a standing link between the EEAS and the Council, allowing a rapid implementation of CFSP policies, in particular as regards CSDP missions. In his/her capacity as chair of the FAC, the HR may convene extraordinary meetings of the Council ‘in cases requiring a rapid decision’ (Article 30(2) TEU), on his/her own initiative or at the request of a Member State. The HR has used this possibility only on very few occasions.10 The fact that the PSC is now endowed with a permanent chair has reinforced its role as the linchpin of the CFSP; its role is particularly relevant for the CSDP, where it exercises a delegated decision-making power, in accordance with Article 38(3) TEU. 1.1.3 The High Representative The HR has been conferred with an impressive array of institutional functions, powers and responsibilities, considering that the post is occupied by a single person, even if the HR benefits from the support provided by the EEAS. The main functions of the HR are threefold: + As HR, s/he exercises a number of specific responsibilities in the area of the CFSP and CSDP (described further below); this is arguably his/her most important and complex function, as it breaks down into a series of specific responsibilities, some of which are linked while others are distinct from each other. + The second main function is that of chairing (‘presiding over’) the Foreign Affairs Council (Article 18(3) TEU). + The third main function is that of being one of the Vice-Presidents (VP) of the Commission, in charge of external relations outside the CFSP (Article 18(4) TEU). This does not in the current practice cover the specific external portfolios of trade, development cooperation and humanitarian affairs (although this does not follow from the Treaty); however, the HR as VP of the Commission is responsible for coordinating all policy areas concerning external relations or having an external dimension. One specific emanation of this responsibility is the creation of the Commissioner’s Group on External Action (CGEA, see Section 2.2), chaired by the HR, which brings together the core group of Commissioners dealing with the above portfolios, and other Commissioners responsible for EU policies with an external dimension on an ad hoc basis.

9 10

Decision of 1 December 2009, OJ 2009 L 322/28 (Annex II). cf. the extraordinary meeting on the situation in Ukraine (29 January 2015, doc. 5755/15).

Institutional framework, legal instruments and decision-making procedures 25 The specific responsibilities of the HR in the CFSP are as follows: + to ‘conduct’ the CFSP (including the CSDP), i.e. to ‘contribute to its development through proposals’ and to implement the decisions of the Council (Articles 18(2), 27(1) TEU). + to exercise a right of initiative vis-à-vis the Council, including the submission of proposals for legal acts in the CFSP (Article 30(1) TEU); this right is shared with the Member States.11 In specific cases the Treaties foresee joint HR/Commission proposals.12 + in accordance with Article 27(2) TEU, to exercise the external representation of the Union in the area of the CFSP;13 this includes the conduct of the ‘political dialogue’ with third partners, as well as expressing the Union’s position at international organizations and conferences.14 The latter role of the HR is among the major innovations of the ToL as it abolished the role previously exercised by the Presidency of the Council. In agreeing to this fundamental change, the Member States took an essential step towards ensuring more coherent and effective action of the Union on the international scene, one of the stated objectives of the reform process launched with the work on the constitutional treaty,15 which survived in the ToL. 1.1.4 The EEAS: central administration and EU delegations The idea of an autonomous service to support the HR emerged during the process leading to the constitutional treaty.16 Previously, the SG/HR set up by the Treaty of Amsterdam could rely on the services of the General Secretariat of the Council, as well as on the small so-called ‘policy unit’ composed of national seconded diplomats, and thus only had limited resources. While the necessity of creating the EEAS was acknowledged, its institutional position remained controversial until the adoption of the Council decision establishing the EEAS. Eventually the compromise between diverging positions of the Commission and the Member States was to place the EEAS in between 11

One significant change to be noted here is that the Commission no longer has a right of initiative in the CFSP, as was the case pre-Lisbon, and is no longer ‘fully associated’ in the work carried out in this domain. Besides Article 30(1), the right of initiative of the HR is laid down in specific legal bases in the CFSP chapter: Article 33 (appointment of EU special representatives), Article 41(3), 3rd subparagraph (setting up of a CFSP start-up fund) and Article 42(4) (decisions relating to the CSDP, including those establishing a crisis management operation). 12 E.g. Article 215(1) TFEU (proposals for sanctions regulations) and Article 222(3) TFEU (implementing arrangements for the solidarity clause). 13 The Commission, for its part, is responsible for the EU’s external representation in all other domains of its external action (Article 17(1) TEU). 14 In the area of the CFSP, this responsibility of the HR is shared with the President of the European Council, who fulfils this role at the level of Heads of State or Governments (Article 16(7) TEU). For example, the annual EU statement at the General Assembly of the United Nations is delivered by the President of the European Council. 15 See G Grevi in Amato, Bribosia and De Witte (n 2) 801; JC Piris, The Treaty of Lisbon, A Legal and Political Analysis (CUP 2010) 245. 16 Grevi (n 15) 796–800.

26 Research handbook on the EU’s common foreign and security policy the Council and the Commission and to confer on it administrative and budgetary autonomy.17 The EEAS was created by a Council decision adopted on 26 July 2010 on the basis of Article 27(3) TEU18 (‘EEAS Decision’) and officially started functioning on 1 January 2011. This provision also sets out the basic mandate of the EEAS to support the HR in the fulfilment of her mandate and the initial ‘tripartite’ composition of its staff, made up at the beginning from former Council and Commission officials, as well as diplomats from the Member States. The EEAS Decision defines it as ‘a functionally autonomous body of the European Union, separate from the General Secretariat of the Council and from the Commission’ and endowed with ‘the legal capacity necessary to perform its tasks and attain its objectives’ (Article 1.2 of the EEAS Decision). Arguably, this is a rather ambiguous definition: if the EEAS is ‘autonomous’, why also specify that it is ‘separate’ from the General Secretariat of the Council and the Commission? It also seems odd to put the General Secretariat of the Commission on the same plane as the Commission in this context. Furthermore, the autonomy of the EEAS appears to be limited by the qualifier ‘functionally’. While the legal nature and institutional position of the EEAS thus seem somewhat unclear,19 this has not created any obstacles in practice: the EEAS, on account of its legal capacity and budgetary autonomy, functions like a quasi-institution (it notably appears as defendant in staff cases before the Court of Justice of the European Union (CJEU)),20 even though it maintains administrative links with the Commission and its services.21 The main tasks of the EEAS are to support the HR in fulfilling his/her mandates in the framework of the CFSP (Article 2(1) paragraph 1, 1st indent of the EEAS Decision), to support the HR in her capacity as President of the FAC (2nd indent), and to support the HR in his/her capacity as VP of the Commission (3rd indent). It is further provided that the EEAS also assists the President of the EUCO, the President of the European Commission and the Commission in the exercise of their functions in the area of external relations (Article 2.2 of the EEAS Decision). Besides these core tasks, the EEAS Decision provides that one important role of the EEAS is to contribute to the programming of external assistance instruments (Article 9 of the EEAS Decision). 17

On the preparatory work on the EEAS see Piris (n 15) 249–252. Council Decision 2010/427/EU establishing the organisation and functioning of the European External Action Service OJ 2010 L 201/30. 19 On the legal status and capacity of the EEAS, see B van Vooren, ‘A Legal-institutional Perspective on the European External Action Service’ (2011) 48(2) CML Rev 475; see also S Blockmans and C Hillion (eds), EEAS 2.0 – A Legal Commentary on Council Decision 2010/427/EU Establishing the Organisation and Functioning of the European External Action Service (SIEPS 2013) 13–19. 20 It may be noted that the EEAS is considered as an institution for the purposes of the EU Staff Regulations as well as the EU Financial Regulation. 21 The EEAS maintains a number of ‘Service Level Agreements’ with the Commission, pursuant to which the latter provides a number of administrative services to the EEAS (the conclusion of such agreements is expressly foreseen by Article 3(3) of the EEAS Decision). 18

Institutional framework, legal instruments and decision-making procedures 27 The EEAS is placed under the authority of the HR and has a central administration at Headquarters, organized in thematic and geographical departments, departments for administration and budget, and horizontal issues such as inter-institutional relations and legal affairs. The Union delegations in third countries and at international organizations, another significant novelty introduced by the ToL (Article 221 TFEU), are in charge of representing the EU in third states and at international organizations, on all matters falling within EU competence, i.e. not only in CFSP matters. Most of the existing 140 delegations were previously Commission delegations. They now form an integral part of the EEAS, are under the direct authority of the HR (Article 5 of the EEAS Decision) and thus constitute an essential diplomatic tool of the HR to implement EU external policies on the ground as well as to coordinate the action of the EU Member States, in line with Articles 3422 and 35 TEU. In the light of its broad range of tasks, which in principle cover all of the HR’s three main responsibilities, the EEAS is thereby in a unique position to ‘bridge the gap’ between the CFSP and other EU external instruments and support the HR in her efforts to ensure coordination between the instruments at the disposal of the EU.23 The participation of the EEAS in the programming of external relations instruments is particularly relevant in this regard. It should also be noted in this respect that the EEAS actively participates in the preparation and conduct of the regular meetings of the CGEA. As regards the political decision-making process, the EEAS is also in a unique position to shape the latter at the Council, through its chairmanship of Council preparatory bodies. It may furthermore be noted that the Commission departments responsible for the CFSP financing, the implementation of the Instrument contributing to Peace and Stability (IcSP), as well as for the drafting of sanctions regulations are collocated with the EEAS (Service for Foreign Policy Instruments), thus allowing a close cooperation and coordination of various initiatives and proposals on behalf of the HR. 1.1.5 The European Commission With the ToL the Commission has lost its institutional role in the CFSP: it is no longer ‘fully associated’ in its work (ex-Article 27 TEU), and no longer has a right of initiative (ex-Article 22 TEU). This loss is to some extent ‘compensated’ by the role of the HR as VP of the Commission, which translates in practice into a close association of the competent Commission services with the EEAS in the context of the preparation of HR initiatives and proposals in the CFSP. The practice has furthermore shown that the Commission increasingly relies on its financial responsibilities to maintain an influence over the CFSP (see Section 2.2).

22 The HR has the specific mandate to ‘organise the coordination’ of Member States at international organizations (Article 34.1, 1st subparagraph, last sentence); this means in practice that EU delegations at international organizations are in charge of organizing regular coordination meetings of Member States. 23 On the coordinating role of the EEAS see M Gatti, European External Action Service: Promoting Coherence Through Autonomy and Coordination (Brill Nijhoff 2016).

28 Research handbook on the EU’s common foreign and security policy 1.1.6 The European Parliament From a legal point of view, the role of the European Parliament (EP) in the CFSP has not significantly changed from before the ToL. Pursuant to Article 36 TEU, the HR has the obligation to regularly inform the EP on the most important aspects of the CFSP, and to duly take into consideration the views expressed by the EP. One novelty is that EU special representatives may also be called on to provide information to the EP. Politically, however, the EP plays a relatively large role in exercising scrutiny over the CFSP, notably on account of the fact that it wields the budgetary power (together with the Council) over the CFSP budget, but also because it exercises political supervision over the Commission, with the possibility of a vote of no-confidence (Article 17(8) TEU). The EP must also approve the composition of the Commission as a college and Article 17(7) TEU makes particular reference to the HR (in her capacity as VP of the Commission). This political supervisory role of the EP is laid down in some detail in the HR’s declaration on political accountability, made following the adoption of the Council decision establishing the EEAS.24 On the financial side, the inter-institutional arrangement on budgetary discipline sets out practical modalities allowing the EP to supervise the implementation of the CFSP budget.25 In practice, the HR regularly participates in the EP’s plenary sessions, and senior EEAS officials regularly appear before the EP’s Foreign Affairs (AFET) and Security and Defence (SEDE) Committees and other Committees as required to brief its members on specific CFSP developments. In some cases, the EP has also requested the HR to provide information on specific issues.26 Moreover, in accordance with Article 36(2) TEU, the EP may address questions and recommendations to the Council and the HR, and holds a semi-annual debate on the development of the CFSP. The EP’s role in exercising ‘democratic scrutiny’ over the CFSP has recently been affirmed by the CJEU in relation to the obligation of the Council to inform the EP on the negotiation and conclusion of agreements in the CFSP, pursuant to Article 218(10) TFEU.27 1.1.7 EU Special Representatives The possibility for the Council to appoint a ‘special representative’ (EUSR) with a specific political mandate was first introduced by the Treaty of Amsterdam in 1999 and was maintained in the ToL (Article 33 TEU). This is one of the cases where the Council can act by qualified majority (Article 31(2), 4th indent TEU). There are currently 7 EUSRs with a geographical mandate (Bosnia and Herzegovina, Kosovo, Horn of Africa, Sahel, South Caucasus and Georgia, Central Asia, Middle East Peace Process) and one with a thematic mandate (human rights). 24

OJ 2010 C 210/1. OJ 2013 C 373/1. 26 See, for example, the so-called Jacqué Report of 31 March 2015, issued following a request form the EP to the HR to examine various allegations in relation to the EULEX Kosovo mission. 27 Case C-263/14, Tanzania Pirate Transfer Agreement ECLI:EU:C:2016/435, paras 80, 84. 25

Institutional framework, legal instruments and decision-making procedures 29 EUSRs generally have the mandate to implement and coordinate EU policies in relation to the country or region concerned and to provide the Council and PSC with regular reports on their implementation. In situations where an EU CSDP mission is present, the EUSR is also mandated to ensure a coordinated approach. EUSRs can generally be considered as ‘extended arms’ of the HR as they act as representatives of the EU, albeit only in relation to a specific country or region (or for the implementation of a horizontal policy such as human rights).28 Even though EUSRs report to the Council, they are placed under the direct authority of the HR (Article 33, 2nd sentence TEU). As such, they also constitute a foreign policy tool at the disposal of the HR. 1.2 Instruments and Procedures Under the ToL As noted in the previous chapter, the CFSP, despite its integration into the EU’s external action, still remains subject to the limits of Article 40 TEU and uses specific procedures and instruments, as described below. 1.2.1 Legal instruments The two main CFSP (legal) instruments are identified in Article 25 TEU, described in terms unchanged in substance since the Treaty of Maastricht, namely ‘positions’ (formerly the common positions) and ‘operational actions’ (formerly ‘joint actions’).29 One instrument surprisingly not mentioned in Article 25 TEU is the international agreements that may be concluded under Article 37 TEU, using the procedure set out in Article 218 TFEU. The ToL has unified the form of the legal act to be used: all legal acts in the CFSP are now in the form of ‘decisions’,30 which are legal acts within the meaning of Article 288 TFEU, meaning that they are subject to the formal requirements set out in Article 296(2) TFEU. 1.2.2 Voting rule and procedures The general procedure for the adoption of legal acts in the CFSP is laid down in Article 31 TEU. The general voting rule is unanimity (paragraph 1), however the possibility of

28 This may explain why the HR has the exclusive right to propose the appointment of an EUSR to the Council. 29 Article 25 also refers to the ‘general guidelines’, i.e. those determined by the EUCO pursuant to Article 26, the ‘strengthening of the systematic cooperation between Member States in the conduct of policy’, and (as a new element) ‘arrangements for the implementation of the decisions referred to in points (i) and (ii)’ (i.e. decisions on positions and actions). It is not entirely clear what this means in practice and what form such ‘arrangements’ would take. The wording is rather that of a general objective, concrete expressions of which are to be found in other provisions of the CFSP chapter (Articles 32, 34, 35, 46). 30 Pre-Lisbon, the legal acts were denominated ‘common positions’ and ‘joint actions’; in some cases, the Council also resorted to ‘decisions’ (based on ex-Article 13.3).

30 Research handbook on the EU’s common foreign and security policy abstention (and of the so-called constructive abstention) introduced by the Treaty of Amsterdam remains. This possibility in principle also applies to the CSDP.31 The second paragraph of Article 31 TEU sets out the cases where an adoption of legal acts by qualified majority voting (QMV) is possible. Two of the cases (decisions implementing positions or actions and decisions appointing EUSRs) existed pre-Lisbon (3rd and 4th indents). The first indent is seemingly new but existed before in substance: ex-Article 23(2), 1st indent TEU provided that legal acts implementing a so-called common strategy adopted by the EUCO could be adopted by QMV.32 The common strategies no longer exist. However, the EUCO has now been endowed with an express power to adopt decisions on the EU’s strategic interests, i.e. legal acts (Article 26(1), 2nd sentence TEU). Such decisions (which would be comparable to the former common strategies) could then be implemented by Council decisions adopted by QMV. This possibility has not been used until now. One new possibility of QMV has been added, with a rather convoluted wording: the Council may also act by qualified majority on a proposal from the HR when the latter has been requested by the EUCO to submit such a proposal; this request, in turn, can be made by the EUCO on its own initiative, or upon that of the HR. This is in principle based on the same logic as the first indent, i.e. that decisions of the Council may be adopted by qualified majority once the EUCO has taken a basic policy orientation to be implemented by further decisions. Another new element is the possibility for the EUCO to decide on further cases where qualified majority could be applied (paragraph 3). In practice, however, there has not been any case of actual voting by qualified majority in the Council since the Treaty of Amsterdam introduced this possibility. However, the Council regularly adopts implementing decisions in the CFSP, notably in the domain of sanctions, and the recitals of such decisions refer to Articles 29 and 31(2) TEU as legal basis, thereby indicating that they fall under one of the cases where the Treaty in principle allows for adoption by QMV.33 The same practice is followed for decisions appointing EUSRs under Article 33 TEU.34 The rule that qualified majority decisions are not possible in the domain of the CSDP has been maintained (paragraph 4), as well as the simple majority rule for procedural decisions (paragraph 5). Regarding the decision-making procedure, as indicated above (see Section 1.1.3), the HR enjoys a right of initiative in the CFSP and the Council normally acts on the basis of proposals from the HR, even where the specific legal 31

As confirmed by the practice of the Council, where Cyprus used the possibility of constructive abstention on the occasion of the adoption of the Joint Action establishing the EULEX Kosovo crisis management mission. 32 This was done in one case: cf. Joint Action 1999/878/CFSP of 17 December 1999 establishing a European Union Cooperation Programme for Non-proliferation and Disarmament in the Russian Federation OJ 1999 L 331/11, adopted in implementation of the EU Common Strategy on Russia. 33 See, for instance, Council implementing decision (CFSP) 2017/975 of 8 June 2017 implementing Decision (CFSP) 2016/849 concerning restrictive measures against the Democratic People’s Republic of Korea, OJ 2017 L 146/145. 34 E.g. Council Decision (CFSP) 2015/2274 of 7 December 2015 appointing the European Union Special Representative for the Sahel OJ 2015 L 322/44.

Institutional framework, legal instruments and decision-making procedures 31 basis does not expressly foresee a proposal from the HR (e.g. for decisions imposing sanctions on the basis of Article 29 TEU). In one case to date, the Council has acted upon a proposal formally submitted by several Member States, when adopting the decision establishing a permanent structured cooperation (PESCO) in December 2017.35 This is consistent with the wording of the legal basis (Article 46(2) TEU), even if it does not expressly provide for a proposal from Member States: as it is foreseen that the Council must consult the HR before the adoption of the decision, it would seem odd that the proposal itself would emanate from the HR. The fact that Member States submitted the proposal is also consistent with their right of initiative under Article 30(1) TEU. The proposals to be submitted by the HR to the Council in the domain of the CFSP are prepared by the relevant services within the EEAS, including legal scrutiny by the EEAS Legal Department, and subject to internal approval by the HR herself or a member of her cabinet on her behalf, before the formal transmission by the SecretaryGeneral of the EEAS to his/her counterpart in the Council. 1.3 The Common Security and Defence Policy The provisions on the CSDP (Title V, chapter 2, section 2) constitute an entirely new section in the TEU as amended by the ToL. They are in substance based on the corresponding provisions in the TCE. The CSDP continues to form an integral part of the CFSP, but has itself a number of distinct features as compared to the ‘traditional’ CFSP. The ToL had also substantially reinforced the provisions on a common defence, and expressly integrated the European Defence Agency (EDA) into the CSDP institutional structures. 1.3.1 Institutional structures of the CSDP While the Council remains the sole decision-making institution also in the CSDP the latter domain nonetheless has a number of institutional specificities. Within the Council, there are two dedicated advisory bodies, one for military operations (the EU Military Committee), one for civilian missions (the Committee on the Civilian Aspects of Crisis Management); a dedicated Working Party dealing with cross-cutting issues, the Political and Military Group (PMG); one horizontal advisory body, the PSC, to which the Council generally delegates decision-making powers for the ‘political control and strategic direction’ of crisis management operations/missions, in accordance with Article 38(3) TEU. Furthermore, the General Secretariat of the Council still administers the so-called Athena mechanism, set up for the financing of the common costs of EU military operations.36 Within the EEAS, there are dedicated administrative structures that were transferred from the Council Secretariat, one for the military side (the EU military staff – EUMS) 35 Council Decision (CFSP) 2017/2315 of 11 December 2017, OJ 2017 L 331/57 (‘PESCO Decision’). 36 Council Decision (CFSP) 2015/528 of 27 March 2015 establishing a mechanism to administer the financing of the common costs of European Union operations having military or defence implications (Athena) and repealing Decision 2011/871/CFSP OJ 2015 L 84/39.

32 Research handbook on the EU’s common foreign and security policy and one for civilian missions (the Civilian Planning and Conduct Capability – CPCC, in charge of mission support), as well as the Crisis Management and Planning Directorate (CMPD), in charge of overall strategic and horizontal issues in relation to CSDP missions, both civilian and military. Of note in this context is the recent creation by the Council of a specific entity within the EUMS for the operational conduct of non-executive military missions, the Military Planning and Conduct Capability (MPCC).37 Prior to the ToL the Council had set up four EU agencies/entities within the framework of the CSDP, which have all been ‘re-established’ by Council decisions taken after the entry into force of the ToL: the EDA,38 the European Union Satellite Centre,39 the European Union Institute for Security Studies (EUISS),40 and the European Security and Defence College (ESDC).41 In relation to these agencies, the HR continues to exercise the responsibilities previously held by the SG/HR.42 These agencies constitute yet another set of important additional tools at the disposal of the HR to implement and support the CFSP and CSDP. The EDA, in particular, initially set up through a joint action on the basis of ex-Article 14 TEU, inter alia has the mandate to contribute to the development of Member States’ military capabilities, notably for the purpose of EU crisis operations with military means (Article 42(3), 2nd subparagraph TEU), as well as in the framework of PESCO. The general mandate of the EDA is now set out in Article 45 TEU, introduced by the ToL. The research activities conducted by the EUISS, which works in close cooperation with the EEAS’ strategic planning division, provide contributions to the development of long-term EU political strategies. It has notably provided input for the EU Global Strategy issued by the HR in 2016. The EU Satellite Centre has been instrumental in providing analysis of satellite imagery for the purpose of EU crisis management operations. The ESDC provides regular training courses for the personnel from Member States and EU institutions that may be seconded to EU crisis management operations. 1.3.2 Means of implementation of the CSDP 1.3.2.1 CSDP missions The CSDP covers two main areas: (i) the establishment by the EU of crisis management operations or missions, using civilian and military assets put at the disposal of the EU by the Member States (Article 42(1) TEU); (ii) the ‘progressive framing’ of a common Union defence policy including a possible ‘common defence’ (Article 42(2) TEU). The EU has mostly been active in the first 37 Council Decision (EU) 2017/971 of 8 June 2017 determining the planning and conduct arrangements for EU non-executive military CSDP missions, OJ 2017 L 146/133. 38 Decision 2011/411/CFSP of 12 July 2011, OJ 2011 L 183/16. 39 Decision 2014/401/CFSP of 26 June 2014, OJ 2014 L 188/73. 40 Decision 2014/75/CFSP of 10 February 2014, OJ 2014 L 41/13. 41 Decision 2013/189/CFSP of 22 April 2013, OJ 2013 L 112/22. 42 Recital 7 of the EEAS Decision provides: ‘The High Representative, or his/her representative, should exercise the responsibilities provided for by the respective acts founding the European Defence Agency, the European Union Satellite Centre, the European Institute for Security Studies, and the European Security and Defence College.’

Institutional framework, legal instruments and decision-making procedures 33 domain and has developed a solid practice in conducting crisis management operations with a wide range of mandates. Under the ToL, the scope of CSDP missions is defined with more detail and covers more tasks than before. They pursue the general objectives of ‘peace-keeping, conflict prevention and strengthening international security in accordance with the principles of the United Nations Charter’ (Article 42(1) TEU); more specifically, the tasks of such missions may include ‘joint disarmament operations, humanitarian and rescue tasks, military advice and assistance tasks, conflict prevention and peace-keeping tasks, task of combat forces in crisis management, including peace-making and post-conflict stabilisation’ (Article 43(1) TEU). A new aspect is that such missions ‘may contribute to the fight against terrorism, including by supporting third countries in combating terrorism in their territories’. In the context of CSDP missions, it may be noted that pursuant to Article 42(5) TEU and Article 44 TEU, the Council may request a group of Member States to implement such a mission, in accordance with modalities to be agreed in consultation with the HR. This possibility has not been used to date. 1.3.2.2 Article 42(6) TEU and Article 46 TEU: PESCO The provisions on PESCO also take their origin in the draft constitutional treaty.43 While the Treaty of Nice had for the first time introduced the possibility of a reinforced cooperation in the CFSP, the domain of the CSDP was expressly excluded from its scope (ex-Article 27b TEU). The new PESCO provisions are thus a significant step forward in the CSDP. This recognizes the reality that in a Union of 28 Member States, not all of them have the same military capabilities and the will to engage them for the purpose of EU military operations.44 Article 46 TEU has to be read in conjunction with Protocol No 10 annexed to the ToL. While Article 46 TEU lays down the procedural aspects to establish PESCO, the criteria for participation and obligations to be fulfilled by the Member States wishing to take part in PESCO are set out in the Protocol. In essence, the latter defined the threshold for being eligible to participate in PESCO (Article 1 of Protocol No 10 – undertaking to develop defence capacities notably through participation in the EDA and have the capacity to supply rapid deployment battle groups) and imposes on Member States obligations to cooperate in the defence sector, in particular with a view to harmonization and interoperability (Article 2 of Protocol No 10). The EDA is mandated in Article 3 of Protocol No 10 to contribute to this process. As regards procedure, Article 46(1) TEU envisages that the Member States concerned must formally notify the Council as well as the HR of their intention to start PESCO. It may be noted that pursuant to paragraph 2, the Council adopts the formal decision on the establishment of PESCO, identifying the participating Member States,

43

See H Bribosia in Amato, Bribosia and De Witte (n 2) 840–846. Nonetheless, a total of 25 Member States decided to participate in the PESCO (cf. Article 2 of Council Decision 2017/2315 – above n 35). 44

34 Research handbook on the EU’s common foreign and security policy by qualified majority, which may seem surprising in a CSDP context.45 The Council acts after consultation with the HR. Following these procedural requirements, PESCO was launched in accordance with the mandate provided by the EUCO in December 2017.46 1.3.2.3 Enhanced cooperation in the CFSP Provisions on enhanced cooperation outside the CSDP have remained in place but have been moved from the CFSP chapter into the general provisions on enhanced cooperation in the TFEU (Article 329(2) TFEU). This can be seen as one illustration of the integration of the CFSP into the EU institutional framework. Article 329(2) TFEU merely sets out the procedure, and as for PESCO, the scope of the enhanced cooperation in the CFSP is not further defined.47 The only limitation is spelled out in Article 326(1) TFEU, which provides that any such cooperation must be in compliance with the Treaties and Union law. 1.4 The Choice of Legal Basis in the CFSP The legal bases allowing the Council to take decisions in the CFSP (including the CSDP) are the following: Article 28 TEU (operational actions), Article 29 TEU (positions), Article 33 TEU (EUSR), Article 37 TEU (international agreements), Articles 42(4) and 43(2) TEU (CSDP missions), Article 45 TEU (EDA) and Article 46 TEU (PESCO). The specific legal bases for the implementation of the CSDP did not exist before and the ToL thus allows a better delimitation between CFSP and CSDP actions. Nonetheless, their respective scope remains rather wide (as compared to legal bases in the TFEU), and the practice of the Council is not always consistent. For example, post-Lisbon the Council continued to base decisions providing EU support to actions carried out by third entities in the field of disarmament and non-proliferation on Article 26(2) TEU.48 Sometimes, however, the same type of decision was based on Article 28 TEU.49 In its more recent practice the Council now systematically bases such decisions on Article 28 TEU. This recent practice is legally correct as the purpose of Article 26 TEU is only to define the respective powers of the EUCO and the Council and Article 28 TEU remains the legal basis for all CFSP actions 45

This may be explained by the fact that by definition not all Member States participate in PESCO. The decisions subsequently taken by the Council to implement PESCO are, however, taken by unanimity (Article 46(6) TEU). 46 See Council Decision (CFSP) 2017/2315 of 11 December 2017 establishing permanent structured cooperation (PESCO) and determining the list of participating Member States, OJ 2017 L 331/57. 47 It is interesting to note (by contrast to the PESCO) that the decision to start a reinforced cooperation requires unanimity (Article 329(2), 2nd subparagraph TFEU). 48 E.g. Council Decision 2012/421/CFSP of 23 July 2012 on support for the Convention on Biological and Toxic Weapons, OJ 2012 L 196/61 (previously, such decisions were based on ex-Article 13.3). On EU action against the non-proliferation of weapons of mass destruction, see further Chapter 9 in this volume. 49 See Council Decision 2012/422/CFSP of 23 July 2012 on support for a nuclear-free Middle-East and Council Decision 2012/423/CFSP of 23 July 2012 on the non-proliferation of ballistic missiles, OJ 2012 L 196/64, 74.

Institutional framework, legal instruments and decision-making procedures 35 outside the CSDP. Even though Article 28 TEU refers to ‘operational action’ by the EU, it may be noted that Article 29 TEU (definition of ‘positions’ of the EU) is also used for measures which arguably have an operational character, i.e. measures imposing targeted sanctions on individuals and entities.50 Issues of delimitation have also arisen between Article 28 TEU (CFSP) and Articles 42(4) and 43(2) TEU (CSDP missions). The Council recently adopted a decision based on Article 28 on a ‘stabilisation action’ in Mali, involving the deployment of a team of advisers to assist the Malian authorities in strengthening their regional administration.51 As this action had certain features that made it comparable to a CSDP mission,52 there was some debate within the Council on whether Articles 42(4) and 43(2) TEU should not be used as lex specialis for this purpose. By contrast, the Council not long afterwards adopted a decision based on these latter legal bases in order to send a CSDP mission to Iraq with a very similar mandate.53 This may be seen as a reflection of the specificities of the CFSP, where the CJEU in principle has no jurisdiction (with the exception of restrictive measures targeting individuals and control of the limits set by Article 40 TEU). As a consequence, the CJEU has no control over the choice of legal basis by the Council, and the latter has a wider margin of discretion in this regard.54 1.5 The Financing of the CFSP The arrangements for the financing of the CFSP are laid down in Article 41 TEU. The basic principles have not changed since the Treaty of Amsterdam: the general rule is that all CFSP acts entailing expenditure (administrative and operating) are charged to the EU budget, with the exception of operating expenditure arising from operations having military or defence implications (i.e. EU military operations), and where the Council unanimously decides otherwise (Article 41(1) and 2 TEU). For EU military operations, their common costs are charged to the Member States through the so-called Athena mechanism, set up for this purpose by a Council decision (based on Articles 26(2) and 41(2) TEU).55 The ToL has added two new elements: the possibility for the Council to adopt decisions (a) establishing procedures for rapid access to the CFSP budget, in particular for preparatory measures (paragraph 3, subparagraph 1), and (b) the setting-up of a start-up fund (based on Member States’ contributions) for preparatory measures that are not funded from the EU budget (subparagraph 3) (the latter by qualified majority on 50 E.g. Council Decision (CFSP) 2016/849 of 27 May 2016 concerning restrictive measures against the Democratic People’s Republic of Korea and repealing Decision 2013/183/CFSP, OJ 2016 L 141/79. 51 Decision (CFSP) 2017/1425 of 4 August 2017, OJ 2017 L 204/90. 52 The EU also operates a civilian mission (EUCAP Sahel Mali, Council Decision 2014/219/CFSP of 15 April 2014, OJ 2014 L 113/21) and a military CSDP mission with an advisory and training mandate in Mali (EUTM Mali, Council Decision 2014/220/CFSP of 17 January 2013, OJ 2013 L 14/19). 53 Council Decision (CFSP) 2017/1869 of 16 October 2017, OJ 2017 L 266/12. 54 See also Chapter 4 in this volume on judicial control over the CFSP. 55 See note 36.

36 Research handbook on the EU’s common foreign and security policy proposal from the HR). It may be noted that if such a fund is set up, the High Representative would be empowered by the Council to make use of it (subparagraph 4). The Council has not availed itself of these possibilities to date.56 This may on the one hand reflect a lack of ambition on the part of the Member States, which could in principle request the HR to take an initiative in this regard. On the other hand, it may also be explained by a certain reluctance on the part of the Commission to agree to procedures for the financing of the CFSP outside the Financial Regulation. It is noteworthy in this context that the Commission has taken the initiative to propose the setting-up of a European Defence Fund to support the development of defence capabilities of the Member States.57 The activation of Article 42(3) TEU would seem coherent with and complementary to this Commission initiative, and in line with the implementation plan for the security and defence aspects of the EU’s Global Strategy.58

2. THE PRACTICE: A MORE EFFICIENT AND BETTER INTEGRATED CFSP? The first part of this chapter has provided an overview of the institutional actors, instruments and procedures of the CFSP/CSDP and the new institutional arrangements. Eight years after the entry into force of the ToL, three mains question arise: (a) whether they have made the EU’s CFSP more efficient and whether their full potential has been used; (b) whether they have succeeded in better integrating the CFSP in the ‘mainstream’ of the EU’s external action; and (c) what role the Member States continue to play in the CFSP. Providing an exhaustive answer to these questions would exceed the scope of this chapter and is also objectively difficult due to the nature of the CFSP as an ad hoc policy. Only a few examples will therefore be addressed. 2.1 The HR and the EEAS The creation of the function of HR and the EEAS as her support service constitute a historic step forward: in matters of foreign policy, the EU is now represented by one single person appointed for five years and not by a different foreign minister every six months. This clearly makes relations with third partners easier and allows for more continuity. The EEAS itself, being the first proper diplomatic service of the EU, also represents a major advancement, in particular since it supports the HR not only in her foreign policy mandate, but also as VP of the Commission and chair of the FAC. The 56

At present, preparatory measures in the CFSP may be financed on the basis of the Financial Regulation (Article 54(2)(c)). 57 Communication from the Commission to the European Parliament, the European Council, the Council, the European Economic and Social Council and the Committee of the Regions on a European Defence Action Plan COM 2016 (950) final. 58 cf. the Implementation Plan on Security and Defence presented by the HR to the Council (Council doc. 14392/16, 14 November 2016, Annex, p. 6, point 11) and the Council Conclusions of 14 November 2016 on implementing the EU’s Global Strategy in the area of security and defence (doc. 14149/16, 14 November 2016, Annex) endorsing the implementation plan.

Institutional framework, legal instruments and decision-making procedures 37 fact that its personnel is partly composed of diplomats from the Member States provides it with relevant diplomatic experience; conversely, the national diplomats present in the EEAS benefit from the experience of long-serving EU officials and gain an understanding of the EU’s institutional functioning. This is mutually beneficial and potentially conducive to more coherence in the EU’s and Member States’ action on the international scene, in accordance with Article 24(2) and (3) TEU. In the same vein, the 140 delegations of the EU which form part of the EEAS and which are placed under the authority of the HR represent an essential tool for the EU’s external action and diplomacy. Whereas the former Commission delegations were essentially involved in the implementation of the EU’s cooperation and development programmes, the new EU delegations not only cover this latter aspect, but are proper diplomatic representations of the EU covering the entire spectrum of economic and political relations with third countries. As they also comprise Commission officials who may receive instructions from their services (albeit under the control of the Head of Delegation, who reports to the HR), they continue to serve for the implementation of EU development policies.59 EU delegations play an equally important role in representing the EU at international and multilateral organizations, in particular at the United Nations. The ToL places particular emphasis on the cooperation between the EU and the UN (cf. Article 21(1), 2nd subparagraph TEU and Article 220 TFEU). It is noteworthy in this regard that following the entry into force of the ToL, the EU succeeded in obtaining an enhanced observer status at the UN General Assembly, allowing the EU to participate in the debates with the same priority as representatives of other groups, notwithstanding its observer status.60 This also allows the EU (for the first time) to intervene in the annual General Debate of the UN General Assembly (UNGA), where EU statements have been delivered by the President of the EUCO since 2012. Furthermore, the ToL has clearly enhanced the range of instruments available for EU action in the CSDP, and the EUCO has requested the competent institutional actors to initiate the necessary actions in implementation of the security and defence aspects of the EU’s Global Strategy.61 The creation of the MPCC is one of the visible results. Another significant development is PESCO, established by the Council in December 201762 in accordance with the mandate from the EUCO.63 As indicated by the HR at the informal meeting of defence ministers on 6–7 September 2017 in Tallinn, PESCO should in substance be ‘output-oriented’ and focus on specific projects submitted by the 59

The Financial Regulation (FR) has been amended to allow a Head of Department to exercise the functions of Authorising Officer by Sub-Delegation to implement Commission development assistance fund (cf. Article 56(2) FR). 60 On the drafting history of this resolution see P Serrano de Haro, ‘Participation of the European Union in the Work of the United Nations: General Assembly Resolution 65/276’, CLEER Working Paper 2014/4. 61 cf. EUCO conclusions of 15 December 2016, doc. EUCO 34/16, para 11. 62 See above n 35. 63 See EUCO conclusions of 22/23 June 2017 (doc. EUCO 8/17, para 8) and of 19 October 2017 (doc. EUCO 14/17, para 13) and Council Decision (CFSP) 2017/2315 of 11 December 2017 establishing permanent structured cooperation (PESCO) and determining the list of participating Member States, OJ 2017 L 331/57.

38 Research handbook on the EU’s common foreign and security policy participating Member States. The PESCO Decision itself does not identify specific projects, but only contains a list of general ‘binding common commitments’ of the participating Member States in line with the PESCO Protocol (No. 10) (Annex to the Decision). Article 4(2)(e) of the PESCO Decision provides that the list of specific projects will be established on the basis of further decisions or recommendations of the Council.64 A number of concrete implementation measures have already been taken: the Council adopted a decision establishing the list of projects as well as a recommendation setting out a roadmap for the implementation of PESCO in March 2018,65 and a decision regarding PESCO governance rules in June 2018.66 An evaluation of the concrete impact of EU CSDP missions and operations would go beyond the scope of this chapter as many EU missions or operations are still ongoing and also cover a number of different mandates.67 However, the fact alone that the EU has given itself the capability to establish such missions or operations that cover a wide range of tasks (civilian missions covering advice, training and capacity building for the police, so-called ‘rule of law’ missions, non-executive military missions providing advice and capacity building for the military, border assistance and monitoring missions, as well as military operations with executive powers, both on land and at sea (operations EUNAVFOR Atalanta and EUNAVFOR MED/Sophia))68 demonstrates that the new instruments and procedures that were gradually put in place following the Treaty of Maastricht allow the EU to be a global actor in foreign policy and security matters, in fulfilment of the EU’s objective to contribute to the maintenance of international peace and security (Article 21(2)(c) TEU). One concrete illustration of the impact of EU missions and operations is that they have fostered EU–UN cooperation in this domain: the UN has in recent years on several occasions requested the EU to provide support through its own crisis management operations and missions and acknowledged EU contributions to international efforts in the maintenance of international peace and security.69 The EU civilian mission in Kosovo (EULEX Kosovo),70 the largest civilian mission until now, is another illustration of CSDP instruments used for the purpose of capacity building in the field of the rule of law, another stated objective of EU external action (Article 21(2)(b) TEU). 64

Pursuant to Article 5 of the PESCO Decision, the HR may also make recommendations regarding the identification and evaluation of individual projects, based on contributions by the EEAS and the EDA. 65 Council Decision (CFSP) 2018/340 of 6 March 2018, OJ 2018 L 65/24; Council Recommendation of 6 March 2018 (2018/C 88/01), OJ 2018 C 88/1. 66 Council Decision (CFSP) 2018/909 of 25 June 2018, OJ 2018 L 161/37. 67 For a detailed overview of EU civilian CSDP missions, see Chapter 5 in this volume. 68 On EU military operations see further Chapter 6 in this volume. 69 E.g. in the framework of the fight against piracy off the coast of Somalia, where the EU is conducting operation EUNAVFOR Atalanta (cf. UN Security Council Resolution 2125 (2013) of 18 November 2013, preambular paragraphs 14 and 17). The EU and the UN are currently finalizing an agreement on mutual cooperation in crisis management operations, which would be the first of its kind. 70 Council Joint Action 2008/124/CFSP of 4 February 2008, OJ 2008 L 42/92 (as amended).

Institutional framework, legal instruments and decision-making procedures 39 The EU, through its new institutional actors, first and foremost through the HR, has also increasingly been able to play a role as facilitator in sensitive political processes in the framework of the UN. Two prominent examples are the dialogue between Serbia and Kosovo concerning a normalization of relations following the declaration of independence by Kosovo, and the negotiations between the five permanent members of the UN Security Council (UNSC) and Germany and Iran, which culminated in the agreement reached in Vienna in July 2015 on the Joint Comprehensive Plan of Action (JCPOA), subsequently endorsed by UNSC Resolution 2231 (2015).71 As regards the instruments and procedures of the CFSP, the ToL has not introduced major innovations. QMV, a possibility that existed before, has not been used to date. It is noteworthy in this context that the Commission recently took the initiative to suggest to the European Council to use this possibility more often.72 The Member States thus appear to continue to treat the CFSP as an intergovernmental process subject to consensus, despite its integration in the realm of the EU’s external action. 2.2 The HR as VP of the Commission One very significant role of the HR is that of being one of the VPs of the Commission, in charge of the external relations of the EU outside the CFSP and for the coordination of all EU policies in the field of external relations (Article 18(4) TEU). This is the function previously exercised by the Commissioner for external relations. The current HR, Federica Mogherini, has reactivated the Commissioner’s group on external action (CGEA), which is chaired by her and normally meets once per month. This group brings together the Commissioners with external portfolios (trade, development cooperation, humanitarian affairs), as well as other Commissioners on an ad hoc basis when external dimensions of internal policies are at stake (e.g. when the group discussed EU action to take with regard to the migration crisis). The CGEA is a concrete manifestation of the so-called ‘double-hat’ of the HR and, even though it has no formal decision-making power, has developed into a useful mechanism on the side of the Commission to ensure a coordinated use of the EU’s external relations instruments.73 Regarding the question of whether the CFSP is better integrated and coordinated with the ‘traditional’ external relations instruments, the picture is a rather mixed one. On the one hand, the ‘double-hat’ of the HR as VP of the Commission has had a number of positive effects. The CGEA referred to above is one of them. The negotiation of international agreements is another one: for comprehensive international 71 For an analysis of the JCPOA and the EU’s role in facilitating the agreement, see S Blockmans and A Viaud, ‘EU Diplomacy and the Iran Nuclear Deal: Staying Power?’, CEPS Policy Insights No 2017-28; see also Chapter 9 in this volume. 72 Communication from the Commission to the European Council, the European Parliament and the Council of 12 September 2018 (COM (2018) 674 final) (‘A stronger global actor: a more efficient decision-making for EU Common Foreign and Security Policy’). The question may be raised in this regard whether this initiative of the Commission is in line with Article 30(1) TEU and the prerogatives of the HR in the CFSP. 73 See further S Blockmans and S Russack, ‘The Commissioner’s Group on External Action – Key Political Facilitator’, CEPS Special Report No 125/2015.

40 Research handbook on the EU’s common foreign and security policy agreements to be concluded with third partners, i.e. agreements covering a substantial CFSP part as well as the ‘classic’ domains of external relations (trade, cooperation, development), the EEAS normally acts as chief negotiator for the EU.74 This follows from the fact that the HR is mandated by the Council to negotiate the CFSP part, the Commission for the remainder of the subject-matters, and acts on behalf of the Commission in her capacity as VP. This is a situation where the double-hat works well in practice and where the EEAS exercises its role as supporting the HR in both these capacities. The HR and the Commission have also regularly made use of their joint right of initiative as generally provided in Article 22(2) TEU: a number of joint HR/Commission ‘communications’ to the Council and the EP have been made since the ToL’s entry into force, covering both CFSP and other aspects of EU external competences. Such joint communications are elaborated and drafted in close cooperation between the EEAS and the competent Commission services.75 In the area of sanctions, where the EU has a long-standing policy of implementing UN sanctions but is also imposing so-called ‘autonomous’ sanctions,76 the new arrangements are also conducive to more efficient decision making. In this domain, the Treaties establish a direct link between the institutional prerogatives of the HR and the Commission, in that Article 215 TFEU provides for joint HR/Commission proposals for the regulations implementing the CFSP decisions adopted by the Council on proposal from the HR. As mentioned above, the competent Commission services are co-located with the EEAS and work hand in hand with the EEAS’ sanctions policy division, ensuring a coordinated approach vis-à-vis the Council. On the other hand, the post-Lisbon institutional practice has shown a certain emergence of turf wars about respective institutional powers. On the side of the Commission, there is a growing tendency to emphasize its financial responsibilities in the CFSP to the detriment of the right of initiative of the HR. This may be explained by the loss of institutional power of the Commission in the CFSP (the Commission no longer being formally ‘fully associated’ in the CFSP). However, this loss should in principle be ‘compensated’ by the fact that the HR is also a member of the Commission; in practice, the latter’s services are closely associated with the preparation of HR initiatives or proposals in the CFSP, and the Commission service directly involved in the CFSP on account of the Commission’s budgetary responsibilities is placed under the authority of the HR in her capacity as VP, and is collocated with the EEAS. 74 To this effect, working arrangements were agreed between the EEAS and the Commission in 2012. The EEAS currently leads negotiations for Association Agreements, Partnership and Cooperation Agreements or similar comprehensive agreements inter alia with Azerbaijan, Chile, Japan, Mercosur, Mexico; negotiations have recently been closed with Armenia and Australia. 75 See, e.g., Joint Communication from the HR and the Commission to the European Parliament and the Council on the EU’s Comprehensive Approach to external conflicts and crises JOIN (2013) 30 final; Joint Communication from the HR and the Commission to the European Parliament and the Council on cybersecurity of the EU, doc. JOIN (2017) 450 final. 76 See the EU Guidelines on implementation and evaluation of restrictive measures (sanctions) in the framework of the EU Common Foreign and Security Policy .

Institutional framework, legal instruments and decision-making procedures 41 The so-called ‘CBSD file’ (capacity building for security and development) illustrates some of these institutional difficulties. While the recent amendment of the IcSP77 would under certain exceptional circumstances cover EU support measures for the benefit of military forces in third countries, it is clear that such measures would not cover the entire range of support needed and that parallel complementary measures would be necessary on the CFSP side. However, the competent Commission services have actively resisted any possible initiative of the HR in this regard (despite requests from the EUCO to the HR),78 on the basis of a rather restrictive interpretation of the limitations imposed by Article 41(2) TEU as regards the use of the CFSP budget for military activities, and on account of the responsibilities of the Commission for the implementation of the EU budget (Article 317 TFEU). The implementation of the so-called ‘integrated approach’79 in the EU’s response to international crises has thus not yet fully materialized. This is still largely due to the fact that CFSP decisionmaking procedures remain in the hands of the Council and PSC and thus the Member States, whereas the Commission has a rather wide measure of discretion in deciding on the priorities in the implementation of EU assistance and cooperation instruments. As a result, on the ground, it may sometimes be difficult to perceive a coherent approach between different EU actors and instruments, even within the CFSP. Hence, while it can be observed that the HR definitely has used her role as VP to the benefit of more coordination between different non-CFSP EU instruments, her role in ensuring more coherence between CFSP instruments, in particular CSDP missions, and other external assistance instruments, has been less clear. 2.3 The Member States in the CFSP One particularity of the CFSP is that its scope is potentially very wide as it may cover ‘all areas of foreign policy and all questions relating to the Union’s security’ (Article 24(1) 1st subparagraph TEU). This can be explained by the nature of foreign policy, which is governed by often unpredictable international developments (cf. Article 28(1) TEU: ‘Where the international situation requires operational action by the Union’). The CFSP is further defined – for the first time – as a ‘competence’ of the EU (Article 24(1) TEU and Article 2(4) TFEU); this confirms that it is a policy that ‘belongs’ to the EU and is separate from the national foreign policies of the Member States. The CFSP (despite the term ‘common’) is thus not a mere intergovernmental process between the

77

Regulation (EU) 2017/2306 of the European Parliament and of the Council of 12 December 2017 amending Regulation (EU) No 230/2014 establishing an instrument contributing to stability and peace COM, OJ L 335, 15.12.2017, p. 6. 78 EUCO conclusions of 15 December 2016 (doc. EUCO 34/16), paragraph 11. 79 This concept was introduced by the EU’s Global Strategy on Foreign and Security Policy (cf. part 3.3: ‘An integrated approach to conflicts and crises’). In substance, this concept is largely identical to the previous concept of the ‘comprehensive approach’ of the EU in relation to external conflicts and crises (cf. Council Conclusions of 12 May 2014, welcoming the Joint HR/Commission communication setting out the approach, doc. 9542/14, 17–21).

42 Research handbook on the EU’s common foreign and security policy Member States, as is sometimes still held in academic literature even after the ToL.80 This is confirmed by the obligation of loyal cooperation on the part of the Member States, which must ‘support the Union’s external and security policy actively and unreservedly in a spirit of loyalty and mutual solidarity’ (Article 24(3) 1st paragraph TEU). This raises the question of the relation between the CFSP and the Member States’ national foreign policies. Clearly, due to the specific nature of the CFSP, the principle of pre-emption cannot apply, i.e. the Member States do not lose their sovereign power to implement their national foreign policies when the EU defines its CFSP. The CFSP is not a case of shared competence. Rather, the respective competences of the EU and Member States in matters of foreign policy can be qualified as parallel competences, the actions of the Member States only being constrained by their duty of loyalty.81 However, this assessment remains rather theoretical. In practice, the CFSP still operates as if it were an intergovernmental process controlled by the Member States. As seen above, the reality until now is that the possibility of QMV in the CFSP has never been used. Furthermore, in line with the prerogatives of the EC, the major strategic decisions in the CFSP are taken at the level of the Heads of State and government by the EUCO, which acts by consensus unless otherwise provided in the Treaties (Article 15(4) TEU). This was illustrated, for instance, in the case of the sanctions imposed against Russia and Ukraine in 2016 in relation to the situation in Ukraine and the Russian annexation of Crimea, where the EC instructed the HR (EEAS) and the Commission several times to submit proposals for legal acts to the Council.82 However, this is not entirely new as the ToL has not changed the role of the EUCO in the CFSP. As stated at the beginning, major developments in the CFSP, and in particular the CSDP, find their origin in strategic decisions taken by the EUCO, and were subsequently reflected in treaty amendments. Thus, despite the existence of the EU’s CFSP, Member States will continue to take foreign policy initiatives on their own, sometimes in coordination with a few partners, even though they could submit initiatives directly to the Council in order for the EU to take action. This is certainly due to the fact that defining an EU policy on a particular matter requires a decision by unanimity (or even consensus at the level of the EC). This is not contrary to the letter and spirit of the TEU: many important developments in the CFSP/CSDP took place upon prior concerted initiatives of a small group of Member States.83 The recent Franco-German initiative on PESCO is one prominent illustration of this process and the role Member States play in the development of the CFSP. This is also in line with the right of any Member State to submit an initiative to the Council (Article 30(1) TEU). In some cases, some Member States have also operated entirely outside the EU framework, e.g. in relation to the situation in Ukraine and the 80 E.g. P Eeckhout, EU External Relations Law (OUP 2011) 166–168; D Thym, ‘The Intergovernmental Constitution of the EU’s Foreign, Security and Defence Executive’ (2011) 7(3) European Constitutional Law Review 466. 81 S Marquardt and J-C Gaedtke, ‘Vor Art. 23–46 EUV’, in H von der Groeben, J Schwarze and A Hatje (eds), Europäische Unionsrecht (Nomos 2015) para 10. 82 See, e.g., EC conclusions of 16 July 2014 (doc. EUCO 147/14), para 6. 83 For instance, a major impetus for the development of the CSDP emanated from a bilateral meeting between France and the UK in 1998 in St Malo.

Institutional framework, legal instruments and decision-making procedures 43 implementation of the Minsk agreement, where two EU Member States (Germany and France) had regular meetings with the Russian Federation and Ukraine in the so-called Normandy format. However, as long as such actions do not contradict or negatively affect EU measures taken in relation to the same issue, they may in the end prove to be complementary. In the end, the EU does not have any ‘exclusivity’ in dealing with regional or global foreign policy and security issues. Finally, there have also been instances of ‘disloyal’ behaviour by Member States, for example in relation to EU statements to be delivered at international organizations where, despite EU positions agreed in Brussels, some Member States on the ground broke ranks and prevented the EU from expressing a unified position (as was the case following the entry into force of the ToL in New York and more recently in the Human Rights Council in Geneva).84 In such instances, in the absence of any competence of the CJEU to examine possible violations of the Member States’ obligations of loyalty, it remains for the HR to address such difficulties in an informal manner.85

3. CONCLUSION The ToL clearly reflects a high level of ambition as regards the development of the CFSP and CSDP. It has created the HR and the EEAS, and this major innovation has clearly increased the presence and visibility of the EU on the international scene. The HR has a wide range of tools at her disposal to make the EU’s CFSP effective. However, it may be questioned whether it is possible for one single person, despite the support of the EEAS, to exercise the multiple responsibilities linked to this function in a satisfactory manner.86 The CFSP and CSDP still remain subject to separate procedures and to the limits imposed by Article 40 TEU. Ensuring consistency between the CFSP and other external policies of the EU, as mandated by Article 21.3 TEU, therefore remains a challenge. Furthermore, a number of tools and procedures destined to make the CFSP more efficient, such as QMV, already existed before but have never been used. Resorting to the whole range of instruments and procedures provided by the ToL remains in the hands, and is ultimately subject to the political will, of the Member States. Some recent developments, such as those in relation to PESCO, however, appear to indicate that a new impetus will be given to the EU’s CFSP, in particular the CSDP, even if the unanimity rule continues to apply in this latter domain.

84 While the concerns raised in New York by one Member State related to the horizontal issues of delimitation of competences between the EU and its Member States in matters of foreign policy and their external representation, the more recent instances of disunity in Geneva were linked to diverging positions on the substance of the matter, notably regarding Palestine. 85 The issue of the lack of unity among Member States in international fora was raised by the HR at the so-called Gymnich meeting (regular informal gatherings of foreign ministers) on 7–8 September 2017. 86 Some early commentators were outright sceptical in this regard: see S Blockmans and M-L Laatsit, ‘The European External Action Service: Enhancing Coherence in EU External Action?’ in J Cardwell (ed.), EU External Relations Law and Policy in the Post-Lisbon Era (TMC Asser Press 2012) 140: ‘mission impossible’.

3. Representing the EU in the area of CFSP: legal and political dynamics Hylke Dijkstra and Peter Van Elsuwege

1. INTRODUCTION When Serbia and Kosovo started a dialogue to improve bilateral cooperation in March 2011, the EU’s High Representative for Foreign Affairs and Security Policy together with the officials in the European External Action Service (EEAS) were tasked with facilitating the process.1 When the crisis in Ukraine escalated in February 2014, following the earlier Maidan protests, the foreign ministers of France, Germany and Poland flew jointly to Kiev to mediate between the government and the opposition on behalf of the EU.2 When the Joint Comprehensive Plan of Action concerning the nuclear programme of Iran was reached in July 2015, the signatories included Iran, the permanent members of the UN Security Council, Germany and the EU.3 The EU is therefore represented on the international stage by different actors and different actor constellations. This applies not only to highly visible ad hoc attempts at conflict resolution, but also to permanent forms of diplomacy. EU representation, for example, varies across international organizations.4 Within international organizations, it can vary even across the different committees.5 Similarly, the upgrading of the approximately 140 EU delegations after the Treaty of Lisbon (2009) has taken place unevenly. The result is that these EU delegations perform very different functions depending on their location. The EU delegation in Baku is not the same as the one in 1 J Bergmann and A Niemann, ‘Mediating International Conflicts: The European Union as an Effective Peacemaker?’ (2015) 53 Journal of Common Market Studies 957. 2 Federal Foreign Office of Germany, ‘Crisis diplomacy in Ukraine’ (21 February 2014) accessed 4 January 2016. 3 Joint Comprehensive Plan of Action (14 July 2015) accessed 2 March 2017; see also Chapter 9 in this volume. 4 E.g. K Verlin Laatikainen and KE Smith (eds), The European Union at the United Nations: Intersecting Multilateralisms (Palgrave Macmillan 2006); KE Jørgensen (ed.), The European Union and International Organizations (Routledge 2009); KE Jørgensen, S Oberthür and J Shahin (eds), ‘The Performance of the EU in International Institutions’ (2011) 33 Journal of European Integration 599; KE Jørgensen and K Verlin Laatikainen (eds), Routledge Handbook on the European Union and International Institutions: Performance, Policy, Power (Routledge 2013). 5 E.g. within the UN General Assembly, see K Verlin Laatikainen, ‘The EU Delegation in New York: A Debut of High Political Drama’ in D Spence and J Batora (eds), The European External Action Service: European Diplomacy Post-Westphalia (Palgrave Macmillan 2015) tables 10.1 and 10.2.

44

Representing the EU in the area of CFSP 45 Beijing.6 EU external representation thus varies significantly and it is not immediately clear why. This chapter attempts to understand how the EU representation in the field of Common Foreign and Security Policy (CFSP) is organized and applied in practice. This question is hardly new. The issue of EU external representation has been at the heart of policy discussions about European foreign policy cooperation since at least the 1970s. And the academic literature discusses inter-institutional politics and the consequences for EU external representation.7 This chapter makes three contributions. First, it seeks to provide a comprehensive overview of EU external representation in the field of CFSP, whereas other publications focus on other areas or specific cases. Second, it approaches the question from a distinctly legal-political perspective, where others have stayed within disciplinary boundaries. Third, it focuses on the Treaty of Lisbon and the post-Lisbon era, whereas most of the scholarship is concerned with the pre-Lisbon institutional set-up of external representation. This chapter starts by discussing EU external representation from a conceptual perspective. It outlines various legal and political dynamics behind the research question who represents the EU. It continues by analysing three instances of external representation: first, multilateral representation of the EU in international organizations and fora; second, bilateral representation of the EU in other countries; third, ad hoc representation and mediation attempts by the EU in international and regional conflicts. The conclusion sheds further light on the interaction between law and politics in understanding EU external representation.

6 R Balfour and K Raik, Equipping the European Union for the 21st Century: National Diplomacies, the European External Action Service and the Making of EU Foreign Policy (Finnish Institute of International Affairs 2013); F Austermann, European Union Delegations in EU Foreign Policy: A Diplomatic Service of Different Speeds (Palgrave Macmillan 2015); Spence and Batora (n 5). 7 E.g. D Allen, ‘“Who Speaks for Europe?”: The Search for an Effective and Coherent External Policy’ in J Peterson and H Sjursen (eds), A Common Foreign Policy for Europe? Competing visions of the CFSP (Routledge 1998); S Meunier and K Nicolaidis, ‘Who Speaks for Europe? The Delegation of Trade Authority in the EU’ (1999) 37 Journal of Common Market Studies 477; I Govaere, J Capiau and A Vermeersch, ‘In-Between Seats: The Participation of the European Union in International Organisations’ [2004] European Foreign Affairs Review 155; H Dijkstra, ‘EU External Representation in Conflict Resolution: When Does the Presidency or the High Representative Speak for Europe?’ (2011) 15 European Integration online Papers (EIoP) accessed 24 February 2017; R Bengtsson and D Allen, ‘Exploring a Triangular Drama: The High Representative, the Council Presidency and the Commission’ in G Mueller-Brandeck-Bocquet and C Rueger (eds), The High Representative for the EU Foreign and Security Policy: Review and Prospects (Nomos 2011); E da Conceição-Heldt and S Meunier, ‘Speaking With a Single Voice: Internal Cohesiveness and External Effectiveness of the EU in Global Governance’ (2014) 21 Journal of European Public Policy 961; C Kaddous (ed.), The European Union in International Organisations and Global Governance (Hart Publishing 2015).

46 Research handbook on the EU’s common foreign and security policy

2. EU EXTERNAL REPRESENTATION IN LAW AND POLITICS EU external representation can be studied through the disciplinary lenses of law and politics. The EU Treaties contain various legal provisions determining which actor represents the EU on which issue. Similarly, from a political perspective, one can examine the leadership of actors in EU external representation, their effectiveness and instances of bureaucratic politics. Rather than analysing legal rules and political dynamics separately, however, we study how they interact. The legal provisions in the Treaties are the result of political choices and many legal rules are codifications of previous norms and practices. Furthermore, political actors tend to use legal provisions to their advantage. This conceptual section starts with some historical background before outlining key legal aspects of the Treaty of Lisbon. It continues by pointing out how political science theories complement legal insights. 2.1 Development of EU External Representation Prior to the Lisbon Treaty The first instances of external representation pre-date the establishment of the European Economic Community (EEC) with the Treaty of Rome in 1958. Jean Monnet, as President of the High Authority of the European Coal and Steel Community (ECSC), was keenly aware of the importance of transatlantic relations and opened an ECSC information office in Washington, DC in 1954.8 Over the next 50 years, the European Commission continued this practice and opened delegations around the world. Diplomatic relations were initially established with important trading partners and countries receiving development assistance. This reflected the EEC’s competence in external commercial policy as well as the Commission’s role in development cooperation following the Lomé agreements of the 1970s. By the time that the Treaty of Lisbon entered into force in 2009, the Commission had more than 130 delegations in non-member countries and international organizations.9 In parallel to the external representation of the EEC, there was a need to represent European Political Cooperation (EPC) established in 1970. The administration of EPC had been delegated to the six-monthly rotating Presidency. It was put in charge of external representation with the London Report of 1981.10 This arrangement was formally confirmed with the Single European Act (SEA).11 With the Treaty of Maastricht of 1993, EPC transformed into the CFSP as a separate ‘pillar’ of the EU. 8

European Commission, Taking Europe to the World: 50 Years of the European Commission’s External Service (Office for Official Publications of the European Communities 2004). 9 M Bruter, ‘Diplomacy Without a State: The External Delegations of the European Commission’ (1999) 6 Journal of European Public Policy 183; D Spence, ‘The European Commission’s External Service’ (2004) 19 Public Policy and Administration 61; European Commission (n 8); C Carta, The European Union Diplomatic Service: Ideas, Preferences and Identities (Routledge 2012); E Drieskens, ‘What’s in a Name? Challenges to the Creation of EU Delegations’ (2012) 7 The Hague Journal of Diplomacy 51; Austermann (n 6). 10 S Nuttall, European Political Cooperation (Clarendon Press 1992); S Nuttall, European Foreign Policy (OUP 2000); ME Smith, Europe’s Foreign and Security Policy: The Institutionalization of Cooperation (CUP 2004). 11 Single European Act [1987] OJ L169, Art. 30.

Representing the EU in the area of CFSP 47 Concomitantly, a purely intergovernmental system of cooperation became part of a legal system established by Treaty law.12 This was also visible with respect to external representation. Article J.5 of the Maastricht Treaty on European Union (TEU) not only articulated the role of the rotating Presidency and the supportive role of the incoming and preceding Presidencies under the so-called ‘Troika’ formula, but also explicitly provided that ‘the Commission shall be fully associated in these tasks’.13 Every subsequent amendment of the Treaties strengthened the incorporation of the CFSP in the EU legal order, while retaining its specific legal rules and institutional mechanisms. The Treaty of Amsterdam of 1999 significantly developed the specific CFSP toolbox through the introduction of new instruments such as Common Strategies, Common Positions and Common Actions. Moreover, the revised Article 11 TEU clarified that the CFSP was no longer to be defined and implemented by ‘the Union and its member states’ but by the Union alone.14 This also implied significant amendments to the external representation of the CFSP. Even though the Council Presidency retained its formal representative function in this area, it was to be assisted by the SecretaryGeneral of the Council, who exercised the new function of ‘High Representative for the Common Foreign and Security Policy’.15 It is no secret that the first High Representative, Javier Solana (1999–2009), often overshadowed the foreign ministers of the Presidency.16 He was assisted by a Policy Unit as well as the officials of the Council Secretariat. The Council could also appoint Special Representatives with a mandate in relation to particular policy issues.17 Last but not least, the EU acquired the capacity to conclude international agreements in the field of CFSP, leading to a (largely academic) debate about the existence of the Union’s international legal personality besides that of the European Community, which was explicitly foreseen in the EC Treaty.18 This discussion had already lost much of its practical relevance after the Treaty of Nice clarified that international agreements in the area of CFSP ‘shall be binding on the institutions of the Union’. The abolition of the pillar structure and the express grant of a single legal personality to the Union with 12

See: R Gosalbo Bono, ‘Some Reflections on the CFSP Legal Order’ (2016) 43 CML Rev

341. 13 PJ Kuijper, J Wouters, F Hoffmeister, G De Baere and T Ramopoulos, The Law of EU External Relations: Cases, Materials and Commentary on the EU as an International Legal Actor (OUP 2013) 25. 14 As argued by Bono, ‘the deletion of the reference to Member States meant that the Treaty of Amsterdam discarded this intergovernmental trait of the CFSP introduced by the Maastricht Treaty’ (n 12) 348. 15 Art. 18(3) TEU (Amsterdam version); H Dijkstra, ‘The Council Secretariat’s Role in the Common Foreign and Security Policy’ (2008) 13 European Foreign Affairs Review 149. 16 Dijkstra (n 7). 17 Art. 18(5) TEU (Amsterdam version). 18 With regard to this academic discussion, see e.g. U Everling, ‘Reflections on the Structure of the European Union’ (1992) 29 CML Rev 1053; A von Bogdandy, ‘The Legal Case for Unity: The European Union as a Single Organization with a Single Legal System’ (1999) CML Rev 887; RA Wessel, ‘Revisiting the International Legal Status of the EU’ (2000) 5 European Foreign Affairs Review 507; R Leal-Arcas, ‘EU Legal Personality in Foreign Policy?’ (2006) 24 Boston University International Law Journal 165.

48 Research handbook on the EU’s common foreign and security policy the Treaty of Lisbon finally put an end to all possible speculations.19 It is now crystal clear that the CFSP is an integral part of the autonomous EU legal order and not a separate framework of cooperation subject to general rules of international law. 2.2 The Treaty Framework of EU External Representation The Lisbon Treaty aimed to enhance the effectiveness and coherence of the Union’s external action.20 The dissolution of the pillar structure (Art. 1 TEU), the introduction of a single legal personality for the Union (Art. 47 TEU) and the grouping of all external action principles and objectives – including those of the CFSP – in a single Chapter21 and a specific Treaty provision (Art. 21 TEU) give the impression of a fully integrated legal framework.22 At the institutional level, the new functions of a permanent European Council President (Art. 15 TEU), a double-hatted High Representative and Vice-President of the European Commission responsible for the consistency of external activities (Art. 18 TEU), assisted by the newly established EEAS (Art. 27 TEU) and Union delegations abroad (Art. 221 TFEU), confirm this picture. However, upon closer inspection, it is clear that ‘the proclaimed unity of EU external action only extends to its general principles and objectives’ whereas the EU’s competence in the field of CFSP remains clearly distinct from other categories of competence set out in the TFEU.23 Also in the post-Lisbon era, the CFSP remains ‘subject to specific rules and procedures’ (Art. 24 TEU) implying a predominant role for the institutions which are assimilated with the executive power (European Council and Council) and a limited possibility of judicial review.24 The distinctiveness of the CFSP is made explicit in the so-called mutual non-affect clause of Article 40 TEU. According to this provision, the implementation of the CFSP shall not affect the application of the procedures and the powers of the institutions laid down by the TFEU and vice versa. However, in the absence of clear criteria to define the scope of CFSP in relation to other fields of the Union’s external action, there is significant potential for inter-institutional conflicts.25 The ‘integration-delimitation’ paradox is also reflected with respect to the EU’s external representation. On the one hand, the role of the High Representative and the EEAS is to bridge the gap between CFSP and the other policy areas in an attempt to 19

Arts 1 and 47 TEU (Lisbon version). See e.g. the Commission’s statement on the occasion of the signature of the Lisbon Treaty, claiming that this new legal framework ‘will enhance efficiency and give the Union a single voice in external relations’ (IP/07/1922, 13 December 2007). For a critical analysis, see P Koutrakos, ‘Primary Law and Policy in EU External Relations: Moving Away From the Big Picture’ (2008) 33 EL Rev 666. 21 Chapter 1 of Title V of the TEU. 22 See also Chapter 1 in this volume. 23 Koutrakos (n 20) 669. 24 C Hillion, ‘A Powerless Court? The European Court of Justice and the Common Foreign and Security Policy’ in M Cremona and A Thies (eds), The European Court of Justice and External Relations Law: Constitutional Challenges (Hart Publishing 2014) 47. 25 P Van Elsuwege, ‘The Potential for Inter-Institutional Conflicts Before the Court of Justice: Impact of the Lisbon Treaty’ in Cremona and Thies (n 24) 123. 20

Representing the EU in the area of CFSP 49 improve the coherence of the EU’s external action. On the other hand, the limits of this integrated approach become obvious in the continued distinction of external representation tasks. The President of the European Council and the High Representative represent the Union for CFSP issues at their respective political levels whereas the Commission ensures external representation for the other policy areas.26 A similar institutional division between CFSP and non-CFSP external action is discernible with regard to the procedure for the negotiation and conclusion of international agreements and the adoption of Union positions in bodies set up by such agreements.27 Depending on the subject of the agreement (or position), the Commission or the High Representative can take the initiative. Reflecting the internal decisionmaking procedures, the role of the European Parliament in the procedure for concluding international agreements in the field of CFSP is limited to a right of information.28 In other words, the rules regarding the EU’s external representation are concomitant with the internal division of powers and the constitutional principles on which the EU legal order is based. This implies, on the one hand, respect for the principles of conferral and institutional balance, implying that the Union and its institutions can only act within the limits of the competences conferred upon them in the Treaties,29 and, on the other hand, respect for the interconnected principles of sincere cooperation and consistency (or coherence), which mitigate (at least partially) the complexities following from the division of competences.30 2.3 The Law and Politics of EU External Representation in Practice As the previous section shows, an analysis of the Treaty rules is not sufficient to understand the practice of the EU’s external representation. It is equally important to outline some of the political dynamics at stake. The first consideration is that of institutional choice. With the Lisbon Treaty, the Member States have established an advanced institutional framework, through which they can pursue common foreign and security policy on the international scene. The CFSP is, however, not the only framework. Most of the Member States still maintain unilateral diplomatic channels. In this respect, it is noteworthy that Declaration 13 to the Treaty of Lisbon underlines that the provisions on the CFSP ‘do not affect the responsibilities of the Member States … for the formulation and conduct of their foreign policy nor of their national representation in third countries and international 26

Art. 15(6) TEU (President of the European Council); Art. 17 TEU (Commission) and Art. 27(2) TEU (High Representative). 27 Art. 218 TFEU. 28 See: Case C-658/11 European Parliament v Council (Pirate Transfer Agreement with Mauritius) EU:C:2014:2025 and comments in P Van Elsuwege, ‘Securing the Institutional Balance in the Procedure for Concluding International Agreements’ (2015) 52 CML Rev 1379. See also Case C-263/14 European Parliament v Council (Pirate Transfer Agreement with Tanzania) EU:C:2016:435. 29 Art. 5(2) TEU and Art. 13(2) TEU. 30 For comments, see: P Van Elsuwege and H Merket, ‘The Role of the Court of Justice in Ensuring the Unity of the EU’s External Representation’ in S Blockmans and RA Wessel (eds), Principles and Practices of EU External Representation (CLEER Working Papers 2012/5) 37.

50 Research handbook on the EU’s common foreign and security policy organisations’. Declaration 14 further specifies that the CFSP ‘will not affect the existing legal basis, responsibilities and powers of each Member State in relation to the formulation and conduct of its foreign policy, its national diplomatic service, relations with third countries and participation in international organisations’. The existence of an explicit EU competence in the area of CFSP (Art. 2(4) TFEU) therefore does not prevent the Member States from deploying their national foreign policy instruments. The Member States are furthermore also bound by their commitments to other international organizations. The UN is a key example and the special responsibilities of France and the United Kingdom in the Security Council are broadly recognized. On issues of security and defence, NATO is furthermore the primary locus of choice for most Member States.31 While the Member States may be happy to support the High Representative and EEAS in general, not all Member States may be happy to support them on all issues. The politics behind institutional choice are interesting as they contrast a functional (which institutional framework can best address the problem?) and political logic (which institutional framework would best serve the interests of a Member State?) with the EU obligations of loyal cooperation.32 The Court of Justice has clarified the implications of the loyalty principle, as expressed in Article 4(3) TEU, with respect to the EU’s non-CFSP external action. In essence, the duty of loyal cooperation implies that Member States may not undermine ‘a concerted Union strategy’ while expressing their national position in international fora.33 Significantly, such a strategy does not require the adoption of a legally binding document. As soon as a matter is discussed within the EU institutions, and even before the formal EU decision-making process enters into force, Member States are subject to special duties of action (i.e. actively supporting the EU’s position) and abstention (i.e. refraining from the adoption of positions or actions undermining the EU’s position or internal decision-making process). Significantly, the Member States’ obligations are of general application, irrespective of whether the EU’s competence is of an exclusive or shared nature.34 The question, of course, is to what extent these obligations in relation to the EU’s non-CFSP external action apply equally to EU representation in the field of CFSP. Reflecting the special status of the CFSP, a specific duty of cooperation is maintained in Article 24(3) TEU. This seems at first sight a redundant repetition taking into account the horizontal nature of the EU’s principles and objectives and the mandatory language used in defining the Member States’ obligations of loyalty in the field of CFSP. Pursuant to Article 24(3), the Member States ‘shall support’ the Union’s external and security policy, they ‘shall comply’ with the Union’s action in this area and ‘shall refrain’ from any action that is contrary to the Union’s interests or is likely to impair the effectiveness of its international action as a cohesive external actor. 31 ‘Shared Vision, Common Action: A Stronger Europe. A Global Strategy for the European Union’s Foreign And Security Policy’ (European External Action Service 2016) 20 accessed 24 February 2017. 32 J Jupille, W Mattli and D Snidal, Institutional Choice and Global Commerce (CUP 2013); JC Morse and RO Keohane, ‘Contested Multilateralism’ [2014] The Review of International Organizations 385. 33 Case C-246/07 Commission v Sweden ECLI:EU:C:2010:203, para 103. 34 ibid, para 71.

Representing the EU in the area of CFSP 51 Moreover, the text leaves little scope for exceptions as suggested by the expressions ‘actively’ and ‘unreservedly’.35 Article 28(2) TEU further specifies that CFSP decisions ‘commit the Member States in the positions they adopt and in the conduct of their activity’. Consequently, it appears that the importance of the distinction between the general principle of loyal or sincere cooperation under Article 4(3) TEU and its CFSP-specific variant under Article 24(3) TEU should not be overestimated. A noticeable difference, of course, is that the Council and the High Representative – rather than the Commission – shall ensure compliance with the loyalty principle in the field of CFSP. Whereas Article 24 TEU thus precludes the Commission from bringing a Member State before the Court of Justice for breaching its duties under the CFSP, Member State actions jeopardizing the attainment of the Union’s external action objectives nevertheless fall within the Court’s jurisdiction in the light of Article 4(3) TEU.36 The second consideration is the control of the High Representative and the EEAS. It is well known in political science and international relations that the delegation of tasks likely results in an agency problem.37 In layman’s terms: the High Representative and the EEAS have ideas of their own and may want to pursue those against the interests of some of the Member States. The Member States are therefore likely to keep the High Representative and the EEAS weak, under-staffed and subject to a whole range of oversight and accountability mechanisms.38 Furthermore, they are likely to bypass the High Representative and the EEAS if they anticipate that both EU actors will not effectively deliver on areas they consider important. A report of the UK House of Lords European Committee was quite explicit in this respect, when it stated that ‘[t]he EEAS should not … seek to project its own foreign policy. The Common Foreign and Security Policy should remain under the control of the Member States.’39 To better understand this so-called agency problem, it is necessary to identify first what the High Representative and the EEAS want to achieve. It is useful to distinguish between institutional interests and policy interests. In terms of institutional interests, there have been significant struggles between High Representatives and other EU actors. Javier Solana, Catherine Ashton and Federica Mogherini have all had their 35

C Hillion and RA Wessel, ‘Restraining External Competences of the Member States under CFSP’ in M Cremona and B De Witte (eds), EU Foreign Relations Law: Constitutional Fundamentals (Hart Publishing 2008) 91. 36 C Hillion, ‘Cohérence et action extérieure de l’Union’ in E Neframi (ed.), Objectifs et compétences de l’Union européenne (Bruylant 2012); see also Hillion (n 24). 37 R Kiewiet and M McCubbins, The Logic of Delegation: Congressional Parties and the Appropriations Process (Chicago University Press 1991); M Pollack, The Engines of European Integration: Agency, Delegation, and Agenda Setting in the EU (OUP 2003); D Hawkins, D Lake, D Nielson and M Tierney (eds), Delegation and Agency in International Organizations (CUP 2006); CA Bradley and JG Kelley, ‘The Concept of International Delegation’ (2008) 71 Law and Contemporary Problems 1. 38 H Dijkstra, Policy-Making in EU Security and Defence Policy: An Institutional Perspective (Palgrave Macmillan 2013); H Dijkstra, International Organizations and Military Affairs (Routledge 2016). 39 European Union Committee, The EU’s External Action Service (HL 2012–13, 147) accessed 24 February 2017.

52 Research handbook on the EU’s common foreign and security policy challenges with the Member States, the rotating Presidency and the European Commission.40 In terms of policy interest, the EU Treaties are an important guide.41 Furthermore, the High Representative and the EEAS have tried to pursue policies in which they have strengths. It is therefore no surprise that they have prioritized non-traditional threats, conflict prevention and the comprehensive approach, stressing the EU’s unique crisis toolkit.42 This has not always pleased the Member States. They notably complained about the disinterest of Ashton in defence, including her skipping a ministerial meeting in 2010.43 The High Representative and the EEAS can only pursue interests different from those of the Member States if they have sufficient resources. Informational advantages are critically important in this respect.44 The Member States accept EU-level representation precisely because they cannot carry out all representative functions themselves. As a result of direct EU representation, Member States lose the ability to control all the interactions that the EU representative has with foreign counterparts. After all, they may not be privy to those interactions. Furthermore, few Member States rival the diplomatic network of the EEAS. This gives the EEAS insight about many third countries. The EEAS has also significant in-house expertise and serves as the point of contact for many third parties. These advantages allow the High Representative and the EEAS to pursue their interests. Member States are keenly aware of this agency problem. They have installed a range of control mechanisms. It is common practice to distinguish between ex ante and ex post control mechanisms. The former determine the scope and procedure of a delegated task, while the latter are about the monitoring and sanctioning of the High Representative and the EEAS.45 EU statements, for example, need to be cleared with the Member States. The EEAS and its delegations are packed with national diplomats. Changes to the budget, personnel policy or organization of the EEAS furthermore need to be cleared with the Member States. Control and oversight are not necessarily explicit. The High Representative and the EEAS typically anticipate that the Member States may be unhappy with their actions and thus refrain from such actions in the first place. They are responsive to the wishes of the membership.

40

H Dijkstra, ‘Commission Versus Council Secretariat: An Analysis of Bureaucratic Rivalry in European Foreign Policy’ (2009) 14 European Foreign Affairs Review 431; Mueller-BrandeckBocquet and Rueger (n 7); N Helwig and C Rueger, ‘In Search of a Role for the High Representative: The Legacy of Catherine Ashton’ (2014) 49 International Spectator 1. 41 Arts 3(2), 21 TEU. 42 European Council, A Security Europe in a Better World: European Security Strategy (12 December 2003) accessed 17 February 2017; S Vanhoonacker and K Pomorska, ‘The European External Action Service and agenda-setting in European foreign policy’ (2013) 20 Journal of European Public Policy 1316. 43 V Pop, ‘Ashton’s absence from defence gathering raises eyebrows’ (EUObserver, 25 February 2010) accessed 22 February 2017. 44 Pollack (n 37); Hawkins and others (n 37). 45 T Delreux and J Adriaensen, ‘Introduction’ in T Delreux and J Adriaensen (eds), The Principal-Agent Model and the European Union (Palgrave 2017).

Representing the EU in the area of CFSP 53 The final consideration is one of bureaucratic politics. This goes a step further than the question of control. Member States may well be in agreement that it is better to be represented by the High Representative and the EEAS in a particular area of the CFSP. Yet this logic may run into trouble when considering the parochial interests within the Member States. In particular, the emergence of the High Representative and the EEAS challenges the foreign ministers and the national diplomatic services, respectively. We know that institutions are sticky:46 they tend to survive even if there is no explicit purpose for them. This can also be said regarding certain tasks of national diplomatic services. Typically among the most prestigious bits of government, they have proven almost immune to EU foreign policy cooperation and the creation of the EEAS. Few if any of the national diplomatic services have downsized as a result of Lisbon.47 If anything, they have tried to remain visible and relevant. Instances of bureaucratic politics result directly from the parallel existence of the EEAS and the national diplomatic services and a lack of clarity about the division of labour.48 The combination of the complex legal framework and political reality implies that the question of who is representing the EU is not always as straightforward as a cursory reading of the Treaties might suggest. The remainder of this chapter analyses different scenarios in order to better understand the mechanisms behind the process of EU representation in practice. It focuses subsequently on multilateral, bilateral and ad hoc EU external representation.

3. EU EXTERNAL REPRESENTATION IN MULTILATERAL FORA The EU’s representation in multilateral fora – understood to include both formal international organizations and international regimes – can take different forms, oscillating between full membership and observer status.49 This is the result of various legal and political factors, both inside and outside the EU. First of all, the EU’s capacity to act on the international stage is a matter of competence. It is therefore no coincidence that full membership is mostly granted in international organizations dealing with issues falling within the scope of the EU’s exclusive competence or within areas where extensive internal harmonization has taken place.50 However, there is no direct correlation between strong internal competences and strong participatory rights

46 P Pierson, Politics in Time: History, Institutions and Social Analysis (Princeton University Press 2004). 47 Balfour and Raik (n 6); Rosa Balfour and Kristi Raik (eds), The European External Action Service and National Diplomacies (European Policy Centre 2013). 48 R Adler-Nissen, ‘Symbolic Power in European Diplomacy: The Struggle Between National Foreign Services and the EU’s External Action Service’ (2014) 40 Review of International Studies 657; Dijkstra (n 7); Dijkstra (n 40). 49 Jørgensen, Oberthür and Shahin (n 4); Jørgensen and Laatikainen (n 4); Kaddous (n 7). 50 F Hoffmeister, ‘Outsider or Frontrunner? Recent Developments under International and European Law on the Status of the European Union in International Organisations and Treaty Bodies’ (2007) 44 CML Rev 54.

54 Research handbook on the EU’s common foreign and security policy in international organizations.51 This is due to the second variable, which is the statute of the international organization concerned. It determines whether or not the EU can participate and in which form.52 Whereas the EU may have the competence to be directly represented, the possibility to make use of this competence depends on the institutional framework of the respective bodies.53 The EU’s status in international organizations also largely depends upon the position of third countries since unanimity, or at least a qualified majority, is normally needed to accept a formal role for non-state actors. Other countries should thus be convinced that direct EU representation is not against their interests.54 Finally, EU Member States are not always keen to give up their position in multilateral fora in favour of a unified EU representation. Membership remains a clear expression of national sovereignty and is deemed important to safeguard unilateral influence on the international stage. Member States are traditionally reluctant to accept a stronger role for the EU if it comes at the expense of their own representation.55 They want to keep their unilateral institutional channels open, allowing them a variety of institutional choices. Arguably, this is even more relevant in the area of CFSP, which is a traditional stronghold of Member State sovereignty, in comparison to non-CFSP contexts where EU external representation more logically follows from the EU’s internal competences. Moreover, the most relevant international organizations dealing with CFSP matters, such as the UN, NATO or the Organization for Security and Cooperation in Europe (OSCE), are state centric. Individual Member States are members of these organizations on their own merits, not as part of some sort of broader EU external presence.56 As a result, attempts to upgrade the EU’s direct representation face legal and political difficulties. This is most clearly illustrated by the development of the EU’s representation in the UN.57 Prior to the Lisbon Treaty, the rotating Presidency represented the EU for matters concerning the CFSP whereas the Commission represented the European Community. The abolition of the pillar structure and the introduction of new rules for external representation under primary EU law (cf. supra) triggered a revision of this arrangement. In particular, the replacement of rotating Presidency representatives with 51

J Wouters and A-L Chané, ‘Brussels Meets Westphalia: The European Union and the United Nations’ in P Eeckhout and M Lopez-Escudero (eds), The European Union’s External Action in Times of Crisis (Hart Publishing 2016) 304. 52 Kaddous (n 7). 53 Wouters and Chané (n 51) 304. 54 J Sack, ‘The European Community’s Membership of International Organisations’ (1995) 32 CML Rev 1234. 55 ibid, 1232–1233. 56 Nevertheless, some formal and informal intra-EU coordination may take place. See, for instance, S Blavoukos and D Bourantonis, ‘The EU’s Performance in the United Nations Security Council’ (2011) 33 Journal of European Integration 731; N Græger and KM Haugevik, ‘The EU’s Performance With and Within NATO: Assessing Objectives, Outcomes and Organisational Practices’ (2011) 33 Journal of European Integration 743. 57 Laatikainen (n 5); E Drieskens, L Van Dievel and Y Reykers, ‘The EU’s Search for Effective Participation at the UN General Assembly and UN Security Council’ in E Drieskens and LG Van Schaik (eds), The EU and Effective Multilateralism: Internal and External Reform Practices (Routledge 2014).

Representing the EU in the area of CFSP 55 EU institutional actors had important repercussions. Whereas the Presidency representatives could benefit from their state’s UN member status to express the EU’s position, this was not automatically guaranteed for the EU’s institutional representatives. In line with the EU’s observer status, the latter would only be allowed to intervene at the end of the debates and would face serious limitations regarding the circulation of official documents and proposals. To avoid such detrimental consequences, a diplomatic initiative resulted in the adoption of UNGA Resolution 65/276 aligning the rules for the participation of the EU in the work of the UN with the provisions of the Lisbon Treaty.58 The new resolution allowed European Council President Van Rompuy to intervene for the first time in the General Debate of the UN General Assembly in September 2011. This was considered ‘a major breakthrough in a forum where the promotion of national sovereignty continues to be a dominant factor’.59 Nevertheless, the process leading to the adoption of this new arrangement also revealed the limits and practical problems of the EU’s representation in multilateral fora. In particular, it proved very difficult to convince third countries of the need to upgrade the EU’s rights of representation. Countries from Africa and the Caribbean Community (Cariforum) feared that allowing greater participation by the EU might undermine the principle of equality among sovereign states and the intergovernmental character of the UN system. This opposition led to significant amendments in comparison to the initial draft of the resolution, watering down the EU’s ambitions.60 The EU remains somewhat ‘procedurally handicapped’ within the UN system.61 For example, as the EU cannot vote in the UNGA, it also cannot give an explanation of a vote on behalf of the members. Apart from the opposition from third countries, the implementation of Resolution 65/276 also encountered resistance from EU Member States. The United Kingdom, in particular, feared that the EU’s direct representation might lead to an erosion of national competences. Accordingly, the British government actively resisted a so-called ‘representation creep’.62 This resulted in controversy about the right of EU delegations to deliver statements in the name of the EU in the area of shared competences. The United Kingdom insisted that such statements could only be made ‘on behalf of the EU and its Member States’ and not simply ‘on behalf of the EU’. A compromise solution 58

PA Serrano de Haro, ‘Participation of the EU in the Work of the UN: General Assembly Resolution 65/276’ (CLEER Working Papers 2012/4) accessed 3 March 2017; J Wouters, J Odermatt and T Ramopoulos, ‘The Status of the European Union at the United Nations General Assembly’ in I Govaere, E Lannon, P Van Elsuwege and S Adam (eds), The European Union in the World: Essays in Honour of Marc Maresceau (Brill 2014) 211. 59 Serano de Haro (n 58); see also Laatikainen (n 5) on all sorts of transitional arrangements and bureaucratic challenges for EU officials to represent the Union. 60 See Jan Wouters, Anna-Louise Chané and Jed Odermatt, ‘Improving the EU’s Status in the UN and the UN System: An Objective Without a Strategy?’ in Kaddous (ed.) (n 7) 52. 61 Wouters, Odermatt and Ramopoulos (n 58) 212. 62 UK Foreign and Commonwealth Office, Review of the Balance of Competences between the United Kingdom and the European Union, July 2013, 41, accessed 3 March 2017.

56 Research handbook on the EU’s common foreign and security policy was found in October 2011 when the Council adopted a document setting out the ‘General Arrangements for EU statements in multilateral organisations’.63 This discussion is evidence of the importance of the alternative and parallel institutional channels that Member States want to keep in addition to EU representation. The Council document provides some general principles and practical guidelines. Most notably, it addresses the UK’s concerns in pointing out that the EU’s external representation does not affect the internal division of competences. The EU can only make a statement when it is competent and after a common position has been agreed in accordance with the relevant Treaty provisions.64 The EU representation is to be exercised from behind the EU nameplate unless prevented by the rules of procedure of the relevant organization. Member States may complement the EU statements as long as they respect the principle of sincere cooperation. When Member States agree, EU actors can thus express statements relating to issues of national competence. In such instances, the statement will be prefaced by ‘on behalf of the Member States’. In areas of shared competence, the prefix ‘on behalf of the EU and its Member States’ will be used and for statements referring exclusively to positions adopted in the EU framework, reference will be made to ‘on behalf of the European Union’. Significantly, the guidelines provide that the latter practice includes the CFSP.65 The discussion surrounding the delivery of statements in multilateral fora revealed the key distinction between internal coordination and external representation. Once an EU position is adopted, this position is to be represented by EU actors externally on the basis of the division of labour foreseen in the Treaties. With respect to CFSP matters, this implies that the President of the European Council, the High Representative or the EU delegation take the floor on behalf of the EU. The Member States are bound to coordinate their action and to uphold the EU’s positions whereas the High Representative shall organize this coordination.66 In practice, this implies that the EU delegations play a crucial role in ensuring ‘on the spot’ coordination. More than 1,000 annual coordination meetings take place in New York, Geneva and Vienna ensuring the coherence of the EU’s position in the UN and the OSCE.67 For international conferences taking place in a location where there is no EU delegation, the Member State holding the Presidency may, on an ad hoc basis, chair the ‘on the spot’ internal coordination meetings.68

63

Council of the EU, ‘EU Statements in multilateral organisations: General Arrangements’, doc. 15901/11, 24 October 2011. 64 ibid. 65 ibid. 66 Art. 34(1) TEU. 67 M Schmidt, ‘The Position of the European Union in the United Nations. A United Nations Perspective’ in Kaddous (ed.) (n 7) 35. D Spence, ‘From the Convention to Lisbon: External Competence and the Uneasy Transition for Geneva Delegations’ in Spence and Batora (eds) (n 5); L-E Lundin, ‘Effective Multilateralism After Lisbon: The Added Value of the EEAS and the EU Delegation in Vienna’ in Spence and Batora (eds) (n 5); Laatikainen (n 5). 68 European Commission, Internal Guidance Note for all Commission Services on External Representation, SEC (2012) 118, 10 February 2012.

Representing the EU in the area of CFSP 57 After some teething problems,69 it seems that the post-Lisbon arrangements are now more or less established. At the same time, the ambition to further improve the EU’s status in international organizations has become more modest. The 2012 Barroso-Aston paper, for instance, called for a piecemeal approach, focusing on a limited number of organizations.70 However, as can be derived from the CJEU’s judgment relating to the International Organisation of Vine and Wine (OIV), nothing prevents the Union from adopting a position to be defended on its behalf in a body set up by an international agreement to which it is not a party, even when not all Member States are a party. The only criterion is that the adopted position concerns an area of law which falls within the EU’s competence.71 Even in a context where EU external representation is not an evident option, the Member States are therefore bound to respect their EU law obligations. This also applies with respect to CFSP matters. Article 34(1) TEU provides that ‘[i]n international organisations and at international conferences where not all the member states participate, those which do take part shall uphold the Union’s positions’. Accordingly, the Member States which are members of the Security Council are under a duty to keep the other Member States and the High Representative fully informed and to defend the positions and the interests of the Union.72 When the EU has a defined position, the EU Member States which sit on the Security Council must request that the High Representative be invited to defend the EU’s position.73 This innovation of the Lisbon Treaty significantly increased the visibility of the EU and allowed the High Representative, and particularly the EU delegation, to intervene on various occasions.74

4. BILATERAL EU EXTERNAL REPRESENTATION While EU external representation in multilateral fora has brought some specific challenges, such as voting and speaking rights, getting the appropriate hearing at the bilateral level has hardly been easier for the EU. Once again, we see significant variation in EU bilateral representation across the globe, which is not easily explained by the EU competences.75 Indeed, to understand how external representation on the 69

PM Kaczynski, ‘Swimming in Murky Waters. Challenges in Developing the EU’s External Representation’, FIIA Briefing Paper 88, September 2011. 70 Communication to the Commission from the President in Agreement with Vice-President Ashton, Strategy for the progressive improvement of the EU status in international organisations and other fora in line with the objectives of the Lisbon Treaty, C(2012) 9420 final, 20 December 2012. 71 Case C-399/12 Germany v Council EU:C:2014:2258, paras 49–52. 72 D Marchesi, ‘The EU Common Foreign and Security Policy in the UN Security Council: Between Representation and Coordination’ (2010) 15 European Foreign Affairs Review 97; Blavoukos and Bourantonis (n 56); Drieskens, Van Dievel and Reykers (n 57). 73 Art. 34(2) TEU. 74 Laatikainen (n 5). 75 Some of the empirical examples in this section are also discussed in H Dijkstra, ‘Non-exclusive Delegation to the European External Action Service’ in Delreux and Adriaensen (eds) (n 45).

58 Research handbook on the EU’s common foreign and security policy bilateral level plays out requires us to account for the host country, both in terms of its importance for individual Member States as well as its attitude towards the EU. As noted earlier, the Commission has long had an extensive network of delegations in other countries. These were offices of the Commission, staffed by Commission officials, trying to implement European policies from trade to development and enlargement. As the permanent bases of the EU abroad, they also became focal points for issues such as press and public diplomacy. The hybrid nature of the Commission delegations, somewhere between technical offices and full-fledged embassies, obviously resulted in challenges with protocol and accreditation. For instance, the Commission staff member appointed as Head of Delegation could not be called an ‘ambassador’.76 Considering the sizeable EU development budgets, it is clear that some of the counterparts would prioritize the Commission Head of Delegation over many of the national ambassadors. The Lisbon Treaty was supposed to address some of these challenges. The purpose was to turn all the bilateral Commission delegations into upgraded EU delegations consisting of both Commission civil servants and seconded European diplomats. This transformation has taken a considerable time and has been uneven at best. That the EU has so many bilateral delegations across the world allows us to run medium or large analyses of external representation. Austermann has done just that.77 She shows how Commission delegations have been upgraded to EU delegations at ‘different speeds’. One of her findings – that EU diplomacy is least centralized in major economic partners – is perhaps surprising from the perspective of the EU’s exclusive competence in this area,78 but it is completely unsurprising when considering the significant interests of the individual Member States in places like Washington or Beijing. Particularly when host countries matter, the Member States are unlikely to let the EEAS and its delegations get in their way. The significance of the host country for the individual Member States is also underlined by Balfour and Raik, who note that EU delegations have ‘been relatively easy to establish in less important and peripheral locations where member states have fewer political interests at play … The easiest cases are locations where one’s own country has no representation.’79 Indeed they find that Washington, Beijing, New Delhi, Moscow, Cairo and Tokyo are places where Member States want to avoid their bilateral channels being compromised.80 Dijkstra argues that these are instances of what he calls ‘non-exclusive delegation’ where Member States have delegated to the EU delegations a role in bilateral representation, but this role is hardly exclusive and is (re)negotiated on a daily basis (not dissimilar to EU representation in multilateral fora).81 It is impossible to describe the local dynamics in all the countries where the EU maintains bilateral relations but scholars have analysed the nitty-gritty details of setting 76

Bruter (n 9); Ramses Wessel, ‘Can the EU Replace its Member States in International Affairs? An International Law Perspective’ in Govaere and others (n 58). 77 Austermann (n 6). 78 C Damro, ‘Market Power Europe’ (2012) Journal of European Public Policy 682. 79 Balfour and Raik (n 6) 44. 80 ibid; see also Austermann (n 6). 81 Dijkstra (n 75).

Representing the EU in the area of CFSP 59 up several EU delegations after the entry into force of the Treaty of Lisbon.82 It is worth highlighting some of their key findings because these findings show us the importance of local variables affecting EU external representation. To start with China, we have witnessed the appearance of an interesting de facto division of labour between the Member States and the EU. Whereas the Member States themselves are in fierce competition regarding trade promotion and investment, they collectively have been happy to leave the difficult political dossiers (arms embargo, climate change, human rights, intellectual property rights, the market economy status) to the EU. The EU has proved a convenient cover shielding the bilateral trading interests. This is in line with intra-EU principal–agent dynamics outlined above. Austermann has analysed this intra-EU political dynamic in greater detail with respect to the institutionalization of the EU delegation in Beijing. She writes that ‘[d]espite the clearly upgraded political role, the EU Delegation cannot do away with … the diverging interests of member states. This makes it difficult to speak always with one voice in Beijing.’83 Indeed, the Member States were pleased to have the Head of Delegation speak about human rights, such as in the case of the detention of Ai Weiwei, but refused to allow him to speak on behalf of the whole EU.84 He could only make a statement on his own behalf. It is clear from the Beijing example that Member States have different channels for bilateral representation and they use them strategically. While the division of labour between the Member States and the EU institutions is thus critically important for bilateral representation, it is not the full story. As with the multilateral fora, one also needs to account for the local environment. A comparison between EU delegations in Washington and Moscow is instructive.85 The way in which the US and Russian administrations systematically play around with the corps diplomatique (in different ways) also affects how the EU is represented externally. Maurer notes the diplomatic circus in Washington is a constant ‘beauty contest’.86 The ambition for diplomatic actors is to gain the attention of the US administration. As such, individual Member States have been deeply suspicious of any bilateral role of the EU delegation, particularly in the area of political and security affairs.87 While this finding seems paradoxical considering that in Beijing Member States delegated precisely such tasks, it shows the significance of the local environments. The EU delegation in Washington has tried to stick to core EU competences, such as transatlantic trade, and to ensure that the EU effort would serve the interests of the Member States. One example has been organizing briefing meetings with US counterparts for all EU diplomats at the same time. The US administration has been supportive of this. As Maurer and Raik write, ‘in Washington it is difficult to reach a high-ranking official … The US had actively pushed the idea of meeting all EU member states … at 82

See particularly individual chapters in Spence and Bátora (n 6). F Austermann, ‘Towards One Voice in Beijing? The Role of the EU’s Diplomatic Representation in China Over Time’ (2012) 18 Journal of European Integration History 83, 101. 84 ibid. 85 H Maurer and K Raik, Pioneers of a European Diplomatic System: EU Delegations in Moscow and Washington (FIIA 2014). 86 H Maurer, ‘Europe in America: An Upgraded EU Delegation in a Reinforced System of European Diplomatic Coordination’ in Spence and Batora (eds) (n 5) 276. 87 ibid, 28. 83

60 Research handbook on the EU’s common foreign and security policy the same time … even before the Lisbon Treaty … such exchanges with the EU28 have become standard practice.’88 The environment in Moscow is quite different. The Russian effort has focused on divide-and-rule and, as such, Russian officials have not been keen to meet the EU28 collectively.89 Access is generally restricted and Moscow has had little interest in EU-level officials, preferring to deal with more high-level national (e.g. German) officials. Maurer and Raik note that this Russian attitude has pushed diplomats of the EU Member States closer together,90 particularly since the annexation of Crimea and the establishment of the EU sanctions regime in 2014. For our understanding of EU external representation more broadly, however, it is once again significant to account for this local environment in which bilateral representation takes place.

5. AD HOC EU EXTERNAL REPRESENTATION While multilateral and bilateral diplomacy tend to be institutionalized, ad hoc forms of external representation may allow the EU institutions a greater presence. Yet what we have seen, particularly with respect to high-level mediation, is a variation of formats through which the EU and its Member States are represented. Indeed, there also seem to be important internal and external variables at play in relation to ad hoc forms of external representation. Furthermore, while ad hoc external representation implies one-off events, it often consists of a process of mediation, persisting for years and resulting in a strong degree of path dependence. There is a relatively long tradition of EU representatives participating in ad hoc diplomacy. Perhaps most famously, the so-called ‘Troika’ consisting at the time of the rotating Presidency and its predecessor and successor sought to mediate in the erupting conflict in former Yugoslavia in 1991.91 Similarly, former Swedish Prime Minister Carl Bildt was the EU envoy and Vice-Chair of the Dayton conference in 1995, even though his position was informally undermined by the British and French representatives also present.92 These experiences, and the lack of EU success in the Western Balkans, provided direct input for the Treaty of Amsterdam of 1999, which established the position of the EU High Representative (a continuous position to assist the rotating Presidency) and codified the EU Special Representatives, who would be appointed by the Member States to give the EU representation in some of the world’s most important conflicts. The first High Representative, Javier Solana, was particularly active in terms of ad hoc external representation. Perhaps because the rules of ad hoc diplomacy are less set in stone, this was an area where he could make a name for himself. As High 88

Maurer and Raik (n 85) 13. ibid. 90 ibid, 13–14. 91 Luxembourg foreign minister Jacques Poos, flanked by his Dutch and Italian counterparts, declared that ‘The hour of Europe has dawned’, cited in T Garton Ash, History of the Present: Essays, Sketches and Dispatches from Europe in the 1990s (Penguin Books 2000) 94. 92 R Holbrooke, To End a War (Random House 1998) 242; Dijkstra (n 75). 89

Representing the EU in the area of CFSP 61 Representative, Solana was instrumental – together with NATO Secretary-General Lord Robertson – in negotiating the OHRID peace agreement for the former Yugoslav Republic of Macedonia in 2001. From 2003, he became the negotiator on behalf of the international community with Iran on nuclear non-proliferation. This paved the way for the later ‘successes’ of Ashton and Mogherini. Solana’s biggest coup, however, was securing a seat on behalf of the EU in the Middle-East Quartet in 2004. Importantly, he did not have to ‘share’ this seat with other European representatives. He was a full member together with the American and Russian foreign ministers as well as the UN Secretary-General. That the ‘appointment’ of EU representatives in ad hoc external representation goes beyond formal competences is perhaps best illustrated by Solana’s mediation during the Israeli–Lebanese war in 2006.93 The Member States were confused and divided but Solana simply stepped on a plane. Asked by reporters on the ground whether he would lead mediation attempts, he stated that ‘I am not going to tell you about any potential way of solving [the conflict]. My visit today here is a visit of friendship, of solidarity and support.’94 Several days later, he was back in Brussels where he briefed the EU foreign ministers, who gave him a lead role with respect to mediation. The situation was different in 2008, when Russia intervened in Georgia, and the French President Nicolas Sarkozy put himself forward as the rotating Presidency in negotiating a peace deal. Solana travelled with Sarkozy to Moscow and Tbilisi, but was all but ignored.95 When Lady Catherine Ashton succeeded Solana in 2009, there were questions as to whether she would continue such activism, as she had less diplomatic experience. Nevertheless, a significant dose of institutional path dependence allowed her to continue the role of the High Representative in the Quartet and the Iran negotiations. The Middle-East Quartet became significantly less important during her tenure due to a lack of developments on the ground but she put significant efforts into the Iran negotiations. Despite the fact that most observers were happy to see her leave Brussels in 2014, there was a suggestion to keep her on as the main negotiator to bring the Iran negotiations to a successful close.96 The new High Representative Mogherini indeed kept her on for another round of negotiations, but quietly replaced her soon afterwards to benefit from the eventual deal. In addition to Iran, mediation in the conflict between Serbia and Kosovo provides a good example of ad hoc diplomacy. Initially, Finnish President Martti Ahtisaari represented the EU in negotiations with Serbia and Russia.97 Significantly, the appointment of Martti Ahtisaari as the EU mediator was an initiative of the German Council Presidency in 1999. It was not based on a formal procedure and, for this reason, his status and mandate always remained somewhat unclear.98 Moreover, the 93

Dijkstra (n 7). Cited in ibid, 8. 95 ibid, 12–15. 96 L Norman, ‘EU’s Mogherini Appoints Catherine Ashton Special Adviser for Iran Talks’ (Wall Street Journal, 5 December 2014) accessed 3 March 2017. 97 See Statement of the European Union on Kosovo, 31 May 1999, Bull. EU 1999-5, 1.3.18. 98 A-S Maass, EU-Russia Relations 1999–2015. From Courtship to Confrontation (Routledge 2016) 14. 94

62 Research handbook on the EU’s common foreign and security policy Council had already formally appointed Wolfgang Petritsch as the EU Special Envoy to Kosovo.99 Notwithstanding this ambiguity surrounding the EU’s representation, the negotiations between Ahtisaari and the Russian ambassador Chernomyrdin paved the way for the adoption of Security Council Resolution 1244. This provided for the departure of Serbian troops and the deployment of an international civilian and security mission. After Kosovo’s unilateral declaration of independence in February 2008, the EU played a crucial role in the process of reconciliation with Serbia. From 2011, under the leadership of Lady Ashton, the EEAS started a so-called ‘facilitated dialogue’ between Belgrade and Pristina.100 The EU’s mediation efforts successfully resulted in a series of agreements concerning Kosovo’s external representation (2012), the acceptance of key principles governing the normalization of relations (2013) and issues such as energy, telecoms, the status of the Serb-majority municipalities in Kosovo, and freedom of movement over the Mitrovica bridge (2015). It was an example of classic, tiresome, long-term diplomacy coupled with sweeteners, such as the EU candidate status for Serbia and the Stabilisation and Association Agreement (SAA) with Kosovo. Significantly, acting through the EU’s institutional structures allowed the Member States to circumvent the delicate issue of recognition. This explains why the SAA with Kosovo was concluded by the EU alone and not by the EU and its Member States.101 While the EU institutions and their representatives have become serious actors in diplomacy, it is also important to point at some setbacks. The conflict between Ukraine and Russia clearly illustrates that the EU can only play a meaningful role if this is acceptable for all parties involved. When Ashton decided to fly to Kiev in January 2014 to mediate between President Yanukovich and the opposition, Russian President Vladimir Putin quickly dismissed her initiative.102 Subsequently, crisis diplomacy was conducted by the foreign ministers of France, Germany and Poland – the so-called ‘Weimar triangle’ – acting ‘in close coordination’ with EU High Representative Ashton.103 Following Russia’s annexation of Crimea and the escalation of violence in the eastern part of Ukraine, the EU’s institutional actors gradually became sidelined. At the Minsk I negotiations, the EU was represented by an unusual troika of Ashton, Trade Commissioner De Gucht and Energy Commissioner Oettinger. At the Minsk II negotiations, there were no EU institutional actors. Instead, the President of France and 99

Council of the EU, Joint Action 1999/239/CFSP in relation to the nomination of a special envoy for Kosovo [1999] OJ L89/1. 100 Bergmann and Niemann (n 1). 101 P Van Elsuwege, ‘Legal Creativity in EU External Relations: The Stabilisation and Association Agreement between the EU and Kosovo’ (2017) 22 European Foreign Affairs Review forthcoming. 102 I Traynor, ‘Vladimir Putin tells Brussels to stay out of Ukraine’s political crisis’, The Guardian (28 January 2014) accessed 1 March 2017. 103 German Ministry of Foreign Affairs, ‘Crisis Diplomacy in Ukraine’ accessed 1 March 2017.

Representing the EU in the area of CFSP 63 the Chancellor of Germany were represented. Switzerland, as the OSCE Chairman-inoffice, played an important coordinating role. This constellation reflected Russia’s reluctance to let EU actors play a role in questions of security and foreign policy. Instead, the preference is for intergovernmental discussions under the so-called ‘Normandy format’, involving the leaders of Germany, France, Russia and Ukraine.

6. CONCLUSION EU external representation remains a puzzle for all but the closest observers. Whereas the EU is represented in one conflict by someone from the EU institutions, in other conflicts several Member States may take a lead role, and in yet other conflicts there is no common EU representation and/or position to speak of. While academics have long been interested in the question ‘who speaks for Europe?’, they have failed to come up with definite answers. The Treaty of Lisbon, rather than solving this question once and for all, has created new challenges for EU external representation across the world. In this chapter we have not provided definite answers, but we have offered a new way of understanding EU external relations by analysing more explicitly the interaction of legal and political dynamics. For us, it is clear that while the EU Treaties have increasingly expanded the scope for collective EU external representation by EU agents, most recently with the Lisbon Treaty, one needs to recognize that there is a range of internal and external political and legal dynamics that constrain the ability of the EU to fully represent itself. For instance, the upgrading of the Commission delegations to EU delegations has been uneven, and the EU delegations have not yet reached their full potential. While some of these dynamics have previously been discussed in the academic literature (and been applied to individual case studies), it is worthwhile to comprehensively sum up the most important limits and enablers for EU external representation. With respect to intra-EU politics, there is an important tension between the ability of the individual Member States to make their own institutional choices and their legal obligations to follow the EU line. Institutional choice by the Member States (whether they act unilaterally or through EU channels) when addressing cooperation problems is well discussed in the international relations literature, whereas legal scholars have been interested in loyal cooperation. Yet to fully understand how the trade-off works out in practice requires us to take an interdisciplinary point of view. Indeed, much of the discussion about ‘representation creep’ in the UN concerns political arguments dressed up in legal terminology. Alternatively, France and the United Kingdom have continually refused to be loyal Member States when it comes to the Security Council, citing their conflicting obligations under the UN Charter. It is, however, not only about how to most effectively promote national interest or solve a cooperation problem. Member States have also been worried about agency drift by the EU agents. In addition, we have seen strong bureaucratic politics dynamics at play. National embassies in places such as Washington have been worried about losing access to local counterparts. The different rotating Presidencies across the multilateral fora have not been all too happy about giving up their own roles. At the same time, the Member States occasionally use EU representation strategically, for instance to shift the

64 Research handbook on the EU’s common foreign and security policy blame. Beijing is a key location where the EU is in charge of addressing some of the more difficult dossiers. What our overview of EU external representation has also shown is that we need to look beyond intra-EU legal and political dynamics. It is not sufficiently well understood in the current academic literature that a critical condition for effective EU representation is in fact external and has to do with the environment in which representation takes place. Simply put, other states should be willing to meet EU representatives. If they are not, there is little that the EU can do about it. The importance of such external conditions is perhaps most obvious when it comes to the EU’s dealings with the great powers. The EU has had a particularly hard time dealing with Russia (be it over Georgia or Ukraine). Yet it is not just about power politics. In the UN context, other (weaker) regional groupings have been able relatively easily to frustrate efforts for more EU representation. The EU, as a non-state actor, remains the odd one out in a world of sovereign states and this significantly conditions its ability to externally represent itself.

4. ‘The Good, the Bad and the Ugly’: three levels of judicial control over the CFSP Christophe Hillion and Ramses A. Wessel

1. INTRODUCTION A quarter of a century ago, a book chapter on judicial scrutiny of the then new Common Foreign and Security Policy (CFSP) of the EU would have raised eyebrows. Times have changed.1 The role of the CJEU in relation to the CFSP is now studied extensively2 as the Court’s CFSP case law is fast expanding.3 This remarkable development is not only due to the Member States’ decision partially to lift the judicial

1 cf. C Hillion and RA Wessel, ‘Restraining External Competences of EU Member States under CFSP’ in M Cremona and B De Witte (eds), EU Foreign Relations Law: Constitutional Fundamentals (Hart Publishing 2008). 2 See for instance S Griller, ‘The Court of Justice and the Common Foreign and Security Policy’ in A Rosas, E Levits and Y Bot (eds), Court of Justice of the European Union – Cour de Justice de l’Union Européenne, The Court of Justice and the Construction of Europe: Analyses and Perspectives on Sixty Years of Case-law – La Cour de Justice et la Construction de l’Europe: Analyses et Perspectives de Soixante Ans de Jurisprudence (TMC Asser Press 2013); G De Baere and P Koutrakos, ‘The Interactions Between the Legislature and the Judiciary in EU External Relations’ in P Syrpis (ed.), The Judiciary, the Legislature and the EU Internal Market (CUP 2012); L Saltinyté, ‘Jurisdiction of the European Court of Justice over Issues Related to the Common Foreign and Security Policy under the Lisbon Treaty’ (2010) Jurisprudence 119 et seq.; A Hinarejos, Judicial Control in the European Union – Reforming Jurisdiction in the Intergovernmental Pillars (OUP 2009); C Hillion, ‘A Powerless Court? The European Court of Justice and the Common Foreign and Security Policy’ in M Cremona and A Thies (eds), The European Court of Justice and External Relations Law (Hart Publishing 2014); C Hillion, ‘Decentralised Integration? Fundamental Rights Protection in the EU Common Foreign and Security Policy’ (2016) European Papers 55; RA Wessel, ‘Resisting Legal Facts: Are CFSP Norms as Soft as They Seem?’ (2015) European Foreign Affairs Review 123; RA Wessel, ‘Lex Imperfecta: Law and Integration in European Foreign and Security Policy’ (2016) 2 European Papers 439; C Eckes, ‘Common Foreign and Security Policy: The Consequences of the Court’s Extended Jurisdiction’ (2016) European Law Journal 492; G Butler, ‘The Coming of Age of the Court’s Jurisdiction in the Common Foreign and Security Policy’ (2017) European Constitutional Law Review 673; M Cremona, ‘Effective Judicial Review is of the Essence of the Rule of Law: Challenging Common Foreign and Security Policy Measures before the Court of Justice’ (2017) European Papers 671; P Koutrakos, ‘Judicial Review in the EU’s Common Foreign And Security Policy’ (2018) International and Comparative Law Quarterly 1; and J Heliskoski, ‘Made in Luxembourg: the Fabrication of the Law on Jurisdiction of the Court of Justice of the European Union in the Field of the Common Foreign and Security Policy’ (2018) Europe and the World: A Law Review (forthcoming). 3 References may be found throughout this chapter.

65

66 Research handbook on the EU’s common foreign and security policy immunity from which the CFSP has traditionally benefited. It is also the result of the incremental integration of the policy in the increasingly constitutionalized EU legal order.4 However far reaching the constitutionalization of the CFSP may have been, gaps nevertheless remain. In particular, the jurisdiction of the Court of Justice, as it has itself recognized,5 is legally limited. Given the many recent studies on that very issue,6 this chapter will not repeat the detailed analyses of the relevant recent cases. By examining possible forms of judicial control over CFSP at different levels (‘the good, the bad, and the ugly’), this chapter rather aims to discuss the Court’s approach to the system of judicial control over the CFSP and to provide a more holistic picture of possibilities and pitfalls. Having recalled the post-Lisbon developments in the CJEU’s jurisdiction in relation to the CFSP, the present contribution thus asks whether and, if so, to what extent remaining gaps in the Court’s control can be filled by involving other courts – both internally at Member States level, and externally by involving international and/or third countries’ courts. Our main argument is that the Court’s suspicion in relation to alternative judicial oversight may be legitimate. However, acknowledged gaps in the EU system of judicial remedies in relation to the CFSP ought to be filled for the Union to meet the requirements of the rule of law.7 The incremental acknowledgement of the Court’s jurisdiction in relation to CFSP might not suffice, which should as a result leave space for complementary solutions. Yet, judging from the case law, the Court of Justice seems generally reluctant to tolerate any other judicial control over the CFSP. For the CJEU, involvement of international courts (or even domestic courts in third states) entails many risks in relation to safeguarding the autonomy of EU law (hence ‘the bad’). This was considered as an obstacle to the Union’s accession to the European Convention on Human Rights (ECHR), even if it could have filled at least that judicial gap. The Court thus held that ‘jurisdiction to carry out a judicial review of acts, actions or omissions on the part of the EU, including in the light of fundamental rights, cannot be conferred exclusively on an international court which is outside the institutional and judicial framework of the EU’.8 Similarly, while acknowledging that the role of Member States is not in itself a bad idea as it is even supported by the Treaties, it is generally seen as an unattractive

4

cf. RA Wessel, ‘The Dynamics of the European Union Legal Order: An Increasingly Coherent Framework of Action and Interpretation’ (2009) European Constitutional Law Review 117; and, more recently, RA Wessel, ‘Integration and Constitutionalisation in EU Foreign and Security Policy’ in R Schütze (ed.), Governance and Globalization: International and European Perspectives (CUP 2018). 5 Opinion 2/13 of the Court of 18 December 2014: Accession by the Union to the European Convention for the Protection of Human Rights and Fundamental Freedoms ECLI: EU:C:2014:2454, para 252: ‘as EU law now stands, certain acts adopted in the context of the CFSP fall outside the ambit of judicial review by the Court of Justice’. 6 See above (n 2). 7 As spelled out in Case 294/83 Les Verts v European Parliament EU:C:1986:166. 8 ibid, para 256.

‘The Good, the Bad and the Ugly’ 67 substitute (‘the ugly’) for the harmonizing role of the Court of Justice itself.9 The fact that the CFSP is part and parcel of the EU’s legal order may explain the underlying Court’s claim that it should itself exercise judicial control over this EU policy (hence ‘the good’), particularly in view of Article 344 TFEU. It remains that this position does not help to fill the constitutional gap, unless it is understood as pressing the Masters of the Treaty to address it by expanding the Court’s jurisdiction.

2. ‘THE GOOD’: THE COURT OF JUSTICE The Court’s view is that it should be the one deciding on the interpretation and validity of all EU law. Indeed, Kirchberg is the place to prevent disparities in EU law from occurring. However, since the Treaty of Lisbon, the Court of Justice has been entrusted only with a limited jurisdiction in relation to the CFSP. According to Article 24(1) TEU it shall not have jurisdiction with respect to these provisions [i.e. ‘specific provisions on the Common Foreign and Security Policy’ enshrined in Chapter 2 of the TEU10] with the exception of its jurisdiction to monitor compliance with Article 40 of this Treaty [TEU] and to review the legality of certain decisions as provided for by the second paragraph of Article 275 of the Treaty on the Functioning of the European Union.

‘The latter provision further specifies that the Court shall have jurisdiction to monitor compliance with Article 40 of the Treaty on European Union and to rule on proceedings, brought in accordance with the conditions laid down in the fourth paragraph of Article 263 of this Treaty, reviewing the legality of decisions providing for restrictive measures against natural or legal persons adopted by the Council on the basis of Chapter 2 of Title V of the Treaty on European Union’.

As argued elsewhere, these provisions have made it possible for the Court, albeit within limits, to exercise judicial control with regard to certain CFSP acts. They also recalibrate its role in patrolling the borders between EU (external) competences based on the TFEU and the CFSP, while generalizing its capacity to enforce the principles underpinning the Union’s legal order.11 The case law that has developed since the entry into force of these provisions displays the Court’s broad conception of its CFSP-related jurisdiction. Its basic understanding is encapsulated in the following formula: [T]he final sentence of the second subparagraph of Article 24(1) TEU and the first paragraph of Article 275 TFEU introduce a derogation from the rule of the general jurisdiction which 9 Obviously, these qualifications are borrowed from the classic 1966 Western movie by Sergio Leone, The Good, The Bad, and the Ugly (Il buono, il brutto, il cattivo) . 10 See Case C-72/15 Rosneft ECLI:EU:C:2017:236, Opinion of AG Wathelet, para 46. 11 Hillion‘’ (n 2); RA Wessel, ‘Lex Imperfecta’ (n 2). For an analysis of the Court’s jurisdiction in relation to sanctions, see Chapter 10 in this volume.

68 Research handbook on the EU’s common foreign and security policy Article 19 TEU confers on the Court to ensure that in the interpretation and application of the Treaties the law is observed, and they must, therefore, be interpreted narrowly’ (emphasis added).12

Articles 24(1) TEU and 275(2) TFEU are thus not interpreted as establishing a distinct Court’s jurisdiction for the purpose of the CFSP. Rather, the judicial control it intends to perform in relation to that policy appears to be the same as the one it exercises generally, as envisaged in Article 19 TEU, albeit within the limits spelled out in those Articles. This ‘generalist’ (so to speak) conception of the Court’s jurisdiction in the area of CFSP led it to consider that its legality control over CFSP restrictive measures is not limited to annulment proceedings envisaged in Article 263(4) TFEU, but includes the possibility for it to give a preliminary ruling on their validity: Since the purpose of the procedure that enables the Court to give preliminary rulings is to ensure that in the interpretation and application of the Treaties the law is observed, in accordance with the duty assigned to the Court under Article 19(1) TEU, it would be contrary to the objectives of that provision and to the principle of effective judicial protection to adopt a strict interpretation of the jurisdiction conferred on the Court by the second paragraph of Article 275 TFEU, to which reference is made by Article 24(1) TEU … . In those circumstances, provided that the Court has, under Article 24(1) TEU and the second paragraph of Article 275 TFEU, jurisdiction ex ratione materiae to rule on the validity of European Union acts, that is, in particular, where such acts relate to restrictive measures against natural or legal persons, it would be inconsistent with the system of effective judicial protection established by the Treaties to interpret the latter provision as excluding the possibility that the courts and tribunals of Member States may refer questions to the Court on the validity of Council decisions prescribing the adoption of such measures.13

The Court’s legality control over certain CFSP acts is therefore the same as the one it exercises over other EU acts. It is an expression of its general mandate as established in Article 19 TEU;14 it is governed by the same principles, in particular the principle of effective judicial remedies enshrined in Article 47 of the Charter of Fundamental Rights.15 The application of the general EU rules on legality control to the CFSP context illustrates that the Court considers the CFSP as firmly embedded in the EU legal order, despite its procedural specificity mentioned in Article 24(1) TEU. Principles and rules of general application would thus be guaranteed through judicial oversight even where applied to a CFSP situation. The latter circumstance does not entail judicial immunity. Three illustrations come to mind: 12 Case C-658/11 EP v Council (Mauritius) ECLI:EU:C:2014:2025, para 70. See also Case C-439/13P Elitaliana ECLI:EU:C:2015:753, para 41; Case C-455/14P H v Council ECLI:EU: C:2016:569, para 40. 13 Rosneft (n 10). 14 Further on this general mandate, see C Hillion, ‘Conferral, Cooperation and Balance in the Institutional Framework of the EU External Action’ in M Cremona (ed.), Structural Principles in EU External Relations Law (Hart Publishing 2018). 15 cf. Cremona (n 2).

‘The Good, the Bad and the Ugly’ 69 First, the Court has made clear that since international agreements in the area of CFSP are concluded on the basis of the general provisions of Article 218 TFEU, albeit subject to some specific arrangements, the Court would exercise judicial control to ensure compliance with the terms of that procedure:16 [T]he obligation imposed by Article 218(10) TFEU, under which the Parliament is to be ‘immediately and fully informed at all stages of the procedure’ for negotiating and concluding international agreements, applies to any procedure for concluding an international agreement, including agreements relating exclusively to the CFSP … Article 218 TFEU, in order to satisfy the requirements of clarity, consistency and rationalisation, lays down a single procedure of general application concerning the negotiation and conclusion of international agreements by the European Union in all the fields of its activity, including the CFSP which, unlike other fields, is not subject to any special procedure.17

The application to a CFSP situation of a TFEU-based procedure does not therefore affect the Court of Justice’s jurisdiction in relation to that procedure.18 Second, and in the same vein, the Court has considered that it would have jurisdiction to control the legality of a decision awarding a public service contract in the context of an EU CSDP Mission given that the contract concerned involved an expenditure to be allocated to the EU budget, and thereby subject to the provisions of the EU Financial Regulation. Confirming the derogatory character of the terms of Articles 24(1) TEU and 275(2) TFEU, and consequently their narrow application, the Court concluded that [h]aving regard to the specific circumstances of the present case, the scope of the limitation, by way of derogation, on the Court’s jurisdiction, which is provided for in the final sentence of the second subparagraph of Article 24(1) TEU and in Article 275 TFEU, cannot be considered to be so extensive as to exclude the Court’s jurisdiction to interpret and apply the provisions of the Financial Regulation with regard to public procurement.19

16 T Tridimas, ‘The European Court of Justice and the Draft Constitution: A Supreme Court for the Union?’ in T Tridimas and P Nebbia (eds), European Union Law for the Twenty-First Century: Rethinking the New Legal Order. Vol. 1: Constitutional and Public Law. External Relations (Hart Publishing 2004) 128; G De Baere, Constitutional Principles of EU External Relations (OUP 2008) 190. 17 Case C-263/14 Parliament v Council (EU-Tanzania Transfer Agreement) ECLI:EU:C: 2016:435, para 68. 18 In the words of Peers: ‘the Court’s ruling means that any CFSP measure can be litigated before it, as long as the legal arguments relate to a procedural rule falling outside the scope of the CFSP provisions of the Treaty (Title V of the TEU). For instance, it arguably means that the Court would have the power to rule on the compatibility of proposed CFSP treaties with EU law, since that jurisdiction is conferred by Article 218 TFEU and not expressly ruled out by Article 275. But such disputes might often include arguments about the substance of the measure concerned (for instance, whether it would breach the EU’s human rights obligations), and it could be awkward to distinguish between procedural and substantive issues in practice.’ See: S Peers, ‘The CJEU Ensures Basic Democratic and Judicial Accountability of the EU’s Foreign Policy’ (EU Law Analysis, 24 June 2014) . 19 Elitaliana (n 12), para 49.

70 Research handbook on the EU’s common foreign and security policy Third, the EU judicature has applied a similar approach in H. v Council and Commission – a case brought by a staff member of the EU Police Mission in Bosnia and Herzegovina (EUPM), established under the CFSP.20 It thus reiterated that the scope of the limitation, by way of derogation, on the Court’s jurisdiction … cannot be considered to be so extensive as to exclude the jurisdiction of the EU judicature to review acts of staff management relating to staff members seconded by the Member States the purpose of which is to meet the needs of that mission at theatre level, when the EU judicature has, in any event, jurisdiction to review such acts where they concern staff members seconded by the EU institutions.21

The Court merely argued that on the basis of Article 270 TFEU, it had jurisdiction in any dispute between the Union and its servants ‘within the limits and under the conditions laid down in the Staff Regulations of Officials and the Conditions of Employment of other servants of the Union’, a provision which does not exclude CFSP-related disputes.22 It also found that its jurisdiction stemmed respectively, as regards the review of the legality of those acts, from Article 263 TFEU and, as regards actions for non-contractual liability, from Article 268 TFEU, read in conjunction with the second paragraph of Article 340 TFEU, taking into account Article 19(1) TEU and Article 47 of the Charter of Fundamental Rights.23

Importantly, Article 2 TEU and Article 21 TEU, to which Article 23 TEU relating to the CFSP refers, were invoked to recall that the European Union is founded, in particular, on the values of equality and the rule of law. The above-mentioned rulings confirm that the Court of Justice considers the CFSP as part and parcel of the Union’s constitutional set-up.24 The CFSP does interact with other EU policies and rules, resulting in the Court’s more complex judicial involvement in CFSP-related situations than the acknowledged jurisdiction envisaged under Articles 24(1) TEU and 275(2) TFEU. Clearly, a CFSP context forms no basis for the Court to disregard general principles of EU law or rules applicable in other policy areas. As it clearly indicated in the H case: ‘While the decisions adopted … have an operational aspect falling within the CFSP, they also constitute, by their very essence, acts of staff management, just like all similar decisions adopted by the EU institutions in the exercise of their competences.’25 This is nothing new. It not only brings back memories 20 See more extensively: P Van Elsuwege, ‘A Court of Justice Upholding the Rule of Law in the Common Foreign and Security Policy: H v. Council’ (2017) 3 CML Rev 841; T Verellen, ‘H v. Council: Strengthening the Rule of Law in the Sphere of the CFSP, One Step at a Time’ (2016) 3 European Papers 1041. 21 H (n 12). Similarly – at least as argued by AG Jääskinen in his Opinion of 21 May 2015, in Elitaliana (n 12) – the EU Courts should be able to hear individuals on budgetary issues, even if a particular decision was taken by an entity established under the CFSP. 22 ibid, para 57. An important argument, also in the context of the present chapter, is that jurisdiction of the CJEU would prevent possible diverging case law of this Court and the domestic courts in the countries of respective staff. 23 ibid, para 58. 24 C Hillion (n 2); RA Wessel, ‘Lex Imperfecta’ (n 2). See also Chapter 1 in this volume. 25 H (n 12), para 54.

‘The Good, the Bad and the Ugly’ 71 of early CFSP case law, such as Hautala on access to documents,26 it is also the application of a well-established case law of the Court on the scope of EU law and application of horizontal principles.27 It is not the purpose of this chapter to comment further upon the Court’s approach.28 In the present discussion, it suffices to underline that, notwithstanding the broad articulation and exercise of the Court’s jurisdiction in relation to the CFSP, gaps remain ‘as EU law stands’.29 The Court notably acknowledged in Opinion 2/13 that: ‘certain acts adopted in the context of the CFSP fall outside the ambit of judicial review by the Court of Justice’.30 Thus, only restrictive measures are subject to the CJEU legality control under the terms of Articles 24(1) TEU and 275(2) TFEU, and these have not been understood (as yet) in a particularly broad fashion to ensure compliance with the requirement of effective judicial protection across the CFSP field.31 Indeed, clarification is needed as regards the Court’s possible oversight of international agreements in the area of CFSP considering that, as mentioned earlier, they are negotiated and concluded in accordance with Article 218 TFEU, in relation to which the Court exercises full jurisdiction. Also, it remains uncertain whether, following the seminal Rosneft ruling, other courses of action, such as the interpretative function of the preliminary ruling procedure, or the action for damages, which play a role in ascertaining effective judicial protection, are available in the context of the CFSP, at least in situations involving acts for which the CJEU has jurisdiction.32 To be sure, the Court lacks jurisdiction in pure CFSP (or CSDP)33 situations, namely when disputes arise about decision-making procedures established in the CFSP chapter. This means, for instance, that the European Parliament can commence proceedings before the Court in cases where the Council has, for example, ignored its powers as envisaged in decision-making procedures set out in the specific CFSP chapter, if the CFSP competence is not otherwise disputed on the basis of Article 40(1) TEU. The interpretation and implementation of the CFSP provisions (including the procedures to be followed) in these situations is left to the Council (or even to individual Member 26

Case C-353/99 P, Council of the European Union v Heidi Hautala ECLI:EU:C:2001:661. See e.g. P Koutrakos, ‘Primary Law and Policy in EU External Relations – Moving Away from the Big Picture’ (2008) 33 EL Rev 666. 28 For various views on this case law see: Van Elsuwege (n 20); T Verellen, ‘H v. Council: Strengthening the Rule of Law in the Sphere of the CFSP, One Step at a Time’ (December 2016) European Papers – European Forum 1; S Øby Johansen, ‘H v. Council et al. – A Minor Expansion of the CJEU’s Jurisdiction over the CFSP’ (October 2016) European Papers – European Forum 1; S Poli, ‘The Common Foreign Security Policy After the Rosneft Ruling: Still Imperfect but Gradually Subject to the Rule of Law’ (2017) CML Rev 1799. 29 Opinion 2/13 (n 5), para 252. 30 ibid. 31 See in this respect the Commission’s views in the context of Opinion 2/13 (n 5), and the Court’s analysis of the provisions of the impugned decision in Rosneft (n 10), paras 75ff. 32 See in this respect: Opinion of AG Wathelet in the Rosneft case (n 10) at footnote 36; Case T-328/14 Jannatian v Council ECLI:EU:T:2016:86, paras 30–31; cf. Case T-602/15 Jenkinson v Council ECLI:EU:T:2016:660, para 45 – under appeal, see Case C-43/17 P. 33 For questions on contractual and non-contractual liability of the Union for CSDP operations, see Chapter 7 in this volume. 27

72 Research handbook on the EU’s common foreign and security policy States), unless the case is framed as a violation of Article 13(2) TEU, which would arguably allow the Court’s involvement.34 Remembering their initial preference for ‘intergovernmental’ cooperation where CFSP is concerned, it may be understandable that Member States intended at the time of the negotiations to prevent a body of ‘CFSP law’ coming into being by way of judicial activism on the part of the Court of Justice, but it is less understandable that they were also reluctant to allow for judicial control of the procedural arrangements they explicitly agreed upon. Though, admittedly, it may be difficult to separate procedures and content. Despite the limits to the Court’s jurisdiction, we may therefore not yet have seen the full picture. For instance, does the reference to Article 263(4) TFEU in Article 275 TFEU limit direct actions to those initiated by individuals, or can actions by the institutions or by Member States inter se be foreseen on the basis of a contextual interpretation by the Court? Earlier, we argued that there are good reasons to apply the principle of sincere cooperation (as currently formulated in Article 4(3) TEU) across the board, including CFSP.35 Despite the current absence of concrete cases, a use of this principle by the Court to settle procedural (or even substantive) conflicts between the institutions and the Member States or the Member States inter se, in a CFSP context cannot be ruled out.36 The recent case law taking general principles of EU law as a starting point only supports this assertion, although it remains clear that the CFSP context should be merely ‘incidental’, allowing the principle of sincere cooperation to be applicable. Gaps nevertheless remain in the CJEU’s control of the CFSP. This is all the more problematic since the Court has not attempted to dissuade the use of CFSP instruments since the entry into force of the Lisbon Treaty. Even if the CFSP course of action entails derogatory limits to judicial oversight which should be understood narrowly, the Court does not seem to develop a case law that gives preference to TFEU-based policies as a result. It has instead signalled that it is taking its post-Lisbon border patrolling function seriously (based on Article 40 TEU)37 deferring to the Treaty drafters’ 34

On Art. 13(2) TEU, see Hillion (n 14). Hillion and Wessel (n 1). 36 One may think of Member States concluding international agreements in areas covered by EU legislation or agreements, but perhaps even of conflicts on agreed (financial) contributions to CFSP/CSDP actions. 37 In 2012, the Court was given a first chance to develop an approach towards the function of Article 40 in Case C-130/10 Parliament v Council ECLI:EU:C:2012:472. It held that Article 215 TFEU (following a previous CFSP decision) rather than Article 75 TFEU (in the Area of Freedom, Security and Justice – AFSJ) was the correct choice, despite the limited role of the European Parliament in relation to the CFSP/Article 215 procedure. The context of peace and security proved to be decisive for the Court’s conclusion. Subsequent cases include Case C-658/11 (Mauritius) (n 12) and Case C-263/14 (Tanzania) (n 17). See more extensively Chapter 1 in this volume; C Matera and RA Wessel, ‘Context or Content? A CFSP or AFSJ Legal Basis for EU International Agreements – Case C-658/11, European Parliament v. Council (Mauritius Agreement)’ (2014) Revista de Derecho Comunitario Europeo 1047; Van Elsuwege (n 20); and C Hillion, ‘Fighting Terrorism through the CFSP’ in I Govaere and S Poli (eds), EU Management of Global Emergencies (Brill 2014). The more recent judgment Kazakhstan case (Case C-244/17 Commission v Council ECLI:EU:C:2018:662) however sheds some light on the limits of that approach. 35

‘The Good, the Bad and the Ugly’ 73 choice to keep the CFSP as a distinct policy framework involving specific institutional balance. The use of the CFSP procedures, and some implications in terms of limited CJEU oversight, are arguably here to stay as long as the Treaties are not modified. Having identified recurrent gaps in the CJEU control over the CFSP, one may then have to examine possible alternative avenues to ensure effective judicial remedies, as required by both Article 47 of the Charter of Fundamental Rights and Article 19 TEU. As the Court of Justice underlined in the H ruling, ‘the very existence of effective judicial review designed to ensure compliance with provisions of EU law is inherent in the existence of the rule of law’.38 In Opinion 2/13, the Court, however, made clear its reluctance in accepting that an international Court, in casu the European Court of Human Rights (ECtHR), could have jurisdiction in relation to CFSP acts over which it could not itself exercise control. It also suggested, although implicitly, that Member States’ courts did not have a role to play. The next two sections further discuss this approach.

3. ‘THE BAD’: INTERNATIONAL/EXTERNAL JUDICIAL CONTROL When introduced by the 1992 Maastricht Treaty, CFSP was often seen as falling outside EU law. In fact, some early publications viewed the then newly established second pillar as international law (if the rules were considered legal at all).39 While the present authors have maintained that CFSP has always been part and parcel of the EU’s legal order,40 since the entry into force of the Lisbon Treaty it seems even more difficult to argue otherwise. The question therefore arises as to whether Article 344 TFEU also applies to disputes on the basis of CFSP acts and obligations and, if so, to what extent. Article 344 TFEU provides that ‘Member States undertake not to submit a dispute concerning the interpretation or application of the Treaties to any method of settlement other than those provided for therein’. This provision has continuously been used by the CJEU to claim its exclusive jurisdiction whenever the interpretation or application of EU law is at stake,41 and the Court’s case law has developed the conditions under which participation of the EU in international dispute settlement systems can be allowed.42 38

See the H case (n 12), para 41; also in Rosneft (n 10), para 77. See for an early analysis of those studies: RA Wessel, The European Union’s Foreign and Security Policy: A Legal Institutional Perspective (Kluwer Law International 1999). 40 See for instance RA Wessel, ‘The Dynamics of the European Union Legal Order: An Increasingly Coherent Framework of Action and Interpretation’ (2009) 1 European Constitutional Law Review 117; C. Hillion, ‘Tous pour un, Un pour tous! Coherence in the External Relations of the European Union’ in M Cremona (ed.), Developments in EU External Relations Law (OUP 2008). 41 See further below and for instance Opinion 1/91 ECLI:EU:C:1991:490, para 35 and Opinion 1/00 ECLI:EU:C:2002:231, paras 11–12; judgments in C-459/03 Commission v Ireland ECLI:EU:C:2006:345, paras 123 and 136, and Kadi and Al Barakaat International Foundation v Council and Commission ECLI:EU:C:2008:461, para 282. 42 See C Hillion and RA Wessel, ‘The European Union and International Dispute Settlement: Mapping Principles and Conditions’ in M Cremona, A Thies and RA Wessel (eds), The 39

74 Research handbook on the EU’s common foreign and security policy Following the present chapter’s metaphor, external judicial involvement has in effect been viewed as ‘bad’ in the eyes of the Court, despite its declared openness towards international jurisdiction.43 Indeed, in Opinion 2/13, the CJEU confirmed the complexities related to the EU’s submission to external judicial scrutiny.44 In answering the question of whether the Union could join the ECHR, the Court pointed to a number of (classic) principles and conditions inherent in the nature of EU law, which in effect encapsulate the difficulties of a combination of EU law and international dispute settlement. For the purpose of the present chapter it is relevant to point to specific paragraphs in which the Court addresses the external judicial review of CFSP measures. As stated earlier, it expressed its displeasure with the idea that the ECtHR would be able to rule on the compatibility with the ECHR of ‘certain acts, actions or omissions performed in the context of the CFSP’ which ‘fall outside the ambit of judicial review by the Court of Justice’.45 The issue was also referred to by Advocate General Kokott when she argued that ‘accession to the ECHR will undoubtedly mean that the EU must respect the fundamental rights protection that stems from the ECHR – and thus also the requirement of effective legal protection in accordance with Articles 6 and 13 ECHR – in all its spheres of activity, including the CFSP’.46 The Opinion also raised the key question of whether the legal protection in the CFSP afforded by the EU legal order could be regarded as effective legal protection for the purposes of Articles 6 and 13 ECHR. The Commission’s view on this point is noteworthy. In the words of AG Kokott: It proposes that [Article 275 TFEU] be understood as meaning that the Court of Justice of the EU not only has jurisdiction over actions for annulment brought by individuals against restrictive measures, but it may in addition deal with actions for damages and reply to requests for preliminary rulings from national courts or tribunals in the sphere of the CFSP. It also advocates handling the options for the legal protection of individuals in the CFSP in such a way as to cover not only acts, within the meaning of the first paragraph of Article 263 TFEU, which produce binding legal effects, but also mere ‘material acts’ (Realakte), that is to say, acts without legal effects.47 European Union and International Dispute Settlement (Hart Publishing 2017); as well as T Lock, The European Court of Justice and International Courts (OUP 2015). 43 Opinion 1/76, European Laying-up Fund for Inland Waterways ECLI:EU:C:1977:63. Further: see Hillion and Wessel (n 42); see also: B De Witte, ‘A Selfish Court? The Court of Justice and the Design of International Dispute Settlement Beyond the European Union’ in M Cremona and A Thies (eds), The European Court of Justice and External Relations Law: Constitutional Challenges (Hart Publishing 2014) 33. 44 Opinion 2/13 (n 5). See also View of AG Kokott, ECLI:EU:C:2014:2475. For an academic appraisal see, inter alia, A Łazowski and RA Wessel, ‘When Caveats turn into Locks: Opinion 2/13 on Accession of the European Union to the ECHR’ (2015) 16(1) German Law Journal 179; as well as A Łazowski and RA Wessel, ‘The European Court of Justice Blocks the EU’s Accession to the ECHR’ (CEPS Commentary, 8 January 2015) . 45 Opinion 2/13 (n 5), para 252. 46 View of AG Kokott (n 44), para 83. 47 ibid, para 86.

‘The Good, the Bad and the Ugly’ 75 While the Commission may be complimented for the daring view that in certain circumstances the extended jurisdiction of the Court flows from the post-Lisbon EU legal order,48 Advocate General Kokott was not convinced. Indeed, and perhaps even more interestingly, she argued: the very wide interpretation of the jurisdiction of the Courts of the EU which it proposes is just not necessary for the purpose of ensuring effective legal protection for individuals in the CFSP. This is because – the entirely accurate – assertion that neither the Member States nor the EU institutions can avoid a review of the question whether the measures adopted by them are in conformity with the Treaties as the basic constitutional charter does not necessarily always have to lead to the conclusion that the Courts of the EU have jurisdiction.49

As will be discussed in the following section, the reason would be that ‘national courts or tribunals have, and will retain, jurisdiction’.50 Despite this option, the question remains – as acknowledged by Advocate General Kokott – of whether effective legal protection in relation to the CFSP can be provided by the EU’s multilevel system itself. And it is this situation that forms the source of the Court’s worries as it would perhaps give room to a ‘non-EU body’ to exercise powers that were consciously left out of the EU Treaties for the CJEU itself. In Opinion 2/13, the Court also aimed to prevent other courts from considering possible human rights violations. But it is doubtful whether this claim can be made. The choice by the EU Treaty drafters at the time to maintain a special position for many CFSP norms as far as their judicial review is concerned does not imply that possible human rights violations in relation to CFSP actions should in general be exempt from judicial scrutiny. Arguably, the reason for the special arrangement was rather to prevent judicial activism in this area of EU competence. These days, the gap in the judicial control over CFSP – which, as we have seen, is recognized by the Court itself – leads to examining the extent to which courts outside the EU can have a role in the judicial scrutiny of CFSP. Given the link between CFSP and other external EU action, CFSP matters could in theory be subject to dispute resolution before several different tribunals.51 However, we will focus first on the European Court of Human Rights, second on the International Court of Justice and third on the national courts of third states. With regard to a possible role for the ECtHR, we have seen that Opinion 2/13 does not exclude it. In fact, this was one of the reasons for the CJEU to advise against the Accession Agreement. In the words of the Court of Justice, the ECtHR could rule on the compatibility with the ECHR of ‘certain acts, actions or omissions performed in the context of the CFSP’.52 Obviously – as the EU is not (yet) a party to the ECHR – these could only be actions by the Member States. It is also clear that the ECtHR will only examine compatibility with the ECHR and cannot function as a tribunal to supervise or 48

C. Hillion, ‘A Powerless Court?’ (n 2) and RA Wessel ‘Resisting Legal Facts’ (n 2). AG View (n 44), para 95. 50 ibid, para 96. 51 For example, EU maritime operations such as EUNAVFOR Somalia/Atalanta, could trigger questions related to the law of the sea which could end up before the International Tribunal for the Law of the Sea (ITLOS). 52 Opinion 2/13 (n 5), para 7. 49

76 Research handbook on the EU’s common foreign and security policy enforce the implementation of CFSP obligations as such. In that sense, the role of the Strasbourg Court in judicial control over CFSP is limited, although it may be confronted with questions on the interpretation and application of EU law. In the context of the present analysis, the current role of the ECtHR is thus circumscribed to possible violations of the Convention by states in the implementation of CFSP, including CSDP. While this role of the Strasbourg Court is similar to that exercised in other policy areas of the EU, an important difference lies in the fact that the Court’s case law has traditionally taken the possibilities for judicial scrutiny by the EU into account when dealing with possible violations of the Convention by Member States related to an implementation of EU decisions. The well-known Bosphorus and Matthews case law serves as the basis for this line of thinking,53 and the judicial protection offered by the EU has generally been seen as providing an adequate alternative. On the basis of this ‘arrangement’ it has been argued that ‘conflicts have been rare, but the threat was ever present’.54 Obviously, this threat is less evident when the CJEU itself cannot offer the full review of CFSP measures. In general, there are no reasons to question the authority of the ECtHR to assess acts or conduct of EU Member States in relation to external action of the European Union. As long as the Union is not a party to the ECHR, it will obviously not be able to breach it, even though the Union’s missions and other external actions are based on EU decisions.55 According to the Behrami and Saramati case law, the Union could nevertheless be responsible for these violations when their authors are acting on behalf of the Union and are under its control; which in turn has consequences for the responsibilities for its Member States (i.e. the parties to the ECHR).56 In recent years, the rules governing the attribution of wrongful acts committed in the context of peace support operations have been the subject of intense discussion, in particular since the Strasbourg case law mentioned above. Although this debate has demonstrated that academic opinion insists on a high level of factual control for holding states and

53 Matthews v United Kingdom (1999) 28 EHRR 361; Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v Ireland (2006) 42 EHRR 1. 54 See JP Jacqué, ‘The Accession of the European Union to the European Convention on Human Rights and Fundamental Freedoms’ (2011) CML Rev 995, 1001. 55 The question of international responsibility of the EU for CFSP actions is still under debate. We would point to the fact that CFSP, like other Union policies, is based on EU decision, pointing towards the responsibility of the EU as such. See Chapter 7 in this volume as well as A Sari and RA Wessel, ‘International Responsibility for EU Military Operations: Finding the EU’s Place in the Global Accountability Regime’ in B Van Vooren, S Blockmans and J Wouters (eds), The EU’s Role in Global Governance: The Legal Dimension (OUP 2013). For others: ‘it would be difficult to argue that an action committed by a Member State when complying with a CFSP measure is to be understood as having been committed by the EU by virtue of its normative control’; see A Delgado Casteleiro, The International Responsibility of the European Union: From Competence to Normative Control (CUP 2016). 56 Behrami v France and Saramati v France, Germany and Norway (2007) 45 EHRR SE10. See also Jacqué (n 54) and A Sari, ‘Autonomy, Attribution and Accountability: Reflections on the Behrami Case’ in ND White and R Collins (eds), International Organizations and the Idea of Autonomy: Institutional Independence in the International Legal Order (Routledge 2011).

‘The Good, the Bad and the Ugly’ 77 international organizations responsible for the conduct of peace operations,57 it has stopped short of addressing the underlying question of whether factual control is the only relevant ground of attribution in this context. In general, states usually transfer only limited powers of operational control over their forces to international organizations, and retain supreme authority, known as full command, for themselves.58 The armed forces of a state thus never lose their institutional status as state organs during their secondment to an international organization and could thus remain responsible even if their actions take place in the context of a military operation initiated by an international organization. However, within the EU context, responsibility questions seem more complex. As a distinct chapter in this volume is specifically devoted to the responsibility and liability for CSDP operations,59 we will limit ourselves to a few general observations with regard to the possible role of external courts. The legal status of the missions and their staff is regulated in Status of Forces Agreements (SOFAs) or Status of Mission Agreements (SOMAs). The absence of an extensive practice60 makes it difficult to present concrete examples, especially since, as stated by Heliskoski ‘virtually all claims lodged under SOFAs and SOMAs in the context of CSDP operations are sorted out by means of amicable settlement’. First of all, civilian missions now have an accepted distinct legal capacity, albeit under EU law only.61 In general, it has been argued that if it can be established that EU military missions constitute ‘subsidiary organs’ of the EU, a rebuttable presumption may be said to exist in favour of attributing their wrongful conduct to the Union, rather than to the contributing states.62 Since all legal acts relating to the launch, conduct and termination of EU military operations are adopted by the Council of the EU,63 the latter is the only EU institution capable, in principle, of establishing military operations as its subsidiary organs (compare the arguments 57

See, for instance, K Mujezinovic´ Larsen, ‘Attribution of Conduct in Peace Operations: The “Ultimate Authority and Control Test”’ (2008) 19 European Journal of International Law 509. 58 See, for instance, B Cathcart, ‘Command and Control in Military Operations’ in T Gill and D Fleck (eds), The Handbook of the International Law of Military Operations (OUP 2010). See also S Johansen, The Accountability Mechanisms of International Organisations – A Framework and Three Case-Studies (Oslo University 2017) esp. 143–201. 59 Chapter 7 in this volume. 60 See F Naert, ‘Shared Responsibility in the Framework of the EU’s CSDP Operations’ in A Nollkaemper and I Plakokefalos (eds), The Practice of Shared Responsibility in International Law, Vol. III (CUP 2017) 3689. 61 These missions are now endowed with a legal personality under EU law, including ‘the capacity to … be a party to legal proceedings, as required in order to implement [the] Joint Action [concerned]’. See, for instance, Art. 15 bis of Council Joint Action 2008/124/CFSP on the European Union Rule of Law Mission in Kosovo, EULEX KOSOVO [2008] OJ L 340/73, as amended by Council Decision 2014/349/CFSP [2014] OJ L 174/42. See further Chapter 7 in this volume. 62 Sari and Wessel (n 55). See also Heliskoski (Chapter 7 in this volume): ‘a presumption that responsibility under public international law for the conduct of CSDP operation or missions is borne by the EU in its own right rather than by the Member States’. 63 cf. F Naert, ‘Legal Aspects of EU Military Operations’ (2011) Journal of International Peacekeeping 218; P Koutrakos, The EU Common Security and Defence Policy (OUP 2013).

78 Research handbook on the EU’s common foreign and security policy used in the H case mentioned above). The decision-making procedures or voting rules (unanimity) do not affect the nature of the Council Decision.64 This seriously limits the possibilities for the ECtHR to scrutinize CFSP/CSDP actions by EU Member States to situations in which Member States would not act on behalf of the Union, but would, for instance, go beyond their mandate. More generally, Naert has pointed to a number of complexities in the application of human rights law to CSDP operations: ‘the extraterritorial application of the European Convention on Human Rights, the question of derogation in times of emergencies and its applicability to peace operations, the relationship between human rights and international humanitarian law and the impact of UN Security Council mandates on human rights’.65 Again, these factors limit the possibilities of seeing the ECtHR as being able to fill possible gaps in the judicial control of CFSP/CSDP. This brings us to the question of whether other international courts could play a role in dealing with CFSP questions. Given the subject matter, the International Court of Justice (ICJ) could be an obvious candidate. In the course of the development of the CFSP, this option has occasionally been mentioned, but also rejected. As held by Denza in 2002: Although reference is sometimes made to the possibility to adjudication of disputes between Member States by the International Court of Justice, this is clearly not a realistic option for most disputes given the time scale required for the ICJ to reach a decision. The actual use by Member States of the ICJ to adjudicate on the intergovernmental provisions of the TEU would, moreover, certainly undermine the role of the ECJ as a constitutional guarantor of the legal order of the European Union – in particular the requirement of unity and consistency.66

Leaving aside the reference to ‘intergovernmental provisions’ in the context of the EU in this respect, the second part of the argument indeed makes sense. It is important to recall that in this context Article 344 TFEU also seems to exclude the possibility for Member States to settle potential CFSP disputes before the ICJ, despite the restrictive role of the CJEU as mentioned in Article 275 TFEU. As was made clear in Opinion 2/13, the prohibition on submitting a dispute to another court is not dependent on the CJEU itself having jurisdiction in that particular field.67 This leaves us with the question of whether third states may initiate proceeding before the International Court and sue EU Member States in cases of, for instance, disagreements related to international agreements concluded in the area of CFSP or CSDP. While, as we have seen, the exclusivity of the Court’s jurisdiction would usually prevent EU Member States from settling disputes involving the interpretation of EU law elsewhere, it is more difficult to find legal arguments to prevent third states from initiating international legal proceedings. As the EU as such has no standing before the

64 This is not to say that Member State liability is always excluded. For a detailed analysis, see Chapter 7 in this volume. 65 Naert (n 63) 237. 66 E Denza, The Intergovernmental Pillars of the European Union (OUP 2002) 322. 67 See also Lock (n 42) 88.

‘The Good, the Bad and the Ugly’ 79 ICJ, these actions could only involve the Member States.68 So far, however, international agreements in the area of CFSP/CSDP have been concluded by the EU alone. Again, the fact that the Council decides on the basis of unanimity does not change the nature of the Decision adopting the agreement. This would only be the case if the Decision were not taken by the Council as such, but by Member States’ representatives acting ‘in the framework of the Council’.69 Perhaps ironically, mixity is not used in this area and hence Member States do not bear individual responsibilities under international law, but only through Union law. This is not to say that CFSP issues may, theoretically, not come up in ICJ proceedings.70 Again one may think of situations in which EU Member States have allegedly violated international law in the margins of activities which fall within the framework of an EU operation. The facts of the case will be decisive in establishing their international responsibilities. One may also think of political decisions on, for instance, individual sanctions which could come into conflict with existing obligations under international diplomatic law. While for EU Member States this could lead to conflicting obligations under EU (CFSP) law and international law, nothing seems to stand in the way of third states using available ICJ procedures to hold other states responsible, even when these other states happen to be EU members. So far, however, practice has not provided any concrete examples. Apart from the ICJ, other international courts could be confronted with questions related to CFSP. One example could be the International Criminal Court in the event of a violation by nationals of EU Member States of the rules on genocide, crimes against humanity and war crimes during the operations of CSDP missions. At present, all EU Member States are States Parties to the Rome Statute. It is also important to remember that the EU and its Member States accept that if EU-led forces become a party to an 68 cf. the cases before the ICJ of Serbia and Montenegro (then composite republics of one single state) against eight NATO member states (Belgium, Canada, France, Germany, Italy, the Netherlands, Portugal and the United Kingdom), asking the ICJ to hold each of the respondent states responsible for international law violations stemming from the NATO bombing campaign against Yugoslavia in March–April 1999. Here also, Serbia and Montenegro chose to sue a number of Member States irrespective of the fact that the actions were based on a NATO decision and also coordinated by NATO. As, according to the ICJ, Serbia and Montenegro, lacked standing, the Court did not have a chance to consider the merits of the case. The cases can be accessed through the website of the ICJ: . 69 We are not aware of existing examples of CFSP Decisions taken in the framework of the Council. Yet, a recent external relations example is the Decision of the Heads of State or Government of the 28 Member States of the European Union, meeting within the European Council, on the Association Agreement between the European Union and the European Atomic Energy Community and their Member States, of the one part, and Ukraine, of the other part, annexed to the European Council Conclusions on Ukraine of 15 December 2016. 70 In general, EU law has not played a role in disputes before the ICJ. Only a very indirect reference to EU law may perhaps be found in the ICJ judgment on the Dispute regarding Navigational and Related Rights (Costa Rica v Nicaragua), where the CJEU’s view on commercial policy as presented in its Opinion 1/76 was accepted by the ICJ. See more extensively RA Wessel, ‘Flipping the Question: The Reception of EU Law in the International Legal Order’ (2016) Oxford Yearbook of European Law 533. While there are indeed no practical examples at this moment, the Gabcíkovo-Nagymaros (Hungary v Slovakia) case, for instance, revealed that international rules may collide with EU obligations.

80 Research handbook on the EU’s common foreign and security policy armed conflict, international humanitarian law will apply to them fully.71 Specific situations could furthermore lead to proceedings before specialized tribunals, such as the International Tribunal for the Law of the Sea (ITLOS), whenever law of the sea disputes arose during or as a result of a CSDP mission. Obviously, however, these disputes could not relate to interpretation of EU law, but merely of – in this case – the rules on the law of the sea. Finally, could one envisage CFSP disputes being brought to and settled by national courts of third states, for instance states where the EU is active through military missions? Again, it is important to underline the obvious: not just because of Article 344 TFEU, but also because of jurisdictional problems, it would not be possible for Member States inter se to have their disputes settled by those courts. However, there is no reason in principle why local or national courts would not have the competence to deal with disputes between third states or local actors and Member States, for instance in relation to the responsibility for wrongful acts committed by CFSP missions or alleged violations of domestic or even international law. In the case of CSDP missions, the SOFAs would generally mention ‘respect’ for local law, although the view is usually taken that ‘respect’ does not mean ‘comply with’ and therefore imposes lower standards.72 Moreover, the application of domestic law would generally be ruled out in the provisions in the agreement dealing with the privileges and immunities of the EU mission.73 While it is not unusual in international law to confer diplomatic privileges and immunities on foreign military and civilian personnel, it has been noted that in the case of EU missions the privileges and immunities are much more extensive and are not only conferred on the higher-ranking staff.74 The EU Model SOFA and SOMA75 grant EU missions’ personnel immunity from the criminal jurisdiction of the host state ‘under all circumstances’.76 At the same time, the sending states retain ‘all the criminal jurisdiction and disciplinary powers conferred on them by the law of the Sending State’.77 Furthermore, the missions’ personnel are exempted from the civil and administrative jurisdiction of the host state ‘in respect of words spoken or written and all acts performed by them in the exercise of their official functions’.78 A role for domestic courts in third states cannot be fully excluded in cases related to so-called ordinary activities, such as driving for private reasons or contracting on the local 71

F Naert, International Law Aspects of the EU’s Security and Defence Policy, with a Particular Focus on the Law of Armed Conflict and Human Rights (Intersentia 2010) 463–540. 72 Naert (n 63) 240. 73 See on these agreements P Koutrakos, The EU Common Security and Defence Policy (OUP 2013) 198–202; Naert (n 63); A Sari, ‘Status of Forces and Status of Mission Agreements under the ESDP: The EU’s Evolving Practice’ (2008) European Journal of International Law 67. 74 Sari (n 73) 78–79; also Naert (n 63) 250, calling the situation ‘unusual’. 75 See Draft Model Agreement on the status of the European Union-led forces between the European Union and a Host State, Council of the European Union (SOFA), 20 July 2007, Doc. 11894/07; and Draft Model Agreement on the Status of the European Union Civilian Crisis Management Mission in a Host State (SOMA), Council of the European Union, 15 December 2008, Doc. 17141/08. 76 Art. 6(3) EU Model SOFA; Art. 6(3) EU Model SOMA. 77 Art. 8 EU Model SOFA; Art. 8 EU Model SOMA. 78 Art. 6(4) EU Model SOFA; Art. 6(4) EU Model SOMA.

‘The Good, the Bad and the Ugly’ 81 market. With regard to lawsuits in third countries, practice offers a variety of situations, including the following: traffic incidents involving EU Delegations’ staff (where in each case the EU examines whether or not to lift immunity for the purpose of local proceedings); criminal proceedings against an international contracted staff member of an EU mission, where the local authorities imprison the person in question, in clear violation of the relevant provisions of the Status of Mission Agreement (but where the host country reminded the EU that the SOMA also calls for mission staff to respect local laws and where the staff member could only be released after some diplomatic effort); and disputes about whether an employment contract with local personnel was concluded by the Head of Delegation in his private or official capacity.79 Overall, however, the role of international courts and courts of third countries in relation to CFSP seems to be limited by the CJEU’s wide interpretation of Article 344 TFEU (as in the case of the ECtHR), the lack of standing of the EU (the ICJ) or the quite extensive rules on EU immunities (courts in third states). The H case also revealed that the proceedings initiated by Ms H before Italian courts gave the CJEU no reason to reconsider the division of judicial tasks.80 This brings us to the question of what role, if any, Member States’ courts could then play in relation to the CFSP.

4. ‘THE UGLY’: MEMBER STATES’ COURTS As is well established, Member States’ systems of remedies are integrated in the EU judicial system. According to Article 19 TEU, Member States must provide remedies sufficient to ensure effective legal protection in the fields covered by Union law.81 Yet, in the eyes of the Court of Justice, it would be difficult (or perhaps ‘ugly’) to allow domestic courts to play a leading role in EU law without at least a harmonizing role for the Court itself. As mentioned in Rosneft: The Court must reject the argument that it falls to national courts and tribunals alone to ensure effective judicial protection if the Court has no jurisdiction to give preliminary rulings on the validity of decisions in the field of the CFSP that prescribe the adoption of restrictive measures against natural or legal persons.82

The Court of Justice has further spelled out the role that national courts are to play in ensuring that in the interpretation and application of the Treaties, the law is observed. Thus, in its Opinion on the Unified Patent Court,83 it held:

79 Informal note issued by the European Commission in March 2010 (available online at ). See more extensively on these examples: RA Wessel, ‘Immunities of the European Union’ (2014) International Organizations Law Review 395. 80 H (n 12), para 19. 81 This section builds on Hillion (n 1). 82 See Rosneft (n 10), para 77. 83 Opinion 1/09 Unified Patent Court ECLI:EU:C:2011:123.

82 Research handbook on the EU’s common foreign and security policy As is evident from Article 19(1) TEU, the guardians of [the] legal order and the judicial system of the European Union are the Court of Justice and the courts and tribunals of the Member States … . It should also be observed that the Member States are obliged, by reason, inter alia, of the principle of sincere cooperation, set out in the first subparagraph of Article 4(3) TEU, to ensure, in their respective territories, the application of and respect for European Union law … . Further, pursuant to the second subparagraph of Article 4(3) TEU, the Member States are to take any appropriate measure, general or particular, to ensure fulfilment of the obligations arising out of the Treaties or resulting from the acts of the institutions of the European Union. In that context, it is for the national courts and tribunals and for the Court of Justice to ensure the full application of European Union law in all Member States and to ensure judicial protection of an individual’s rights under that law. The national court, in collaboration with the Court of Justice, fulfils a duty entrusted to them both of ensuring that in the interpretation and application of the Treaties the law is observed (emphases added).

Based on this general statement, it is arguable that as co-‘guardians of [the] legal order and the judicial system of the European Union’, Member States’ courts and tribunals should be called upon to ensure compliance with provisions of EU law in the context of the CFSP where the CJEU itself does not have jurisdiction. Nothing in the Treaties suggests that the restrictions applicable to Court of Justice’s powers, based on Articles 24(1) TEU and 275(1) TFEU, concern in any way the jurisdiction of Member States’ courts. On the contrary, Article 274 TFEU stipulates: ‘Save where jurisdiction is conferred on the Court of Justice of the European Union by the Treaties, disputes to which the Union is a party shall not on that ground be excluded from the jurisdiction of the courts or tribunals of the Member States.’ In this sense, Advocate General Kokott underlined in her View in the Opinion procedure 2/13, that ‘this follows from the principle of conferral, according to which competences not conferred upon the EU in the Treaties remain with the Member States’.84 The restricted jurisdiction of the Court of Justice per Articles 24(1) TEU and 275(2) TFEU should thus involve the commensurate involvement of Member States’ judiciaries precisely to offset the Court’s inability to ensure that the law is observed in the interpretation and application of some aspects of the CFSP. Article 19 TEU indeed points to this complementary role, inspired by the Court of Justice’s case law,85 when requiring Member States to ‘provide remedies sufficient to ensure effective legal protection in the fields covered by Union law’, to avoid ‘a lacuna … in the legal protection system’,86 and thus to fulfil the requirement of Article 47 of the EU Charter of Fundamental Rights (CFR).87 Given that the Court of Justice itself cannot provide 84

See View of AG Kokott (n 44), esp. para 96. For an academic appraisal see, inter alia, Łazowski and Wessel (n 44). 85 See e.g. Case C-583/11 Inuit Tapiriit Kanatami and Others v Parliament and Council ECLI:EU:C:2013:62; Case C-50/00 P Unión de Pequeños Agricultores v Council ECLI:EU:C: 2002:462. 86 View of AG Kokott (n 44), para 85. 87 Art. 47 CFR stipulates: ‘Everyone whose rights and freedoms guaranteed by the law of the Union are violated has the right to an effective remedy before a tribunal in compliance with the conditions laid down in this Article.

‘The Good, the Bad and the Ugly’ 83 legal protection, the notion of sufficiency entails that it is for the Member States to provide effective remedies. Advocate General Kokott discussed the role of national judiciaries extensively in her ECHR Opinion.88 The Court, by contrast, did not. In mentioning that ‘accession would effectively entrust the judicial review of those acts, actions or omissions on the part of the EU exclusively to a non-EU body’ (emphasis added),89 it suggested instead, albeit obliquely, that Member States’ courts are not able to review the legality of CFSP acts, even those that fall outside its jurisdiction. For the Court of Justice, its exclusion from certain aspects of the CFSP sphere is seemingly tantamount to an exclusion of the whole EU judicial system, including Member States’ courts as EU courts, despite the express provision of Article 274 TFEU, the unequivocal language of Opinion 1/09, and the obligations enshrined in Article 19 TEU. Admittedly, allowing Member States’ courts to review the legality of certain EU acts would undoubtedly complicate the functioning of the EU legal order. This is a well-known concern for the Court of Justice, which was forcefully expressed in its Foto-Frost judgment in which it concluded that those courts do not have the power to declare acts of the Community institutions invalid. As the Court emphasized in the judgment of 13 May 1981 in Case 66/80 International Chemical Corporation v Amministrazione delle Finanze [1981] ECR 1191, the main purpose of the powers accorded to the Court by Article [267 TFEU] is to ensure that Community law is applied uniformly by national courts. That requirement of uniformity is particularly imperative when the validity of a Community act is in question. Divergences between courts in the Member States as to the validity of Community acts would be liable to place in jeopardy the very unity of the community legal order and detract from the fundamental requirement of legal certainty. The same conclusion is dictated by consideration of the necessary coherence of the system of judicial protection established by the Treaty. In that regard it must be observed that requests for preliminary rulings, like actions for annulment, constitute means for reviewing the legality of acts of the community institutions … . Since Article [263] gives the Court exclusive jurisdiction to declare void an act of a Community institution, the coherence of the system requires that where the validity of a community act is challenged before a national court the power to declare the act invalid must also be reserved to the Court of Justice.90

The invalidation of CFSP acts by Member States’ courts would have implications comparable to those evoked in Foto-Frost as regards the unity of the EU legal order and Everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal previously established by law. Everyone shall have the possibility of being advised, defended and represented. Legal aid shall be made available to those who lack sufficient resources in so far as such aid is necessary to ensure effective access to justice.’ 88 View of AG Kokott (n 44), esp. paras 96–103. 89 See Opinion 2/13 (n 5), para 255. It reiterated that point in the following paragraph. 90 Case C-314/85 Foto Frost v Hauptzollamt Lübeck-Ost ECLI:EU:C:1987:452. For a recent reiteration of the doctrine it contains, see e.g. Case C-362/14 Maximillian Schrems v Data Protection Commissioner ECLI:EU:C:2015:650, and Rosneft (n 10).

84 Research handbook on the EU’s common foreign and security policy legal certainty. This could indeed explain the Court’s implicit position on Member States’ courts in Opinion 2/13. That said, how could the Foto-Frost solution operate in a situation where the Court of Justice has no jurisdiction? How can it guarantee the unity of the EU legal order, and particularly the uniformity of application of CFSP rules, if the Court cannot review those rules in the first place? Arguably, ‘the necessary coherence of the system of judicial protection established by the Treaty’ requires that if the Court does not have the ‘jurisdiction to declare void’ certain CFSP acts, it cannot claim the power to declare such acts invalid, and a fortiori that such power be reserved to it. The application of the Foto-Frost doctrine presupposes the CJEU’s jurisdiction. Thus, Advocate General Kokott considered: [I]n the context of the CFSP, the Court of Justice cannot claim its otherwise recognised monopoly on reviews of the legality of the activities of EU institutions, bodies, offices and agencies. The settled case-law of the Court, stemming from the judgment in Foto-Frost, cannot, therefore, in my view, be applied to the CFSP. Unlike in supranational areas of EU law, there is no general principle in the CFSP that only the Courts of the EU may review acts of the EU institutions as to their legality.91

Admittedly, the Foto-Frost doctrine does apply to certain CFSP-related situations. Member States’ courts are thus precluded from invalidating CFSP acts that fall under the Court of Justice’s jurisdiction. The Court made that point clear in the Rosneft judgment: The necessary coherence of the system of judicial protection requires, in accordance with settled case-law, that when the validity of acts of the European Union institutions is raised before a national court or tribunal, the power to declare such acts invalid should be reserved to the Court under Article 267 TFEU (see, to that effect, judgments of 22 October 1987, Foto-Frost, 314/85, EU:C:1987:452, paragraph 17, and of 6 October 2015, Schrems, C362/14, EU:C:2015:650, paragraph 62). The same conclusion is imperative with respect to decisions in the field of the CFSP where the Treaties confer on the Court jurisdiction to review their legality (emphasis added).92

The tenets of the Foto-Frost jurisprudence, applied to the CFSP context, are therefore strongly reaffirmed – the judgment indeed spells them out almost in full. But this application is envisaged only where it itself has jurisdiction. One may thus infer from the above dictum that for CFSP-related cases falling outside the scope of Article 275(2) TFEU, by contrast, Member States’ courts are able to exercise what remains their judicial power. To be sure, Member States’ judicatures can always invalidate unlawful national measures taken in the context of a CFSP act.93 EU principles and rules, including the 91

View of AG Kokott (n 44), para 100. See Rosneft (n 10), para 78. 93 The High Administrative Court of Nordrhein Westfalen was asked to rule on the alleged responsibility of Germany for the transfer of suspected Somali pirates to Kenya, carried out in the framework of the EUNAVFOR Atalanta mission (Oberverwaltungsgericht NRW, 4 A 2948/11, 18 September 2014). For an insightful analysis of this case see: E Sommario, 92

‘The Good, the Bad and the Ugly’ 85 CFR, are then of relevance given that the Member State would be acting within the scope of EU law within the meaning of Article 51(1) CFR. But beyond the national implementation measures, Member States’ courts, qua EU courts, are arguably the only EU judicature able to control the validity of CFSP acts as such, though possibly with the Court of Justice’s aid. In particular, the Court may assist the national judge’s review of a CFSP act, or its national implementation, through the preliminary ruling procedure. In particular, it may provide an interpretation of any EU law, such as a provision of the Charter or a provision on the EU institutions’ essential role enshrined in Title III TEU, which would be relevant for deciding on the case at hand.94 After all, the limits enshrined in Article 275(2) TFEU cannot entail restrictions on the Court’s jurisdiction in relation to other (that is, non-CFSP) domains of EU law without potentially breaching the rule of Article 40(1) TEU, while negating the exceptional nature of the judicial arrangements of Article 275 TFEU, and their consequent narrow interpretation.95 In sum, there are legal elements to support Member State courts’ involvement, as EU courts, in exercising complementary judicial control over the CFSP, where and as long as the Court of Justice is not allowed to exercise it itself.96 That this approach involves complications for the functioning of the legal order cannot in itself disqualify the only judicial protection against CFSP acts that is available under EU law as it stands. The contrary would amount to a denial of legal protection which would be equally problematic for the EU legal order, based as it is on the rule of law.97 Indeed, the implications of a decentralized judicial control of the CFSP might be less damaging for the EU legal order than a judicial review by national courts limited to the domestic implementation measures. While in the latter case, national courts would be adjudicating by reference to national and EU law, in the former situation, they would ‘Attribution of Conduct in the Framework of CSDP Missions: Reflections on a Recent Judgment by the Higher Administrative Court of Nordrhein Westfalen’ in S Poli (ed.), Protecting Human Rights in the European Union’s External Relations (CLEER paper 2016/5). 94 AG Wahl considered in the H Case that when the CJEU does not have jurisdiction it is for the national courts ‘to examine the lawfulness of the contested decisions and rule on the related claim for damages’ (para 89). In doing so, they may have to ask preliminary questions: ‘90. … it cannot be excluded that the competent national courts may have doubts as to the extent of their review of the contested decisions as well as on the possible consequences of that review. 91. Should that be the case, I would remind those courts that they are at liberty – and they may sometimes be obliged – to submit a request for a preliminary ruling to the Court under Article 267 TFEU. In that connection, the Court may still be able to assist those courts in deciding the case before them, while remaining within the boundaries established by Articles 24(1) TEU and 275 TFEU. It occurs to me that such requests for a preliminary ruling ought to be welcomed …’. 95 As mentioned above, the Court of Justice considered that ‘the final sentence of the second subparagraph of Article 24(1) TEU and the first paragraph of Article 275 TFEU introduce a derogation from the rule of the general jurisdiction which Article 19 TEU confers on the Court to ensure that in the interpretation and application of the Treaties the law is observed, and they must, therefore, be interpreted narrowly’ (emphasis added); see at para 70 of its Mauritius judgment. 96 For a possible role of domestic courts in settling questions of liability for CSDP operations, see Chapter 7 in this volume. 97 See Article 2 TEU, and e.g. Schrems (n 90), para 60.

86 Research handbook on the EU’s common foreign and security policy review the legality of the CFSP measure on the basis of EU law only, including the CFR, thus acting in the interest of the Union and in line with their duty of cooperation. In other words, the shared power of national courts in exercising judicial review of CFSP acts may contribute to securing the primacy of EU norms, including the Charter, in situations where the Court does not have jurisdiction.98

5. CONCLUSION While Article 24(1) TEU refers to ‘specific rules and procedures’ for the CFSP, it is equally clear that, in the absence of such specific rules and procedures, the general rules apply. If anything, the case law of the Court referred to above consistently underlined this. References to CFSP provisions in Articles 24(1) TEU and 275 TFEU are indeed meant to limit the Court’s jurisdiction to acts that are not of a ‘pure’ CFSP nature (based on Title V, Chapter 2 TEU).99 Indeed, as Cremona puts it, ‘this allows the Court – while granting the CFSP full scope as a policy field – to ensure that “CFSP exceptionalism” with respect to its own jurisdiction does not creep beyond its proper bounds’.100 While one may discuss what these ‘proper bounds’ are, it is at least clear that the Court’s general jurisdiction is not limited by the fact that a certain act was adopted in the context of the CFSP. Yet it is widely acknowledged that, despite the Court’s clear jurisdiction in relation to CFSP-related issues, it cannot yet provide ‘full review’. If so, there is a systemic gap in the EU system of judicial remedies, as seemingly recognized by the Court itself. How could one then fill this gap? One option is simply to grant the Court full judicial oversight over the CFSP and thus suppress the current derogatory provisions of Articles 24(1) TEU and 275 TFEU.101 The Court may, in the meantime, have to elaborate on its current CFSP-related case law, premised on the narrow interpretation of the derogations enshrined in the above-mentioned Articles, in an attempt to meet, as far as possible, the standards of Article 47 CFR without circumventing the limits set out by the Treaty drafters. The hope in Luxembourg may indeed be that, in the medium term, the Masters of the Treaties realize that the Court of Justice is trustworthy in the CFSP context, the integrity of which it scrupulously protects, in that it is capable of exercising judicial control over the EU foreign policy without overshadowing the authority of the political protagonists. 98

In this respect, see Case C-399/11 Melloni v Ministerio Fiscal ECLI:EU:C:2013:107. cf. the Opinion of AG Wathelet in the Rosneft case (n 10), paras 42–46. The question whether ‘pure’ CFSP sanctions (e.g. arms embargoes) would be covered by the ‘restrictive measures’ mentioned in Art. 275 remains unanswered, but given the Court’s restrictive approach to the exceptions and the importance attached to EU principles (such as equality and access to court), we would see no reasons to exclude sanctions with a mere CFSP legal basis from the Court’s jurisdiction. 100 Cremona (n 2). 101 See in this respect, the intervention of President of the CJEU Lenaerts at the ICON-S Conference 2016, Day 3, Plenary Session 3: ‘Judicial Interview and Dialogue’: . 99

‘The Good, the Bad and the Ugly’ 87 A droit constant, the Court of Justice may have to accept that in the current system, an additional role for external or other EU judicatures must be acknowledged. In that respect, Opinion 2/13 was not very helpful for those waiting for the possibility of the ECtHR stepping in and filling the gaps. As we have argued elsewhere,102 the criteria to allow external courts to deal with EU (including CFSP) law are hard to meet: they comprise, inter alia, the prevention of an adverse effect on the autonomy of the EU legal order; respect for the allocation of powers between the EU and its Member States; and the absence of jurisdiction to interpret EU law. At the same time, other international courts or national courts in third states may de facto be confronted with CFSP-related questions and it may be difficult for EU Member States to draw on their EU membership card to escape international obligations where situations have not been regulated otherwise. While a substantial role for ‘outside’ courts may indeed be difficult considering the terms of Article 344 TFEU, the contrary seems to hold true for the domestic courts in the EU Member States. On the basis of Article 19 TEU, they do have a role to play. To quote the Court once more: ‘As is evident from Article 19(1) TEU, the guardians of [the] legal order and the judicial system of the European Union are the Court of Justice and the courts and tribunals of the Member States.’103 The Rosneft case confirmed that the CFSP is not excluded from this general role of the national courts. Thus, full access to court is ensured as regards CFSP acts, in line with requirements of the rule of law. Judicial control over the CFSP may thus take place at different levels and we would maintain that, as the law stands, the role of the national courts is not so much ‘ugly’ as, in certain situations, necessary. In addition, to ensure full review of CFSP measures, it can only be hoped that the Court will not consider all external review as ‘bad’ and at least allow for other courts (the ECtHR in particular) to fill the gaps, thus giving preference to effective judicial review over autonomy in situations where its own jurisdiction is limited.

102 103

Hillion and Wessel (n 42). Opinion 1/09 (n 83), para 66 (emphasis added).

PART B THE PRACTICE OF CSDP

5. Civilian CSDP missions: ‘the good, the bad and the ugly’ Ana E. Juncos*

1. INTRODUCTION Since the launch of the Common Security and Defence Policy (CSDP) in 1999,1 the European Union (EU) has developed the capabilities to plan and conduct civilian crisis missions that cover a wide range of areas, from policing to border monitoring to judicial reform. Even though civilian crisis management developed in the shadow of EU military capabilities, progress in this area has been rapid and almost two-thirds of the EU missions and operations launched so far have been civilian in nature or have had a civilian component. Demonstrating the credibility of commitments laid down at the Helsinki Council (1999), and further developed at Feira (2000) and Gothenburg (2001), was of some importance. At the Feira Council, the EU Member States identified four priority areas in civilian crisis management (policing, rule of law, civilian administration and civil protection) and made a commitment to provide 5,000 police personnel by 2003.2 The launch of the EU Police Mission in Bosnia in 2003 made the EU’s commitment to develop autonomous civilian crisis management capabilities operational. While much of the discussion in the literature has focused on the merits and obstacles to developing a defence dimension at the EU level,3 civilian CSDP has received relatively little attention.4 This is despite the fact that some of the EU’s key achievements in the security arena have been accomplished through the civilian 1

Prior to the Lisbon Treaty, the CSDP was referred to as the European Security and Defence Policy (ESDP); for the sake of consistency, the term CSDP is used throughout the chapter. 2 European Council, Council conclusions (Santa Maria de Feira, 19–20 June 2000), available at http://www.consilium.europa.eu/uedocs/cms_data/docs/pressdata/en/ec/00200-r1.en 0.htm, accessed 23 July 2018. 3 A Menon, ‘Empowering Paradise? The ESDP at Ten’ (2009) International Affairs 227; J Howorth, ‘The European Union’s Security and Defence Policy: The Quest for Purpose’ in C Hill, MH Smith and S Vanhoonacker, International Relations and the European Union (OUP 2017). 4 Exceptions include R Bossong, ‘EU Civilian Crisis Management and Organisational Learning’ (2013) European Security 94; CS Chivvis, EU Civilian Crisis Management: The Record So Far (RAND 2010); I Ioannides, ’EU Civilian Capabilities and Cooperation with the Military Sector’ in E Greco, N Pirozzi and S Silvestri (eds), EU Crisis Management: Institutions and Capabilities in the Making (Istituto Affari Internazionali 2010; A Nowak (ed.), Civilian Crisis Management: The EU Way (Chaillot Paper No. 90, European Union Institute for Security Studies 2006).

89

90 Research handbook on the EU’s common foreign and security policy dimension of the CSDP. This chimes well with a general trend whereby international intervention has become more focused on long-term support and capacity-building rather than short-term external imposition. Yet the emphasis on civilian crisis management does not signify a return to the image of the EU as a ‘civilian power’. Civilian CSDP is seen as contributing to both to the EU’s soft and hard power. In the words of Federica Mogherini: The European Union has always prided itself on its soft power – and it will keep doing so, because we are the best in this field. However, the idea that Europe is an exclusively ‘civilian power’ does not do justice to an evolving reality. For instance, the European Union currently deploys seventeen military and civilian operations, with thousands of men and women serving under the European flag for peace and security – our own security, and our partners’. For Europe, soft and hard power go hand in hand.5

This chapter seeks to examine progress in this area by focusing on its achievements so far, and especially its contribution to the role of the EU as an international actor (the good); the operational and capability problems it has faced in implementing CSDP civilian missions (the bad); and the politics of civilian crisis management (the ugly). Because of the political nature of civilian CSDP, it has been contested since its origins. In particular, this chapter focuses on three types of political contestation, namely intergovernmental, bureaucratic and local politics, and how these have affected the implementation of civilian CSDP. It is necessary, however, to first discuss the origins of civilian CSDP and the record so far.

2. FROM AMBITION TO REALITY: THE DEVELOPMENT OF THE CIVILIAN DIMENSION OF THE CSDP Despite the traditional depiction of the EU as a civilian power, the development of a civilian dimension to what was then known as the European Security and Defence Policy (ESDP)6 only came about as a result of the pressure from the Nordic countries, which saw this as a counterweight to the militarization of the EU. Yet as this section goes on to show, EU civilian crisis management missions have since become a key instrument in the EU’s toolbox to support its role in conflict prevention and peacebuilding.

5 ‘Shared Vision, Common Action: A Stronger Europe. A Global Strategy for the European Union’s Foreign and Security Policy’ (European External Action Service 2016) 4. https:// europa.eu/globalstrategy/sites/globalstrategy/files/eugs_review_web.pdf, accessed 28 November 2016. 6 With the entry into force of the Lisbon Treaty, this policy area has been renamed as Common Security and Defence Policy (CSDP).

Civilian CSDP missions 91 The development of civilian crisis management7 coincided with changes in the security context in the post-Cold War period. During the 1990s and 2000s, the liberal peace, with its emphasis on stabilizing conflict countries and promoting free market democracies, generated an increasing demand for externally led interventions. Deployment of United Nations (UN) missions continued to grow over this period to over 95,000 uniformed deployed personnel in 2017 and a total of 71 peacekeeping operations since 1948.8 Moreover, international intervention has not just increased in size, but has also widened its scope from its traditional peacekeeping role to peacebuilding and with it the type of missions deployed: monitoring a ceasefire, electoral observation missions, disarmament, demobilization and reintegration (DDR) and security sector reform (SSR). While peacekeeping forces are still considered crucial in the first stages of a conflict, civilian crisis management has become increasingly relevant as an instrument to support transition and long-term stabilization and development. Regional organizations have also stepped up to this challenge, becoming more involved in peacekeeping and crisis management. In the case of the EU, the development of the Common Foreign and Security Policy (CFSP) with the Maastricht Treaty, and later the CSDP, can be placed in this context. Cooperation in security and defence matters has been one of the most recent additions to the process of European integration, with the CSDP only having been formally enshrined in the Treaty of Nice (2000). However, at that point the emphasis was on the development of ‘credible military forces’ in order to develop the EU’s ‘capacity for autonomous action’.9 Yet at the Helsinki European Council (December 1999), the Council agreed to establish a non-military crisis management mechanism at the disposal of the Union as a compromise between those seeking to enhance the EU’s military dimension and the neutrals, led by Sweden and Finland, which advocated for a 7 A terminological note is in order here. While the literature on international conflict usually refers to conflict management, in EU circles the preferred term is that of ‘crisis management’. Crisis/conflict management usually refers to those short-term measures facilitating ‘the settlement and containment of violent conflict’, which might include inter alia peace enforcement, peacekeeping and post-conflict stabilization (O Ramsbotham, T Woodhouse and H Miall, Contemporary Conflict Resolution (2nd edn, Polity Press 2005)). By contrast, conflict prevention is the term preferred by the European Commission. In the Commission’s parlance, conflict prevention refers to both long-term and short-term measures; it covers the stages before and after the break-out of a conflict (Commission, ‘Communication from the Commission on Conflict Prevention’ COM(2001) 211 final). In this view, only a phase of the conflict cycle escapes from the Commission’s definition of conflict prevention: the management of an open crisis (war). From the Council’s point of view, crisis management includes not only peacemaking and peacekeeping, but also other state-building, confidence-building and monitoring activities taking place in the post-conflict stabilization phase, such as police missions, monitoring missions and border assistance missions. Thus, the possibility of overlaps between Commission and Council’s instruments is a real one in the post-conflict stabilization stage. See also E Gross and AE Juncos (eds), EU Conflict Prevention and Crisis Management (Routledge 2011). 8 United Nations, Peacekeeping Factsheet 2017, http://www.un.org/en/peacekeeping/ resources/statistics/factsheet.shtml, accessed 10 July 2017. 9 EU-ISS (2001) From St-Malo to Nice. European defence: core documents, Chaillot Paper 47, Paris: EU-ISS. Available at: https://www.iss.europa.eu/sites/default/files/EUISSFiles/cp047e. pdf accessed 10 July 2017.

92 Research handbook on the EU’s common foreign and security policy civilian dimension to crisis management. At the Feira Council (2000), the EU Member States identified four priority areas in civilian crisis management (policing, rule of law, civilian administration and civil protection) and made a commitment to provide 5,000 police personnel by 2003. Significant improvements have taken place in the last decade with the adoption of the Civilian Headline Goal 2008 and the Headline Goal 2010, a list of generic civilian CSDP tasks, the creation of a permanent CSDP warehouse, and the adoption of a civilian CSDP Compact, among other things (civilian capability development is explored in Chapter 8 in this volume).10 The civilian missions carried out in the context of CSDP are governed by Articles 42–46 of the Treaty on European Union (TEU) under the heading ‘Provisions on the Common Security and Defence Policy’ (Title V, Chapter 2, Section 2). CSDP constitutes an integral part of the CFSP (Art. 42(1) TEU) and it continues to be governed within an intergovernmental framework. As such, decisions in the Council are taken by unanimity (Art. 42(4) TEU). The European Parliament is consulted on and informed of CSDP developments, including operational developments, with the specification that its views are to be ‘duly taken into consideration’ (Art. 36 TEU). Civilian mission mandates are set by the Council of the European Union and agreed with host states. Although not legally required, this usually includes the signing of a Status of Mission Agreement or SOMA. While the Council is responsible for setting out ‘the objectives and scope and the general conditions for [the] implementation’ of civilian CSDP missions, ‘the High Representative of the Union for Foreign Affairs and Security Policy, acting under the authority of the Council and in close and constant contact with the Political and Security Committee, shall ensure coordination of the civilian and military aspects of such tasks’ (Art. 43(2) TEU). The rules for the financing of the missions are outlined in Article 41(1) TEU. According to this provision, administrative expenditure is automatically charged to the EU budget. Operational expenditure for civilian missions is also charged to the EU budget (Art. 41(2) TEU) as long as it does not arise from operations having military or defence implications. To ensure the rapid disbursement of funds for preparatory actions for civilian CSDP missions, the Lisbon Treaty also introduced a mechanism in Article 41(3) TEU according to which the Council would adopt a decision on ‘specific procedures for guaranteeing rapid access to appropriations in the Union budget’, after consulting the European Parliament. According to the Treaty, the operational implementation of civilian CSDP should rely on the capabilities provided by the Member States (Art. 42(1) and 42(3) TEU). The EU may use civilian and military assets outside the EU for peacekeeping, conflict prevention and strengthening international security (Art. 42(1) TEU). The Lisbon Treaty also extended the scope and type of civilian and military missions to be carried out under the CSDP framework (the so-called ‘Petersberg tasks’) (Art. 43 TEU) to 10 Council of the EU, ‘Civilian Headline Goal 2008’ 15863/04, 5–7, http://eeas.europa.eu/ csdp/about-csdp/civilian_headline_goals/index_en.htm. GAERC (2007) ‘Civilian Headline Goal 2010’ (approved by the ministerial Civilian Capabilities Improvement Conference and noted by the General Affairs and External Relations Council) 14823/07, 2; EEAS (2015), ‘Draft List of Generic Civilian CSDP Tasks’, EEAS (2015) 654 Rev 1, http://www.statewatch.org/news/2015/ may/eu-csdp-revised-cvilian-tasks-7656-rev2-15.pdf, accessed 18 April 2016.

Civilian CSDP missions 93 include humanitarian and rescue tasks; conflict prevention; tasks for combat forces in crisis management; joint disarmament operations; military advice and assistance tasks; and tasks in post-conflict stabilization. The Treaty adds that ‘these tasks may contribute to the fight against terrorism, including by supporting third countries in combating terrorism in their territories’ (Art. 43(1) TEU). At the Laeken Council, in December 2001, the CSDP was declared operational. Since then, civilian CSDP has become a reality, with the EU being involved in over 20 civilian crisis management missions out of a total of 36 – including police missions, civilian administration, training missions and rule of law missions.11 At the time of writing, the EU is conducting ten civilian missions and six military missions. The last civilian CSDP mission launched by the EU was the EUAM Iraq, an EU Advisory Mission in support of Security Sector Reform, established in 2017.12 One of the EU’s key strengths has been its ability to combine both civilian and military instruments and to develop a ‘comprehensive approach’ to conflict management (see below). Over time, the structures dealing with civilian crisis management have seen the most significant changes in the EU foreign policy machinery, some of them prior to its operationalization (e.g. CIVCOM), some of them as a result of lessons learned.13 CIVCOM, the Committee on Civilian Aspects of Crisis Management, was established in June 2000. From the original Police Unit created within the Directorate General (DG) IX in the Council Secretariat, which had eight people working for it, a Civilian Planning and Conduct Capability was established in 2007, headed by the Civilian Operations Commander, and currently manned by 75 staff. In 2009, the Crisis Management and Planning Directorate (CMPD) was also created, in charge of political-strategic crisis management planning and joint civilian-military capability development. The Lisbon Treaty did not alter this institutional structure, although it led to a reorganization of the crisis management bodies within the European External 11 This total includes the EUBAM Moldova and Ukraine and the EUMM to former Yugoslavia, which are not strictly CSDP missions, but share most of their features. On EU crisis management operations prior to the launch of the CSDP, see W van Eekelen and S Blockmans, ‘European Crisis Management Avant la Lettre’ in S Blockmans (ed.), The European Union and Crisis Management (TMC Asser Press 2008); and AE Juncos, EU Foreign and Security Policy in Bosnia. The Politics of Coherence and Effectiveness (Manchester University Press 2015). The same can be said of EUJUST Themis in Georgia, which had more the character of a Commission project (see F Hoffmeister, ‘Inter-pillar Coherence in the European Union’s Crisis Management’ in S Blockmans (ed.), The European Union and Crisis Management (above) 1651–66). 12 See EEAS, ‘Military and Civilian Missions and Operations’ (2017) https://eeas.europa. eu/topics/military-and-civilian-missions-and-operations/430/military-and-civilian-missions-andoperations_en, accessed 10 July 2017. 13 A Kammel, ‘The EEAS and Its Crisis Management Component’ in J Rehrl (ed.), Handbook on CSDP: The Common Security and Defence Policy of the European Union (Federal Ministry of Defence and Sports of the Republic of Austria 2017); R Bossong and T Benner, ‘Capacity-building at the Headquarter Level: The Case of EU Civilian Peace Operations’ (2012) Journal of Intervention and Statebuilding 351. On learning, see G Faleg, The EU’s Common Security and Defence Policy: Learning Communities in International Organizations (Palgrave Macmillan 2017); Juncos (n 11); ME Smith, Europe’s Common Security and Defence Policy: Capacity-Building, Experiential Learning and Institutional Change (CUP 2017).

94 Research handbook on the EU’s common foreign and security policy Action Service (EEAS), and under the authority of the new position of High Representative for Foreign Affairs and Security Policy. The establishment of this double-hatted position (as the Vice-President of the Commission) also meant that stronger links could be sought with the work of the Commission in security and development areas – although the legal dichotomy remains. Related to this, the reactivation of the Commissioners’ Group on External Action, led by the HR/VP, has facilitated inter-institutional cooperation within the Commission and with the EEAS.14 Evidence of better inter-institutional relations can be found in the fact that the Commission and the EEAS have worked closely on different initiatives including on SSR, Capacity Building for Security and Development, and resilience, to name but a few. Yet, there are still difficulties regarding complex budgetary procedures and the rapid disbursement of funds, for instance. However, the turf wars that accompanied the implementation of the Lisbon Treaty and the consequences of the economic and financial crisis led to a slowdown in the pace and the level of ambition of CSDP. More recent attempts to reinvigorate the CSDP at the December 2013 European Council were mostly focused on the military dimension of CSDP (this is explored in Chapter 6 in this volume). Nevertheless, the EU Global Strategy (EUGS) refers to civilian missions as ‘a trademark of CSDP’ and calls for the further development of civilian CSDP by ‘encouraging force generation, speeding up deployment, and providing adequate training based on EU-wide curricula’.15 This suggests that civilian crisis management will continue to be a key priority in the EU’s external action in the years to come as demonstrated by the launch of a civilian CSDP Compact in 2018.

3. CIVILIAN CSDP MISSIONS: THE RECORD SO FAR As mentioned above, civilian CSDP missions have become a trademark of CSDP, with over 20 deployed to date. This section provides an overview of the record of civilian CSDP missions so far by examining the geographical areas of deployment, types of missions (such as police, rule of law, and monitoring), and personnel deployed. The first point worth noting is that the average size of the current missions is relatively small, with around 247 personnel (including contracted and local staff), although this figure would be even smaller if we were to exclude EULEX (which amounts to more than half of the total) (see Table 5.1). Their budgets are also relatively modest: €250 million was spent in 2015 out of the total CSDP budget. The Member States also contribute directly through the salaries of their seconded personnel. The top contributing Member States are Germany, Poland, Sweden, Italy and Finland.16 14 See, for instance, S Blockmans and S Russack, The Commissioners’ Group on External Action – Key Political Facilitator (Special Report No. 125, CEPS 2015). 15 ‘Shared Vision, Common Action’ (n 5) 47. 16 EEAS, Personnel Figures of the Civilian CSDP Missions (31.08.2016). For a discussion of the UK’s contribution to civilian CSDP and the likely impact of Brexit, see the analysis in Chapter 8 in this volume.

Civilian CSDP missions 95 Table 5.1 Deployed personnel in civilian CSDP missions EULEX EUAM Kosovo Ukraine

EUMM Georgia

EUBAM EUPOL Rafah COPPS

EUCAP EUCAP EUBAM EUCAP Nestor Niger Libya Sahel Mali

Seconded personnel

387

83

178

2

42

23

38

10

57

820

Contracted international

197

26

27

1

15

28

22

2

27

345

Local staff

749

79

109

7

42

3

34

2

34

1,059

Total

1,333

188

314

10

99

54

94

14

118

2,224

Note:

This table does not include EUPOL Afghanistan, which was completed on 31 December 2016.

Source: EEAS, Personnel Figures of the Civilian CSDP Missions (31.08.2016).

In terms of the mandate, the first trend that we have seen in civilian CSDP missions has been towards a more holistic approach towards SSR. While most of the first CSDP missions deployed by the EU focused only on one security sector (e.g. the police or border monitoring), there was soon a realization that a more integrated approach was required. For instance, the first mandate of EU Police Mission (EUPM) Bosnia was too narrow in the sense that it only focused on policing issues. By contrast, later mandates responded to criticisms that EUPM had neglected the rule of law dimension in the fight against criminality. Thus, a Criminal Justice Interface Unit was created to address problems regarding cooperation between the police and the prosecutorial authorities. Later missions have had a much broader remit such as rule of law, SSR or capacity-building missions (e.g. EULEX Kosovo, EUAM Ukraine and EUCAP Sahel, to name a few) (see also Table 5.2 on the nature and type of missions). The second trend has also had to do with changes in the security environment and the nature of Western interventions. The demise of the liberal peace17 has now led to calls for local ownership and building the resilience of external partners.18 Resilience seeks to move beyond previous institution-building programmes that sought holistic and externally driven changes to building the capacity of local actors from the bottom up.19 Through the lenses of resilience, interventions now should be focused at the level of actors’ capacities rather than on the external or international environment.20 The EU’s role is to help states and societies build their resilience through training, 17 R Mac Ginty and OP Richmond, ‘The Local Turn in Peace Building: A Critical Agenda for Peace’ (2013) Third World Quarterly 763. 18 Commission and High Representative of the Union for Foreign Affairs and Security Policy, ‘Joint Communication to the European Parliament and the Council. A Strategic Approach to Resilience in the EU’s External Action’, JOIN (2017) 21 final. 19 AE Juncos, ‘Resilience as the New EU Foreign Policy Paradigm: a Pragmatist Turn?’ (2017) European Security 1; D Chandler, ‘International Statebuilding and the Ideology of Resilience’ (2013) Politics 276. SV Haldrup and F Rosén, ‘Developing Resilience: A Retreat from Grand Planning’ (2013) Resilience: International Policies, Practices and Discourses 130. 20 ‘Shared Vision, Common Action’ (n 5); Juncos (n 19).

96 Research handbook on the EU’s common foreign and security policy Table 5.2 EU civilian and military missions and operations, 1991–2017 Name

Location

Nature

Type

Duration

ECMM/EUMM

Western Balkans

Civilian

Monitoring

1991–2007

EUPM

Bosnia

Civilian

Police

2003–2012

Concordia

FYROM

Military

Military

2003

Artemis

RD Congo

Military

Military

2003

EUPOL Proxima

FYROM

Civilian

Police

2004–2005

EUJUS Themis

Georgia

Civilian

Rule of law

2004–2005

EUFOR Althea

Bosnia

Military

Military

Since 2004

EUPOL Kinshasa

RD Congo

Civilian

Police

2005–2007

EUSEC RD

RD Congo

Civil-military

Security sector reform

2005–2016

EUJUST LEX

Iraq/Brussels

Civilian

Rule of law

2005–2013

Support to AMIS

Sudan

Civil-military

Assistance

2005–2006

AMM

Aceh/Indonesia

Civilian

Monitoring

2005–2006 Since 2005

EUBAM Rafah

Palestinian Territories

Civilian

Border

EUBAM

Ukraine-Moldova

Civilian

Border

Since 2005

EUPOL COPPS

Palestinian Territories

Civilian

Police

Since 2006

EUPAT

FYROM

Civilian

Police

2006

EUPT

Kosovo

Civilian

Planning

2006–2008

EUFOR

RD Congo

Military

Military

2006

EUPOL

RD Congo

Civilian

Police

2007–2014

EUPOL

Afghanistan

Civilian

Police

2007–2017

EUFOR

Tchad/RCA

Military

Military

2008–2009

EU SSR

Guinea-Bissau

Civil-military

Security sector reform

2008–2010

EULEX

Kosovo

Civilian

Rule of law

Since 2008

EUMM

Georgia

Civilian

Monitoring

Since 2008

EUNAVCO

Somalia

Military

Assistance

2008

EUNAVFOR

Somalia

Military

Maritime

Since 2008

EUTM

Somalia

Military

Capacity-building

Since 2010

EUAVSEC

South Sudan

Civilian

Security sector reform

2012–2014

EUCAP NESTOR

Horn of Africa

Civilian

Capacity-building

Since 2012

EUCAP Sahel Niger

Niger

Civilian

Capacity-building

Since 2012

EUBAM Libya

Libya

Civilian

Border

Since 2013 Since 2013

EUTM Mali

Mali

Military

Capacity-building

EUCAP Sahel Mali

Mali

Civilian

Capacity-building

Since 2014

EUAM

Ukraine

Civilian

Security sector reform

Since 2016

EUFOR RCA

CAR

Military

Military

2014–2015

EUMAM RCA

CAR

Military

Security sector reform

2015–2016

EUNAVFOR MED

Mediterranean

Military

Maritime

Since 2015

EUTM RCA

CAR

Military

Capacity-building

Since 2016

Source: EEAS, ‘Military and Civilian Missions and Operations’ (2017) https://eeas.europa.eu/topics/militaryand-civilian-missions-and-operations/430/military-and-civilian-missions-and-operations_en.

Civilian CSDP missions 97 monitoring, mentoring and advising.21 Thus, the role of the EU appears as one of a facilitator, a mentor and a partner moving away from discourses of ‘external intervention’ and ‘transformation’. Civilian CSDP operations therefore play a key role in terms of supporting capacity-building, in particular in the security sector (e.g. EUCAP Mali, EUCAP Niger, EUCAP Nestor/Somalia). Third, other changes in the security context have also led to changes in the roles of civilian CSDP missions. For instance, a focus on hybrid threats, maritime security and aviation security explains the deployment of missions such as EULEX, EUAM Ukraine, EUCAP Nestor or EUAVSEC. There also seems to be more emphasis on the protection of EU interests through its CSDP operations. The Implementation Plan on Security and Defence is very clear about this. CSDP missions and operations must contribute not only to responding to external conflicts and crisis and building the capacities of partners, but also to protecting the Union and its citizens (see also Art. 21 TEU). According to this document, ‘[t]hese three priorities are mutually reinforcing. A single CSDP mission or operation in fact can potentially contribute to all three priorities: they constitute a coherent whole.’22 Thus, it is not surprising that CSDP civilian missions have not only focused on strengthening the rule of law in third countries and building their capacities, but also on fighting organized crime and border management (EULEX Kosovo and EUBAM Rafah), illegal immigration (EUCAP Sahel Niger and EUCAP Sahel Mali), counter-terrorism (EUPOL Afghanistan and EUCAP Sahel Niger) and anti-piracy (EUCAP Nestor/Somalia). These changes in the functions and scope of civilian CSDP have put into question the four priority areas agreed at Feira. These key priority areas included police, rule of law, civilian administration and civil protection. At the November 2004 Civilian Capabilities Commitment Conference, two more areas for civilian crisis management were identified: monitoring and support to EU Special Representatives.23 Later work in the context of the Civilian Capability Development Plan identified some ‘generic civilian CSDP tasks’ to guide the development of requirement lists of capabilities, although it did not amend the overall priorities identified at Feira.24 The Implementation Plan of November 2016, however, calls on the ‘EEAS to make proposals for Member States’ consideration on revisiting the Feira priority areas for civilian missions in light of the profoundly changed security environment’.25 The preference for monitoring, mentoring and advising (MMA) mandates is also noteworthy. While in the first years of the CSDP there was intense discussion among 21

On ‘train and equip’, see Chapter 6 in this volume. Council of the EU, ‘Implementation Plan on Security and Defence’ (14 November 2016) 3, https://eeas.europa.eu/sites/eeas/files/eugs_implementation_plan_st14392.en16_0.pdf, accessed 28 November 2016. 23 ‘Ministerial Declaration’ (Civilian Capabilities Commitment Conference, Brussels, 22 November 2004), https://www.consilium.europa.eu/uedocs/cmsUpload/COMMITMENT%20 CONFERENCE%20MINISTERIAL%20DECLARATION%2022.11.04.pdf, accessed 27 February 2017. 24 EEAS Revised draft list of generic civilian CSDP tasks. 7656/2/15, Brussels, 8 March 2015. Available at http://www.statewatch.org/news/2015/may/eu-csdp-revised-cvilian-tasks-76 56-rev2-15.pdf, accessed 27 February 2017. 25 Council of the EU, ‘Implementation Plan on Security and Defence’ (n 22). 22

98 Research handbook on the EU’s common foreign and security policy the Member States about the merits of strengthening vs executive missions,26 the default position nowadays seems to be for non-executive MMA mandates,27 especially given the problems experienced with the only executive civilian mission so far: EULEX Kosovo. Under its executive mandate, EULEX judges and prosecutors carry out investigations and adjudicate criminal cases ‘until the progress of local authorities allows a complete transition of executive functions to them’.28 Finally, over time the EU has gradually expanded the geographical scope of its operations, originally centred in Europe (Western Balkans and South Caucasus) and Africa, to the Middle East and South East Asia (see Table 5.2). However, the majority of its missions are still deployed in Europe and Africa. This brief overview of the main features of civilian CSDP already illustrates the flexibility of these missions, which constitutes one of its strengths in terms of the EU’s global role. The following section examines in more detail the positive contribution that civilian CSDP makes to international security.

4. THE GOOD: CIVILIAN CSDP AND ITS CONTRIBUTION TO THE EU’S ROLE AS AN INTERNATIONAL ACTOR EU civilian crisis management has overall contributed to the EU’s role as a normative power. First, the objectives of these missions largely support the normative dimension of the EU and are in line with the CFSP objectives as established in the Treaties – democracy, the rule of law, the universality and indivisibility of human rights and fundamental freedoms, respect for human dignity, equality and solidarity, and respect for the principles of the UN Charter of 1945 and international law (Art. 21 TEU). The mandates of these missions have focused on responding to crises and promoting peace; supporting the stabilization of post-conflict societies and strengthening the rule of law and the security sector according to democratic principles. This is not to say that civilian CSDP missions have not sought to promote EU ‘interests’. Indeed, many of the missions have aimed to protect the security of the Union by addressing related security threats such as organized crime (mainly in the Western Balkans), migration (in the Sahel and Horn of Africa), piracy (Horn of Africa) and counter-terrorism (in Afghanistan and Africa). While this is in line with the EUGS, which argues that ‘interest and values go hand in hand’,29 this has inevitably led to tensions in the implementation of civilian CSDP. In particular, where missions are seen to pursue EU geopolitical and economic interests, this might undermine the legitimacy and buy-in from the local authorities and populations. In the Balkans, for instance, a focus on organized crime 26

On the case of EUPM, see Juncos (n 11). Indicative of this preference is the establishment of the Military Planning and Conduct Capability’ (MPCC) to oversee all military non-executive missions as outlined in the Council conclusions of 14 November 2016. See T Tardy, ‘MPCC: towards an EU military command?’ (Brief Issue, EU-ISS, 7 June 2017). 28 EULEX, Basic Facts Factsheet, http://www.eulex-kosovo.eu/eul/repository/docs/2016 1209_EULEX-FactSheet.pdf, accessed 2 February 2017. 29 ‘Shared Vison, Common Action’ (n 5) 13. 27

Civilian CSDP missions 99 and the strengthening of border controls by the civilian missions in the region (EUPM in Bosnian and Herzegovina, Proxima in FYROM, EULEX Kosovo) was seen to be at odds with the key priorities of the local populations, which were more concerned with petty crime and unemployment.30 The preference for strengthening mandates and the long-term approach of many of these missions bode well for the EU’s role as a normative power. Regarding the first point, as discussed earlier, with the exception of the rule of law mission in Kosovo (EULEX), all civilian CSDP missions have had a strengthening mandate. Rather than supplanting local actors, CSDP operations have sought to nurture local capacities through mentoring, monitoring and advising. For instance, in Ukraine, EUAM has supported the implementation of a national community policing programme; EUCAP Sahel Niger has helped with the revision of the police training manual; and EUPOL COPPS in Palestine has strengthened the capacities of the Palestinian Civil Police.31 Secondly, while civilian missions are officially part of the EU’s ‘crisis management’ responses, in practice they have focused on longer-term peacebuilding and preventative efforts. This in turn raises the question of whether the label ‘crisis management’ is an appropriate one for civilian missions and the issue of coordination with other long-term Commission instruments.32 In fact, in the case of Bosnia, there was not always full cooperation in harmonizing the projects launched by the Commission and EUPM, leading, for example, to the duplication at times of advisors to some Bosnian institutions, with one co-located police officer coming from EUPM and one advisor from the Commission’s twinning projects. Another related problem is that, even though many of the issues civilian CSDP missions face require a long-term approach, not all the missions have been deployed long enough to have a long-lasting impact. This was the case with EUJUST THEMIS, a CSDP operation which, despite having very ambitious objectives, was given only a year to achieve them. It therefore came as no surprise that by the end of the mission many of its objectives remained unaccomplished.33 But even in the cases of EUPM Bosnia and EUPOL Afghanistan, which were deployed for ten years, their impact is questionable. In the latter case, according to Thierry Tardy, the nature and scale of the challenges that EUPOL faced during its ten-year presence were huge, and the Mission was not designed and resourced to effectively tackle those challenges.

30 AE Juncos, ‘Of Cops and Robbers: European Union Policy and the Problem of Organized Crime in Bosnia Herzegovina’ in B Balamir-Coskun and B Demirtas-Coskun (eds), Neighborhood Challenge: The European Union and its Neighbors (Universal Publishers 2009). 31 K Fearon and S Picavet, ‘Challenges for Civilian CSDP Missions’ in J Rehrl (ed.), Handbook on CSDP: The Common Security and Defence Policy of the European Union (Federal Ministry of Defence and Sports 2017). 32 See also n 8. 33 X Kurowska, ‘The Rule-of-Law Mission in Georgia (EUJUST Themis)’ in G Grevi, D Helly and D Keohane (eds), European Security and Defence Policy: The First Ten Years (1999–2009) (European Union Institute for Security Studies 2009).

100 Research handbook on the EU’s common foreign and security policy In the end EUPOL’s achievements were limited and the medium-term sustainability of what has been achieved is far from being guaranteed.34

Finally, civilian CSDP missions have also shown a high degree of adaptability. The evolution of their mandates from a narrow focus on police reform and SSR to more multifaceted activities (aviation security, maritime capacity-building, training) provide evidence of this adaptability. In principle, it is also possible to deploy them in a relatively short space of time;35 in practice, however, the planning process has often taken several months (six months in the case of EUAM Ukraine and ten months in the case of EUCAP Sahel Mali).36 It is also possible to change and adapt their mandates throughout their deployment (see, for instance, several changes to the mandate and name of EUCAP Somalia).37 Another positive development has been the increasing interaction with other EU instruments and actors, particularly with the Commission (EUBAM Moldova and Ukraine is a case in point), FRONTEX (e.g. EUBAM Libya) or Europol (in the case of EULEX Kosovo), to name but a few. Civilian CSDP missions also work closely with other international actors deployed on the ground, such as the UN (in Mali, for instance), the OSCE (in Ukraine) or NATO (in Afghanistan or Kosovo38). As far as EU–NATO relations are concerned, the possibility of NATO resorting to EU civilian capabilities has been floated in the past (the so-called ‘Berlin Plus in reverse’), but has never materialized.39 Instead, due to the bilateral dispute between Turkey and Cyprus, EU–NATO relations have been characterized by a political impasse since 2004–2016, with formal cooperation being limited to EUFOR Althea in Bosnia Herzegovina (the only CSDP operation launched under the ‘Berlin Plus’ arrangements to date). In other cases (Afghanistan, Kosovo, the Horn of Africa or the Mediterranean Sea), cooperation between EU and NATO operations has relied on ad hoc and informal agreements on the ground rather than formalized procedures.40 The political declaration signed at the NATO Warsaw Summit in 2016 appears to have breathed new life into cooperation between the two organizations. The NATO 34 T Tardy, ‘EUPOL Afghanistan 2007/16: Mission Impossible?’ (Brief Issue, EU-ISS, 5 July 2017). 35 For instance, the planning of AMM Aceh took a matter of weeks before its operational deployment on 15 September 2005. H Dijkstra, Policy-Making in EU Security and Defense. An Institutional Perspective (Palgrave Macmillan 2013) ch.6. 36 AI Xavier and J Rehrl, ‘How to Launch a CSDP Mission or Operation’ in J Rehrl (ed.), Handbook on CSDP: The Common Security and Defence Policy of the European Union (Federal Ministry of Defence and Sports of the Republic of Austria 2017). 37 EUCAP Somalia (2017) Legal basis, http://www.eucap-som.eu/legal-basis/, accessed 11 July 2017. 38 In the case of Kosovo and Afghanistan, and due to the lack of communication between the two organizations at the strategic level, draft technical agreements were informally negotiated at the operational level between CSDP missions and NATO operations. 39 L Michel, ‘NATO and the United States: Working with the EU to Strengthen EuroAtlantic Security’ in S Biscop and R Whitman (eds), The Routledge Handbook of European Security (Routledge 2012). 40 N Graeger, ‘Security. EU-NATO Relations: Informal Cooperation as a Common Lifestyle’ in A Orsini (ed.), The European Union With(in) International Organisations (Routledge 2014).

Civilian CSDP missions 101 Warsaw Declaration foresees increasing cooperation in areas such as hybrid threats and capacity-building, which could open new avenues for cooperation between civilian CSDP and NATO.41 On 6 December 2016, the Council conclusions endorsed 42 concrete proposals to implement the EU-NATO Joint Declaration, although the implementation of such proposals will still depend on the political will of the Member States, including making the required financial resources available, and other political developments in the neighbourhood and beyond, including Turkey’s willingness to cooperate with its European partners, Russia’s foreign policy in the East, and Donald Trump’s commitment to transatlantic relations.42 There have also been limits as to how flexible civilian CSDP missions can be. For one thing, integration with the instruments and actors from the area of Freedom, Security and Justice remains limited despite common challenges.43 Moreover, institutional constraints have also slowed down deployment and resulted in procurement problems due to the fact that while missions are planned by the EEAS, the funding is still administered by the Commission and subject to Commission procurement rules.44 The intergovernmental nature of CSDP missions also affects its ability to adapt to changing conditions on the ground, which in turn is seen as key in shaping local perceptions of effectiveness, as shown by the case of EUAM Ukraine.45 While the need for increasing synergies between civilian and military instruments has been repeatedly mentioned as a necessary requirement,46 coordination between civilian and military CSDP missions remains limited. For instance, civ-mil operations such as EUSEC RD and EU SSR Guinea-Bissau remain the exception rather than the norm (see Table 5.2). While there are regular institutional contacts at the strategic and operational level to facilitate coordination between missions, a more integrated approach similar to that of UN peacekeeping missions still remains off the table.47

41

‘Joint Declaration by the President of the European Council, the President of the European Commission and the Secretary-General of the North Atlantic Treaty Organization’ (NATO Declaration, Warsaw, 8 July 2016). 42 E Lazarou, ‘EU-NATO cooperation and European defence after the Warsaw Summit’ (Briefing, European Parliament Research Service 2016). 43 T Tardy (ed.), Recasting EU Civilian Crisis Management (EU-ISS 2017). 44 See Implementation Plan on Security and Defence on the need to improve the flexibility of financial mechanisms. Council of the EU, ‘Implementation Plan on Security and Defence’ (n 22). 45 K Zarembo, ‘Perceptions of CSDP Effectiveness in Ukraine: A Host State Perspective’ (2017) European Security 190. 46 See, for instance, ‘Promoting synergies between the EU civil and military capability development’ (EEAS 2011) https://www.consilium.europa.eu/uedocs/cmsUpload/110223%20 factsheet%20Civil-Military%20synergies%20-%20version%201_EN.pdf, accessed 27 February 2017. 47 The new Joint Communication on an Integrated Approach will seek to improve these issues but it does not foresee a vertical integration of different EU actors on the ground as is the case with the UN model. See also H Dijkstra, P Petrov and E Mahr, ‘Reacting to Conflict: Civilian Capabilities in the EU, UN and OSCE’ (2016) EU-CIVCAP Report DL4.1, https:// eucivcap.files.wordpress.com/2017/06/eu-civcap_deliverable_4-1.pdf.

102 Research handbook on the EU’s common foreign and security policy

5. THE BAD: BRIDGING THE CAPABILITIES–EXPECTATION GAP Despite its contribution to international security, civilian CSDP – and EU foreign policy, more generally – has suffered from what Christopher Hill described as the ‘capabilities–expectation gap’, in other words, ‘the significant difference which had come about between the myriad hopes for and demands of the EU as an international actor, and its relatively limited ability to deliver’.48 Although successive rounds of Treaty and institutional reforms have improved matters, there remains a gap between what the EU is set to achieve and the capabilities it has at its disposal, in particular in the area of civilian crisis management.49 Civilian CSDP missions have been affected by training and recruitment problems.50 As far as training is concerned, a number of initiatives, such as the European Security and Defence College and Europe’s New Training Initiative for CCM (ENTRi), a programme funded by the European Commission since 2011, have improved the training provision for civilian CSDP personnel. However, training activities lack standardization and coordination, thus impacting on the quality and quantity of personnel available for deployment.51 These problems are exacerbated by problems of recruitment. The lack of an EU roster of experts for civilian crisis management means that it is difficult to fill all the available vacancies, especially when it comes to highly specialized tasks. The fact that resources are dispersed among different national ministries (Interior, Justice, Civil Protection) makes things even more difficult. Finally, issues pertaining to lack of language skills, lack of career incentives, and availability of experts from the private sector and NGOs also hinder recruitment for civilian CSDP.52 The challenges faced in this area at the national level are somehow understandable. They have to do with the fact that by contrast with their military counterparts, civilian crisis management institutions had to be established from scratch at the national level.53 Moreover, there remains a lack of will by some Member States (in particular, some large Member States) to contribute to CSDP missions in a way commensurate with 48 C Hill, ‘Closing the Capabilities-expectations Gap?’ in J Peterson and H Sjursen (eds), A Common Foreign Policy for Europe? Competing Visions of the CFSP (Routledge 1998). 49 G Grevi and D Keohane, ‘ESDP Resources’ in Grevi, Helly and Keohane (eds) (n 33); D Korski and R Gowan, Can the EU Rebuild Failing States? A Review of Europe’s Civilian Capacities (European Council on Foreign Relations 2009). N Pirozzi, ‘The European Union and Civilian Crisis Management after Lisbon’ (2015) European Foreign Affairs Review 287. 50 See, for instance, Civilian and Military Personnel in CSDP Missions and Operations (European Parliament 2017) 15–20, http://www.europarl.europa.eu/RegData/etudes/STUD/2017/ 578035/EXPO_STU(2017)578035_EN.pdf, accessed 11 July 2017. 51 See T De Zan, P Tessari and B Venturi, ‘Procedures, Personnel and Technologies for Conflict Prevention and Peacebuilding: An Assessment of EU Member States’ Capabilities’ (2016) EU-CIVCAP Report DL2.1, http://www.eu-civcap.net/files/2016/11/Procedures_ Personnel_Technologies_Conflict_Prevention_Peacebuilding-Assessment_EU_Member_States_ Capabilities.pdf. 52 ibid; Dijkstra, Petrov and Mahr, ‘Reacting to Conflict’ (n 47); T Tardy, ‘Civilian CSDP: what next?’ (Brief Issue, EU-ISS, 10 November 2016). 53 See, for instance, S Vanhoonacker and AD Jacobs, ‘ESDP and Institutional Change: The Case of Belgium’ (2010) 41 Security Dialogue 559.

Civilian CSDP missions 103 their resources. For instance, while the Nordic countries have contributed significant numbers of civilian personnel, some larger Member States like France or Spain have a very low ratio of deployed personnel per head of population.54 At the EU level, the emphasis placed on the military dimension in the initial stages of the CSDP meant that the civilian dimension did not receive as much attention, not to mention resources. Yet the EU has put in place a number of mechanisms to support capability development on the civilian side with varied success, including the recently agreed civilian CSDP Compact (see Chapter 8 of this volume on capability development). Another related problem refers to the limited scale and lack of ambition of most of these missions (perhaps with the exception of EULEX).55 A quick look at the number of personnel and the time frame of these operations demonstrates that the majority of these missions have been short lived and small scale. Given the complexity of some of the tasks and the size of some of the countries where the EU has deployed these missions, this obviously raises the issue of whether the EU is really serious about these undertakings, with many of these missions falling short of meeting their objectives. Having said that, one should not underestimate the role played by symbolism. Civilian CSDP missions have not only been a way to address specific key threats, but have played a symbolic role as test cases and to demonstrate to the international community that CSDP is not simply a talking shop. In the case of EUJUST THEMIS, this was seen as ‘a good opportunity to test civilian crisis management capabilities in the field of rule of law, in a relatively stable area, with a small-scale mission’ as well as to serve as an ‘important test for EU relations with Russia’.56 However, as recently shown by Ukraine, civilian personnel deployed in missions are also faced with increasing risks.57 Other civilian missions operate or have operated in theatres with a high security risk rating58 such as Niger, Mali, Libya and Afghanistan,59 so the question is how long will civilian CSDP be able to benefit from symbolic engagements? This is linked to the issue of success. There is an incentive for the EU to declare its missions a success as mission success is clearly linked to the EU’s credibility as an international actor. In other words, labelling an operation as ‘successful’ constitutes a way to legitimize the EU’s international role. As a result, the EU has often been quick to declare CSDP missions a ‘success’. This has been even more so in the case of the Balkans and the Eastern neighbourhood, where the EU deployed its first missions.60 This is problematic, not only because it inflates expectations about future missions – widening the capabilities–expectation gap further – but also because this uncritical 54

De Zan, Tessari and Venturi (n 51). Menon (n 3). 56 C Gourlay, ‘The Politics of EU Civilian Interventions and the Strategic Deficit of CSDP’ (EU Crisis Management Paper Series, DCAF Brussels-ISIS Europe 2011) 13. 57 In April, an OSCE monitor was killed in East Ukraine. ‘American OSCE monitor killed in rebel east Ukraine’ (Euroactiv.com, 24 April 2017) https://www.euractiv.com/section/globaleurope/news/american-osce-monitor-killed-in-rebel-east-ukraine/, last accessed 23 July 2018. 58 Fearon and Picavet, ‘Challenges for civilian CSDP missions’ (n 31) 90. 59 In the case of EUPOL Afghanistan, three casualties were reported during its ten-year mandate: Tardy (n 34). 60 Juncos (n 11); X Kurowska and B Tallis ‘EU Border Assistance Mission: Beyond Border Monitoring?’ (2009) 14(1) European Foreign Affairs Review 47. 55

104 Research handbook on the EU’s common foreign and security policy position might prevent the EU from learning some of the lessons and best practices identified during the implementation process.

6. THE UGLY: THE POLITICS OF CIVILIAN CRISIS MANAGEMENT Because of the political nature of civilian CSDP, it has been politicized and contested since its origins. In particular, this section focuses on three types of political contestation – intergovernmental, bureaucratic and local politics – and how this has affected the implementation of civilian CSDP. 6.1 Intergovernmental Politics From its very beginnings, the development of civilian crisis management has been shaped by the different perceptions held by the Member States on the role of civilian instruments (as opposed to military ones), to the point that it has been argued that, in the case of the CSDP, ‘Member States have engaged in almost perpetual conflict over the nature of this policy area’.61 While France has traditionally emphasized the role of military instruments in international security and sought to develop the military dimension of CSDP, Nordic countries have, on the other hand, stressed the importance of civilian instruments. As mentioned earlier, the EU has developed over time an integrated approach, with the official rhetoric emphasizing the EU’s ability to deploy both civilian and military instruments. However, this should not hide the fact that civilian crisis management has been seen by some Member States as a way to counter-balance the militarization of the EU (e.g. as perceived by the neutrals) or to better complement NATO (the UK’s view). As far as leadership is concerned, it is interesting to note that it has been small and medium-sized countries, rather than the Big Three (France, Germany, UK), which have taken a leading role in the institutional development of EU civilian crisis management and, in particular, the Nordic countries. Not only did they put civilian crisis management firmly on the agenda at Cologne despite French opposition, they also played a key role in specific initiatives aimed at strengthening the decision-making and operational capabilities of the civilian dimension of the CSDP. For instance, the creation of a committee for civilian crisis management (CIVCOM) agreed at Helsinki (1999) was a Swedish proposal.62 The Nordic EU Member States have also taken the lead on capability development. For example, Sweden was one of the Member States

61 A Menon, ‘Power, Institutions and the CSDP: The Promise of Institutionalist Theory’ (2011) 49 Journal of Common Market Studies 89. 62 PV Jakobsen, ‘Small States, Big Influence: The Overlooked Nordic Influence on the Civilian ESDP’ (2009) 47 Journal of Common Market Studies 93; F Lee-Ohlsson, ‘Sweden and Development of the European Security and Defence Policy: A Bi-directional Process of Europeanization’ (2009) 44 Cooperation and Conflict 129.

Civilian CSDP missions 105 successfully sponsoring the Civilian Headline Goal 2008 and, together with Germany, it launched the idea of Civilian Response Teams.63 Regarding specific operations, one also needs to acknowledge the leading role of the Member States. Just as with military operations, civilian CSDP missions are considered by the Member States as a way to upload their own interests, as in Lithuania in the case of Themis or Poland in the case of EUAM Ukraine. However, it is worth noting that, as a general rule, missions have tended to be reactive rather than proactive. In other words, in most cases, a proposal by a Member State was introduced as a response to a request by another international organization (mainly the UN) or another third state, e.g. the host nation.64 The intergovernmental politics of civilian crisis management are also evident in the fact that Member States will seek to shape the mandates of the missions in accordance with their own interests. This might sometimes lead to inefficiencies on the ground, especially when the lack of consensus among conflicting Member State interests leads to vague or weak mandates. In other cases, civilian crisis management operations are seen by the Member States as a substitute for action where there is a lack of consensus as to a more robust response. As noted by Nicoletta Pirozzi, civilian CSDP missions are ‘less controversial from a political and financial point of view at national level, as they do not imply the use of military force and can be financed through the common EU budget (CFSP budget line and other financial instruments)’(see also Chapter 6 in this volume).65 For instance, in the case of Africa, political considerations have meant that, despite French pressure to launch military operations, the majority of EU Member States have opted for civilian capacity-building missions. Moreover, financial considerations have also played a key role during times of austerity. From the point of view of the Member States, civilian missions also have a comparative advantage vis-à-vis Commission initiatives since they remain an intergovernmental tool, in other words ‘instruments in the hands of Member States, which grants them political clout and credibility vis-à-vis recipient countries that other instruments may not enjoy’.66 While ensuring a sense of ownership by the Member States is always a positive element, this can have negative consequences in terms of effectiveness and coherence. For instance, Member States might seek to shape the mission during the implementation process by micromanaging missions from Brussels.67 6.2 Bureaucratic Politics Civilian CSDP missions have also suffered from bureaucratic politics. From this perspective, each bureaucratic organization has its own specific interests (often in the 63

ibid. Gourlay, ‘The Politics of EU Civilian Interventions’ (n 56) 7. 65 Pirozzi, ‘The European Union and Civilian Crisis Management’ (n 49) 300. 66 Tardy (n 52) 2. 67 F Ejdus, ‘Planning Ahead: EU Crisis Management Interventions Require Local Ownership at the Earliest Stage’ (Policy Briefing 38, University of Bristol 2016) http://www.bristol.ac.uk/ media-library/sites/policybristol/documents/PolicyBristol_Briefing_38_EU_crisis_management. pdf, accessed 27 February 2017. 64

106 Research handbook on the EU’s common foreign and security policy form of increasing competences, size, budget, prestige) that it will seek to promote, resulting in constant competition among organizations.68 This was particularly the case during the pre-Lisbon period, when both the European Commission and the Council Secretariat sought to protect/extend their areas of competence in civilian crisis management.69 Whether it was EUPM or EUJUST THEMIS, CSDP missions suffered from problems of institutional turf wars between the Commission and the Council.70 Problems of coherence were also evident between civilian and military operations as in the case of EUFOR Althea and EUPM in Bosnia, where both missions were active in the fight against organized crime.71 The Lisbon Treaty and the development of the comprehensive/integrated approach have gone some way to dealing with these problems. However, some overlaps and inconsistencies remain and a culture of coordination still to needs to be developed. As such, the potential offered by the newly established EEAS has not always been fulfilled. According to Major and Mölling, ‘disputes over competences and insufficient coordination inside the EEAS impair its capacity to act’.72 For one thing, there have been some problems of coordination between CMPD and Civilian Planning and Conduct Capability (CPCC).73 Thematic and regional expertise are not always appropriately brought together to deal with specific crisis in the neighbourhood and the connection between the CSDP structures and the rest of the EEAS machinery has remained ambiguous.74 The most recent reorganization of the EEAS in 2017 led to the establishment of a new thematic division for conflicts and crises: ‘PRISM’ (Prevention of Conflicts, Rule of Law/Security Sector Reform, Integrated Approach, Stabilisation and Mediation). The new division is the result of a merger between the division for Conflict Prevention, Peacebuilding and Mediation (SECPOL.2) and the division for Common Security and Defence Policy Coordination and Support division (CSDP.1). How PRISM will coordinate with other divisions in charge of civilian CSDP missions (CPCC and CMPD) remains to be seen. The new Joint Communication on the Integrated Approach should also help bring more coherence into the system. Problems have also related to financing and procurement, which have affected the implementation of the missions since the early days of the CSDP and even before.75 These also provided indications of bureaucratic turf wars between the Council and the 68

G Allison and P Zelikow, Essence of Decision: Explaining the Cuban Missile Crisis (2nd edn, Addison Wesley Longman 1999). 69 N Klein, European Agents Out of Control? Delegation and Agency in the Civil-military Crisis Management of the European Union 1999–2008 (Nomos 2010). 70 E Gross, ‘EU Military Operation in the Former Yugoslav Republic of Macedonia (Concordia)’ in Grevi, Helly and Keohane (eds) (n 33); I Ioannides, ‘EUPOL PROXIMA/EUPAT (fYROM)’ in Grevi, Helly and Keohane (eds) (n 33). 71 Juncos (n 11). 72 C Major and C Mölling, ‘Towards an EU Peacebuilding Strategy: The Effects of the Lisbon Treaty on the Comprehensive Approach of the EU in the Area of Civilian Crisis Management’ (2013) 18 European Foreign Affairs Review 57. 73 Pirozzi (n 49) 303. 74 Major and Mölling (n 72) 57. 75 For problems relating to the financing of the EU Administration in Mostar, see Juncos (n 60).

Civilian CSDP missions 107 Commission, and between the Council and the European Parliament. Moreover, these problems also demonstrated tensions between CSDP and EU development policies despite the EU’s efforts to implement the security-development nexus and a comprehensive approach. Accordingly, the Implementation Plan for Security and Defence of 2016 includes recommendations ‘on increased financial incentives and solidarity, more flexibility as well as a more seamless range of funding options’.76 The adoption of the EUGS and the Implementation Plan on Security and Defence (as well as other key recent documents such as the Capacity Building for Security and Development and the new SSR Strategy) offers some prioritization and guidelines as to how to move forward with the implementation of a strategic vision in security and defence policies, which should help reduce some of the past problems of coordination. 6.3 Local Politics Finally, one needs to consider the impact of local politics: the role that political conflict among local elites might have on the implementation of civilian CSDP. Obviously, EU missions are not implemented in a vacuum and thus the political climate (from latent conflict to open war) in the target country is of utmost importance. For instance, non-cooperative elites might obstruct civilian CSDP missions in order to safeguard their political and/or economic interests. This is because, despite attempts to portray some EU initiatives as technocratic and/or neutral, CSDP initiatives are inherently political. They touch upon highly sensitive issues such as who exercises the monopoly over the use of force or control over a given territory, affecting the balance of power among conflict parties. By promoting a specific policy, the EU empowers and disempowers one side or the other of any given conflict.77 For these reasons, local politics have affected civilian missions, and above all their effectiveness. In the worst cases, opposition to the EU’s intervention put the safety of EU personnel at risk. The shooting down of a helicopter of the European Community Monitoring Mission in 1991 is the most tragic example of the impact of local politics on EU civilian missions. EUPOL Afghanistan also provides an illustration of how a difficult war-like context can undermine the effectiveness of the mission and the medium- and long-term sustainability of the reforms.78 It is noteworthy that the EU has opted for technocracy as a substitute for strategy, with the EU choosing to deliberately keep its profile low and to de-politicize its engagement. This has particularly been the case in the Eastern neighbourhood so as to not antagonize Russia.79 Here, the EU has relied on the European Neighbourhood Policy as the main framework to deal with conflicts in the region since it is seen as a long-term and largely technocratic (non-political) exercise and thus a more palatable instrument. In some cases, the EU shied away from using the ‘CSDP’ denominator and 76

Council of the EU, ‘Implementation Plan on Security and Defence’ (n 22) 28. AE Juncos, ‘Europeanization by Decree? The Case of Police Reform in Bosnia’ (2011) 49 Journal of Common Market Studies 367. 78 Tardy (n 34). 79 A Huff, The role of EU defence policy in the Eastern neighbourhood (Occasional Paper 91, EU–ISS 2011). 77

108 Research handbook on the EU’s common foreign and security policy instead used the Commission as an umbrella – e.g. with the EUBAM mission to Moldova and Ukraine and the EU Border Support Team in Georgia. But more telling of the EU’s current approach in Eastern Europe are those cases of ‘missed opportunities’, where the EU could have launched a civilian mission and decided against doing so, such as in Georgia in 2005. In this case, the Themis mission in Georgia deliberately excluded from its mandate the frozen conflicts of South Ossetia and Abkhazia. When the existing OSCE Border Monitoring Operation on the Georgian–Russian border had to be withdrawn because of the Russian veto in 2005, the EU was unable to deploy a civilian CSDP mission to replace the OSCE operation.80 As a result, EUJUST Themis lacked a real strategic outlook. In other cases, what we have seen is an attempt by the EU to portray its civilian missions as non-political in nature. A case in point is the EUPM in Bosnia (2002– 2012). The projects undertaken by the EUPM were presented as technical reforms and justified as a way to increase the efficiency and rationalization of Bosnian police forces. The language of the EUPM mandate avoided references to political elements and instead chose neutral terms such as ‘sustainable’, ‘best practice’, and so on. No mention, however, was made of how these reforms might impact on the balance of power among the conflict parties or the legitimacy of the EU’s intervention more generally.81 The politicization of the EUPM had the opposite effect of increasing opposition by the local elites and, in particular, Bosnian Serbs, who saw the police reform as an attempt to centralize police structures. In other cases, however, the technical nature of EU interventions has not hindered the (perceptions of) effectiveness of the mission such as in the case of EUBAM Moldova Ukraine.82 More recently, the EU has sought to increase local ownership as an attempt to increase buy-in and improve the legitimacy and effectiveness of its missions. Local ownership has thus become a buzzword in Brussels, with an increased emphasis at the EU level on the need to increase the involvement of the host governments and societies. This is in line with the idea of building the resilience and capacity of the EU’s neighbouring countries (see above). Thus, the EUGS calls for a ‘bottom up approach’83 – or at least to ‘blend top-down and bottom-up efforts’ – to foster local agency. As put by the EUGS, ‘[p]ositive change can only be home-grown, and may take years to materialize’.84 As mentioned earlier, most of the CSDP civilian missions (with the exception of EULEX Kosovo) have favoured an approach that emphasizes supporting rather than replacing local capacities in line with the idea of local ownership. In practice, however, as shown by the case of EUCAP Nestor, civilian CSDP missions have fallen short of promoting local ownership, especially when it comes to the design (and not just the implementation) of the missions.85 80

ibid, 19. AE Juncos, ‘Member State-building vs Peacebuilding. The Contradictions of EU Statebuilding in Bosnia and Herzegovina’ (2012) 29 East European Politics 58. 82 Zarembo (n 45). 83 ‘Shared Vision, Common Action’ (n 5) 31. 84 ibid, 27. 85 F Ejdus, ‘“Here is your mission, now own it!” The rhetoric and practice of local ownership in EU interventions’ (2017) 26(4) European Security, http://dx.doi.org/10.1080/ 09662839.2017.1333495. 81

Civilian CSDP missions 109

7. CONCLUSION This chapter has shown how civilian CSDP has become an important component of this policy. While it has not always figured in political and academic debates about the future of CSDP, the civilian dimension of CSDP has made significant progress in terms of institutionalization and operationalization. In particular, the reforms brought forward by the Lisbon Treaty, including the establishment of the double-hatted High Representative, have opened the way for a more flexible approach and more interactions with Commission instruments. However, institutional reforms alone have not been sufficient to mobilize the necessary political will and resources to make the EU’s ambitions in civilian crisis management a reality.86 Other factors, such as increasing insecurity in the neighbourhood (including the Ukrainian crisis and the refugee crisis to name a few) and internal events (Eurozone crisis, rise of Euroscepticism, Brexit) which have informed the EUGS and the Implementation Plan for Security and Defence, have had (and promise to have) more of an impact on the development of this policy. There are many reasons to believe that civilian CSDP will continue to be a key instrument in the EU’s security toolbox. First, the EU prides itself on the ability to adopt an integrated approach to security, combining civilian and military means.87 This constitutes a comparative advantage vis-à-vis other security organizations such as NATO. Secondly, the idea of the EU as a ‘civilian’ or ‘normative power’ also relies to a great extent on the ability to use civilian means to achieve its objectives. Civilian crisis management missions help promote the EU’s values and norms beyond its borders by promoting and strengthening the rule of law, human rights and democracy, which, more often than not, require a long-term and non-executive approach such as the one privileged by civilian CSDP. The EU’s missions have also shown some degree of flexibility and adaptability to new security challenges, although this flexibility has been limited by institutional constraints, for example, limited integration with justice and home affairs instruments. The future of civilian CSDP will be determined by how well the EU addresses some of the challenges identified in this chapter and, in particular, the need to bridge the capabilities–expectation gap and effectively manage the politics of civilian crisis management. The first challenge has been acknowledged at all levels. However, for years the EU and its Member States have failed to commit the quality and quantity of personnel required for civilian CSDP, widening further the capability–expectation gap. Although expectations were lowered after the entry into force of the Lisbon Treaty, an increasing sense of insecurity as a result of the crises affecting the Eastern and Southern neighbourhood could well provide the necessary impetus to improve the recruitment and training of civilian personnel serving in CSDP missions as demonstrated by the new initiative establishing a civilian CSDP Compact. More difficult will be to address the problems relating to the politics of civilian crisis management. Intergovernmental politics will continue to play a key role in shaping this policy area although, as before, the insecurities relating to external and internal (Brexit) crises might also have a unifying effect, at least in the short and medium term. A move 86 87

In this regard, I agree with the overall assessment of Pirozzi (n 49). ‘Shared Vision, Common Action’ (n 5).

110 Research handbook on the EU’s common foreign and security policy towards a more integrated approach could also lessen the effects of bureaucratic politics although they will never be completely eliminated. Finally, a stronger commitment to and implementation of the local ownership principle could also go some way to addressing the impact of local politics.

* Research for this article has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no.: 653227. The content reflects only the author’s views, and the European Commission is not responsible for any use that may be made of the information it contains.

6. Military CSDP operations: strategy, financing, effectiveness Daniel Fiott

1. INTRODUCTION On the back of successful military deployments to the Democratic Republic of the Congo (Operation Artemis), Chad and the Central African Republic (EUFOR Chad/ RCA) and the Horn of Africa (Operation Atalanta), the EU can draw on an increasing wealth of operational knowledge and expertise. Yet, as the EU develops a more ‘comprehensive’ and ‘integrated’ approach to crisis response, a number of operational, institutional and legal issues present themselves. The nature and political objectives of CSDP military operations are driven by the evolving security context in the EU’s near and wider neighbourhood. In many respects, the scope, extent and objectives of CSDP military operations are evolving in light of a shifting institutional context with a growing role for the European External Action Service (EEAS) and the High Representative/Vice-President for the EU’s foreign and security policy (HR/VP). Furthermore, CSDP military operations occur in an evolving politico-military context in which the military approach of the EU is increasingly tailored to crisis situations. Not only have non-EU, ad hoc operations such as those in Iraq and Afghanistan blunted the appetite for large-scale deployments, but there is also recognition on the part of military planners that to simply address the symptoms of crises is not strategically sustainable. What is often termed the ‘comprehensive approach’ in EU parlance is, when translated into politico-military planning, a desire to not see any backsliding into crises or instability once a sustainable level of security has been achieved.1 The aim of this chapter is to look at the intersection of the legal and operational parameters in which military operations under the CSDP are deployed. To this end, the chapter focuses on three broad challenges and it draws on relevant case studies to provide greater empirical context. The first challenge relates to the relationship between the initial legal parameters of deployment and how operational decisions may lead to and require politico-strategic flexibility. In this context, the chapter draws on early lessons from Operation Sophia to show how the original political objectives for deployment (i.e. reconnaissance and the disruption of trafficker networks) are increasingly giving way to new operational needs (i.e. less about littoral operations near Libya and more about solace operations). The second challenge relates to the available funding and legal issues surrounding the deployment of CSDP operations. In this context, the chapter looks at some of the challenges facing the EU’s Training Missions (EUTMs) and what a lack of funding and legal clarity for such operations means in 1 Commission/EEAS, ‘Joint Staff Working Document: Taking Forward the EU’s Comprehensive Approach to External Conflict and Crises – Action Plan 2015’ SWD (2016) 254 final.

111

112 Research handbook on the EU’s common foreign and security policy terms of mission effectiveness, lessons learned and the comprehensive approach. The final challenge relates to how geopolitical and strategic shifts affect the nature and remit of EU institutions involved in the deployment of CSDP military operations.

2. EVOLVING STRATEGIC CONTEXTS 2.1 Politico-strategic Planning Structures One important feature of CSDP military operations is how the politico-strategic objectives and legal parameters set down by the Council of the EU in advance of a CSDP deployment are liable to change during the life-cycle of a military operation. Such change is driven by the strategic context in which CSDP operations are deployed. Indeed, a feature of any military operation is the interaction between the politicostrategic level of operations deployment (‘political guidance’) and operational and/or tactical requirements (‘the facts on the ground’). Within the EU there is a well-defined political process guiding this politico-strategic and operational/tactical interaction (see Figure 6.1). To begin with, Article 42(4) Treaty on European Union (TEU) states that the Council – acting unanimously on a proposal from the HR/VP – can launch an operation under the CSDP. Any decision to deploy a CSDP military operation begins with the unanimous consent of the Political and Security Committee (PSC) and formal adoption by the Foreign Affairs Council (FAC). Along with the EU Military Committee (EUMC), which is a Council body composed of EU Member State military officials, the PSC and FAC retain politico-strategic oversight of CSDP military deployments, both at the inception of these operations and during the conduct of them. As the chair of the FAC, the HR/VP plays an important role in overseeing the relevant EU institutions through the planning phase of a military operation. After the PSC, FAC, HR/VP and EUMC have agreed in principle that an EU military operation is needed, work begins on drafting what is termed the Crisis Management Concept (CMC). The CMC is a planning document that brings together the EU’s overall strategic objective in a given crisis situation, with the tools the EU has available to conduct its CSDP operations (i.e. military capabilities). The CMC is, in essence, designed to offer the PSC, FAC and HR/VP a range of possible operational options. The overall body responsible for drafting the CMC is the EEAS’ Crisis Management and Planning Directorate (CMPD). The CMPD is a post-Lisbon creation designed to ensure an integrated civilian-military planning structure for the EU. Among other duties, the CMPD ensures strategic planning for CSDP operational deployment and it reviews ongoing missions. Informing the work of the CMPD on the CMC are the EU’s very own military staff (EUMS) and the Civilian Planning and Conduct Capability (CPCC), which provides advice and coordination on civilian and civ-mil operations deployed under the CSDP. The EUMS is crucial to the deployment of CSDP military operations because it serves as the EU’s early warning, situational awareness and strategic planning hub. Although the EUMS provides strategic advice to the HR/VP, the Military Staff mainly work with the EUMC – a body composed of the Chiefs of Defence (CHODs) of each EU Member State.

Military CSDP operations 113 Once in possession of the CMC, the PSC is able to take a decision on the military-strategic option best suited to the crisis situation. The PSC may consult a number of working groups in the Council of the EU when deciding on which option to pursue. These bodies usually include the EUMC and the Politico-Military Group (PMG) and the working group of foreign relations counsellors (RELEX). Once a decision has been taken by the PSC, the FAC publishes a Council Decision and on this basis a Head of Mission and/or Operational Commander is/are appointed and decisions regarding financial arrangements and capabilities for the military operation are taken.2 Following the Council Decision, the Concept of Operations (CONOPs) and Operation Plan (OPLAN) are drafted to outline the tactical and operational specifics of the military operation (i.e. capabilities, logistics, supply, etc.). Before a CSDP military operation is deployed, however, the EU also has to decide on command and control (C2) and this usually rests on one of three possible C2 or EU Operation Headquarters (EU OHQ) options: the EU cell based at NATO’s Supreme Headquarters Allied Powers Europe (SHAPE), national OHQs provided by an EU Member State/s3 or the EU Operations Centre (OPCEN). On this basis, the EU can undertake executive and/or non-executive operations under the CSDP.4 Of course, the UK’s decision to leave the EU will affect the nature of the EU’s OHQs and the newly created Military Planning and Conduct Capability (MPCC). Created after the UK’s EU referendum, the MPCC has subsumed OPCEN and it is based at the EEAS for the purposes of taking command of all the EU’s non-executive operations, including the EUTMs for the Central African Republic, Mali and Somalia.5 Unless arrangements between the EU and the UK on CSDP military operations and C2 can be found once the UK officially leaves the EU in March 2019, then it is likely that the UK will no longer be able to plug into, draw from or lead the EU’s C2 capabilities. For example, the Council of the EU decided on 30 July 2018 to transfer the operational HQ of EUNAVFOR Atalanta from Northwood (UK) to Rota (Spain) and it also replaced the former British commander of the operation with a Spanish Vice Admiral.6 Furthermore, NATO’s Deputy Supreme Allied Commander Europe (DSACEUR) has traditionally been a UK citizen and he has also occupied command posts for certain EU military operations in the past (e.g. EUFOR Althea), as part of the Berlin Plus 2

A Mattelaer, ‘Decision Making in the Field of CSDP’ in J Rehrl and HB Weisserth (eds), Handbook on CSDP: The Common Security and Defence Policy of the European Union (Federal Ministry of Defence and Sports of the Republic of Austria 2013). 3 France, Germany, Greece, Italy and the UK. 4 The EU Concept for EU-led Military Operations and Missions makes a distinction between executive and non-executive operations. Executive missions refer to military operations that directly replace host nation forces, whereas non-executive missions are designed to support the host nation in an advisory role only. See: Council of the EU, ‘European Union Concept for EU-led Military Operations and Missions’ 17107/14 (Brussels, 19 December 2014) 9. 5 Council of the EU, ‘Concept Note: Operational Planning and Conduct Capabilities for CSDP Missions and Operations’ 6881/17 (Brussels, 6 March 2017) 8–9. 6 Council of the EU, ‘EUNAVFOR Somalia Operation Atalanta: Council prolongs the Operation and decides on new headquarters and new Operation Commander’ accessed 10 September 2018.

114 Research handbook on the EU’s common foreign and security policy arrangements between NATO and the EU. Any such double-hatted role may be called into question in the future depending on the terms of the UK’s EU departure.7 Political and operational guidance for EU military operations are linked through the relationship of different bodies and individuals (the ‘chain of command’). After the PSC/FAC have given their political consent they appoint an EU Operation Commander (OpCdr), who is responsible for developing both the CONOPs and OPLAN. The OpCdr takes charge of the military-strategic level in the chain of command, which refers mainly to the coordination of the deployment and sustainability of EU forces that have been deployed. Acting under the authority of the OpCdr after appointment by the PSC/FAC is the EU Force Commander (FCdr). The FCdr serves at the operational level and executes the military operation by commanding the assigned forces (land, air, maritime) during the operation.8 Both the OpCdr and FCdr are continuously supported by the EUMS with regular monitoring of the strategic environment.9 In addition to the EUMS, a number of other bodies and agencies provide situational awareness for CSDP military operations. For example, the EEAS’ EU Intelligence and Situation Centre (INTCEN) provide both shared national intelligence reports and open source material. The EU’s Satellite Centre (SATCEN) also features in CSDP operations planning and oversight through the provision of satellite imagery. 2.2 Operation Sophia Despite this well-defined chain of command, the tactical needs of military operations can test the adaptability of CSDP operations when they are deployed in theatre. The recent example of the EU’s naval deployment to the Mediterranean through EUNAVFOR MED Operation Sophia is indicative of the need to have a flexible CONOPs and OPLAN, which can adapt politico-strategic guidance on the basis of operational and tactical requirements. This is especially true in the case of the EU’s naval deployments (of which there are currently two). As far as the history of the EU’s military deployments is concerned, naval operations are relatively new military tasks for the EU. For example, of the 11 military operations deployed by the EU since 2003 (e.g. EUFOR Chad/RCA, Artemis and Althea) nine have been land-based missions focusing on separation of forces, stability missions and military training. This is not to say that these land-based operations are not subject to similar politico-strategic shifts to those observed during maritime operations, but the EU is well versed in operating in inhospitable and hostile land environments. What is new about maritime operations – and especially Operation Sophia – is that there are far fewer naval operational experiences from which the EU can draw lessons, even though Europe’s navies have a wealth of maritime experience. Nevertheless, from 7

J Black, A Hall, K Cox, M Kepe and E Silfversten, Defence and Security After Brexit: Understanding the Possible Implications of the UK’s Decision to Leave the EU (RAND Corporation 2017) 26. 8 Council of the EU, ‘EU Concept for Military Command and Control’ EEAS 02021/7/14 (Brussels, 5 January 2015) 8–9. 9 Council of the EU, ‘EU Concept for Military Planning at the Political Strategic Level’ 6432/15 (Brussels, 23 February 2015) 11.

Military CSDP operations 115

Source: European Security and Defence College, 2016.

Figure 6.1 CSDP military operations decision-making bodies a geopolitical point of view the maritime domain has emerged as a key operational theatre for the EU. Indeed, the majority of CSDP operations that have been launched on land in the wider neighbourhood (i.e. to areas such as the Sahel) are premised on a more traditional crisis management paradigm that seeks to contain crisis. What is interesting about the CSDP’s more recent focus on maritime security is that it is more of a reflection of the internal–external security nexus. The maritime domain is one where flows of people, arms, trade and so on find their way from external spaces into the EU. In this sense, the EU is strategically connecting geopolitical spaces in its immediate and wider neighbourhood (i.e. connecting land and maritime domains).10 Interestingly, EU Member States have been far more forthcoming with the types of naval and air assets (see Chapter 8 in this volume) required to effectively secure maritime domains. EU Member States have also not approached CSDP naval operations with the same degree of hesitation that has accompanied the deployment of the (as yet unused) EU Battlegroups for land operations. This is perhaps because EU naval operations are designed to respond to security concerns that directly (and sometimes quite visibly) 10 K Schilde, ‘European Military Capabilities: Enablers and Constraints on EU Power?’ (2017) 55 Journal of Common Market Studies 37.

116 Research handbook on the EU’s common foreign and security policy impact European citizens in Europe itself.11 EU Battlegroups are generally associated with out-of-area operations, meaning that deployment is bedevilled by concerns about a potential loss of troops (the ‘CNN effect’) and the ethics of military intervention more broadly.12 Indeed, EUNAVFOR MED Sophia was deployed as a response to human smuggling and trafficking through the Mediterranean corridor between Libya and EU territory. Operation Sophia is but one element of the EU’s broader approach to relations with Libya, which includes financial support for the protection of migrants and migration management in Libya. Operation Sophia was established by the EU Council on 18 May 2015 and it deployed on 22 June 2015. The core mandate of the operation is thus to contribute ‘to the disruption of the business model of human smuggling and trafficking networks in the Southern Central Mediterranean’.13 At this stage no UN Security Council Resolution (UNSCR) was required because the first phase of the operation – surveillance – occurred in international waters.14 As of 11 September 2018, five surface vessels are on operations in the Mediterranean and these naval assets are supported by two helicopters and four aircraft. However, since 2015 Operation Sophia has seen the deployment of 50 naval vessels and 19 aircraft at intervals to ensure average force strength of approximately five naval units and six air assets at any point during the operation. Twenty-five Member States have contributed to the operation. Belgium, France, Germany, Greece, Italy, Luxembourg, the Netherlands, Portugal, Slovenia, Spain and the United Kingdom have dedicated hard assets to the operation in the form of naval vessels and aircraft.15 Other Member States have contributed specialist personnel (e.g. medical personnel) and equipment. Operation Sophia is structured around four specific operational phases: phase one – to undertake surveillance tasks to better understand smuggling activities and methods (June–October 2015); phase two – seizure and diversion of smugglers’ vessels (October 2015–June 2016); phase three – destruction of smuggling vessels and related assets (June 2016–present); and phase four – withdrawal of forces and completion. When compared to the deployment of other CSDP military operations, the time taken between the CMC and operational deployment was remarkably short: it took one month to generate the forces required for the operation.16 Despite this level of engagement with the operation, however, a key challenge has been to ensure an operational focus in a geopolitical area where numerous security and political challenges compete for a response. First, securing a mandate for the 11 M Riddervold, ‘New Threats, Different Response: EU and NATO and Somali Piracy’ (2014) 23 European Security 546. 12 D Fiott, ‘The Use of Force and the Third Pillar’ in D Fiott and JA Koops (eds), The Responsibility to Protect and the Third Pillar: Legitimacy and Operationalization (Palgrave Macmillan 2015) 130–145. 13 Council of the EU, ‘Council Decision on a European Union military operation in the Southern Central Mediterranean’ CFSP 2015/778 (Brussels, 18 May 2015) 2. 14 S Blockmans, ‘New Thrust for the CSDP from the Refugee and Migrant Crisis’ (2016) CEPS Special Report No. 142, 5. 15 European External Action Service, ‘EUNAVFOR MED operation SOPHIA – Assets’ accessed 20 April 2017. 16 Blockmans (n 14) 4.

Military CSDP operations 117 second phase of the operation was difficult. CSDP military operations are deployed either under a UNSCR, by invitation from a host state and/or through a combination of the two. As late as September 2015, however, the Russian government had vetoed a UNSCR giving a mandate to the EU to act in the southern central Mediterranean with military force, if required. At the same time, the Libyan unity government was still unformed – and rival power centres were located in Tripoli and Tobruk – which made the potential use of force especially challenging.17 To secure both a UNSCR mandate and the blessing of the Libyan authorities, the EU had to augment the operation to include capacity-building training for Libyan coastguards and navy and to assist with the UN arms embargo in Libya. These tasks were important additional tasks for Operation Sophia, which addressed some of the specific needs of the Libyan government. This was not the only example of EUNAVFOR Sophia being subject to the interplay between security and humanitarianism in the Mediterranean.18 Even before the physical deployment of the operation, analysts were pointing to the seeming mismatch between the specific objectives of the operation and the overall strategy for what the EU wanted to achieve in Libya.19 Without consensus in the UNSC and consent from the Libyan authorities, there was only so much the EU could do to stop smugglers and traffickers operating from a maritime position located outside of Libyan territorial waters. Not only did the initial ‘narrowness of the operation’s mandate’ raise questions about how smuggling networks would adapt their behaviour considering the EU’s operational restrictions, but the operation would have to play the dual role of meeting politicostrategic and tactical objectives as well as acting to prevent loss of life at sea.20 Relevant EU bodies clearly understood this dilemma. In its guidance to the PSC of 15 May 2015, for example, the PMG recognized that even though the core mandate of the operation related to smuggling and trafficking, naval vessels participating in the operation could not neglect their international obligations under the Safety of Life at Sea (SOLAS), the UN Convention on the Law of the Sea (UNCLOS) and the Search and Rescue (SAR) Convention – that is, they had to be ready and able to conduct search and rescue tasks.21 Yet this sort of ‘mission creep’ was not the only operational issue facing the EU, for the ability of the EU to develop a comprehensive approach to the crisis in Libya was 17

T Tardy, ‘Operation Sophia: Tackling the Refugee Crisis with Military Means’ (Brief Issue, EU-ISS 30 September 2015) accessed 14 April 2017. 18 N Perkowski, ‘Deaths, Interventions, Humanitarianism and Human Rights in the Mediterranean “Migration Crisis”’ (2016) 21 Mediterranean Politics 331. 19 G Faleg and S Blockmans, ‘EU Naval Force EUNAVFOR MED sets Sail in Troubled Waters’ (CEPS Commentary, 26 June 2015) accessed 19 April 2017. 20 Tardy (n 17) 1 and 4. 21 Council of the EU, ‘Draft Crisis Management Concept for a Possible CSDP Operation to Disrupt Human Smuggling Networks in the Southern Central Mediterranean’ 8824/15 (Brussels, 12 May 2015) accessed 14 April 2017.

118 Research handbook on the EU’s common foreign and security policy also challenging. The ‘comprehensive approach’ aims to provide an encompassing diplomatic, political, economic and humanitarian approach to crises – i.e. not just ‘sending in the marines’ but looking to longer-term stability. Following on from the experiences of the EU’s other naval operation to the Horn of Africa – EUNAVFOR Atalanta – there was some hope that the EU could bring development aid, capacitybuilding and security together in the same way in Libya. Indeed, the Political Framework for Crisis Approach (PFCA) for Libya, which was drafted in October 2014, clearly spelled out the need to improve security in Libya with a view to laying down the conditions for economic development.22 Implementing a comprehensive approach in Libya would be extremely challenging, however, especially given the shifting political state in Libya itself. True, the EU had deployed EUBAM Libya in 2013 to help build border management capacity in the country, but the mission is civilian in nature and it is not even located in Libya itself: it is currently based in Tunis, Tunisia. It was clear that no ‘cut and paste’ from the EU’s experiences in the Horn of Africa would be possible in Libya. Operation Sophia has also been deployed in the context of increased interinstitutional cooperation between the EU and NATO. Although Operation Sophia was deployed before the 8 July 2016 signing of the EU-NATO Joint Declaration and the initial 42 action points agreed between the two organizations on 6 December 2016, this drive towards enhanced cooperation on defence has seen NATO deploy alongside the EU in the Mediterranean. Indeed, by launching Operation Sea Guardian in October 2016 as a non-Article 5 operation, NATO aims to enhance situational awareness and capacity-building in the Mediterranean. In a more practical sense, Sea Guardian not only manages to increase maritime surveillance in the Western and Eastern parts of the Mediterranean (in order to provide strategic depth and width to EUNAVFOR Sophia), but also plays a technical role by providing submarine and refuelling services. For NATO and the EU, maritime cooperation has emerged as a key element of interinstitutional cooperation; one must not forget that NATO is also working closely with the EU’s Frontex agency in the Aegean Sea to assist with the migration crisis.23 Both the EU and NATO operate in the context of SHADE MED to improve shared awareness and de-confliction in the Mediterranean.24 While cooperation is a political statement by both organizations, EU-NATO maritime operations are still affected by the constraints of the aforementioned rules of engagement in the Mediterranean.25 Finally, although the EU and NATO are enhancing maritime cooperation there are still 22

Council of the EU, ‘Libya, a Political Framework for a Crisis Approach’ 13829/14 (Brussels, 1 October 2014). 23 MJ McNerney, G Persi Paoli and S Grand-Clement, Cross-Cutting Challenges and Their Implications for the Mediterranean Region (Cambridge/RAND 2017). 24 SHADE MED is a forum for interested governments and organizations dedicated to sharing information on migration and maritime security in the Mediterranean basin. To date, SHADE MED has brought together 16 non-governmental organizations, 20 international organizations, 39 governments, 12 military representatives and/or missions, one university and four companies. See accessed 7 July 2017. 25 L Watanabe, ‘Libya’s Future: Uncertain, Despite a Political Agreement’ (2016) 23 Middle East Policy 114.

Military CSDP operations 119 questions about the ultimate strategic objectives both organizations share for maritime security in the Mediterranean given their distinct mandates.

3. FINANCING MILITARY OPERATIONS The evolving politico-strategic context is not the only important factor that affects the nature and effectiveness of CSDP military operations. Indeed, one of the perennial challenges facing EU policy planners is how to finance CSDP operations. Article 41(1) of the TEU states that all administrative expenditure associated with CSDP operations is charged to the EU budget, and CSDP civilian operations are funded directly from the Union budget under the ‘Global Europe’ budget line (Chapter 19 03 01). For CSDP military operations, however, a different set of procedures are in place as Article 41(2) TEU specifically forbids military operations from being funded from the EU budget (unless the Council of the EU specifically decides otherwise, which it has not done so far). Thus, for CSDP military operations the principle of ‘costs lie where they fall’ is observed, which means that those Member States that want to launch and be part of a military operation cover the costs. In addition to this principle, the Member States can also draw on the Athena mechanism26 for operations – this is a pooled fund calculated through a share of the gross national income index, which can fund common costs such as HQ running costs, fuel and even medical evacuation. This mechanism is an ad hoc fund and does not derive from the EU budget. 3.1 ‘Train and Equip’ Yet debates about funding CSDP military operations are well rehearsed. Arguments to date largely focus on burden-sharing within the Athena mechanism27 and even parliamentary scrutiny of the Athena mechanism and military operations.28 These factors do not require any further appraisal here but an area of increasing interest is the financing of particular types of CSDP military operations. More specifically, a number of military operations that have been deployed in recent years focus on security capacity-building in third countries. Such operations seek to ‘train and equip’ security forces in third countries with a view to stabilizing state structures. Indeed, EUTMs have 26 Council Decision (CFSP) 2015/528 of 27 March 2015 establishing a mechanism to administer the financing of the common costs of European Union operations having military or defence implications (Athena) and repealing Decision 2011/871/CFSP [2015] OJ L84/39. 27 F Terpan, ‘Financing Common Security and Defence Policy Operations: Explaining Change and Inertia in a Fragmented and Flexible Structure’ (2015) 24 European Security 221; NIM Nováky, ‘Who Wants to Pay More? The European Union’s Military Operations and the Dispute over Financial Burden Sharing’ (2016) 25 European Security 216; and F Mérand and A Rayroux, ‘The Practice of Burden Sharing in European Crisis Management Operations’ (2016) 25 European Security 442. 28 J Wouters and K Raube, ‘Seeking CSDP Accountability Through Interparliamentary Scrutiny’ (2012) 47 The International Spectator: Italian Journal of International Affairs 149; and A Huff, ‘Executive Privilege Reaffirmed? Parliamentary Scrutiny of the CFSP and CSDP’ (2015) 38 West European Politics 396.

120 Research handbook on the EU’s common foreign and security policy been deployed to Mali (in 2013), the Central African Republic (in 2016) and Somalia (in 2010) to help modernize and professionalize forces into ethical and democratically accountable forces, and to promote effective responses to threats such as terrorism. Quite apart from these operations to Africa, the Council of the EU also views capacity-building through training as a way to improve partner capacities in ‘the areas of strategic communication, cyber security and border security’.29 Financing EUTMs has created a particular challenge because there are restrictions on how far EU funds (the EU budget or otherwise) can be used to procure arms, ammunition and equipment for military forces in third countries. As has been stated, Article 41(2) TEU strictly forbids using the EU budget to fund military operations unless the Council of the EU so decides. This implies that a range of security-relevant financial instruments that derive from the EU budget cannot be used for military operations. For example, while the Instrument Contributing to Stability and Peace (IcSP) has been used in the past to train and equip police forces in places such as Niger, it is not entirely suited to military ‘train and equip’ missions. While the IcSP Regulation does not specifically forbid military ‘train and equip’ funding,30 the emphasis is clearly placed on civilian measures (see, for example, Article 3(2)(d and g)). Unlike the IcSP, however, Article 3(13) of the Development Cooperation Instrument (DCI) Regulation is much clearer on the prohibition of EU funding for the procurement of arms and/or ammunition.31 Consequently, ‘there is currently no EU budget instrument designed to provide a comprehensive financing to security capacity building in partner countries, in particular its military component’.32 Outside of the EU budget, however, more flexible responses can be found, even if restrictions on military ‘train and equip’ initiatives still exist. For example, in the past the African Peace Facility (APF) has provided the African Union (AU) with financing for training and exercises for peacekeeping operations such as that in Somalia (AMISOM). Funded under the European Development Fund (EDF), the APF is a good example of where non-EU budget financing mechanisms can be used to support military capacity-building initiatives in third countries although even the APF has ‘never been used to pay for equipment’.33 However, it should be considered that the EDF and the APF were not designed to conveniently circumvent the restrictions on the 29 Council of the EU, ‘Implementing the EU Global Strategy in the Area of Security and Defence’ 14149/16 (Brussels, 14 November 2016) 5. 30 In places the IcSP regulation is vague on the specific issue of funding for military ‘train and equip’ operations although this vagueness cannot realistically serve as a legal basis for the funding of ‘train and equip’ missions. For example, Article 4(2)(c) states that EU support for crisis response in third countries ‘may include technical and financial assistance for the implementation of peace-building and State-building support actions’ [author’s emphasis]. State-building actions would arguably include support for military services. Council of the EU, ‘Regulation Establishing an Instrument Contributing to Stability and Peace’ No. 230/2014 (Brussels, 11 March 2014). 31 Council of the EU, ‘Regulation Establishing a Financing Instrument for Development Cooperation for the Period 2014–2020’ No. 233/2014 (Brussels, 11 March 2014). 32 Commission/EEAS, ‘Joint Communication: Capacity Building in Support of Security and Development – Enabling Partners to Prevent and Manage Crises’ JOIN (2015) 17 final 8. 33 T Tardy, ‘Enabling Partners to Manage Crises – From “Train and Equip” to CapacityBuilding’, EU Institute for Security Studies Policy Brief, No. 18 (2015) 2 accessed 15 April 2017. 34 H Merket, The EU and the Security-Development Nexus: Bridging the Legal Divide (Brill Nijhoff 2016) 111. 35 European Commission, ‘The Cotonou Agreement’ accessed 20 April 2017. 36 C Lamb, S Arkin and S Scudder, The Bosnian Train and Equip Program: A Lesson in Interagency Integration of Hard and Soft Power (National Defense University Press 2014) 17–18. 37 Commission/EEAS (n 32) 3. 38 Tardy (n 33) 2 and JI Moreno Zamora, ‘Perspectivas de Futuro de la Iniciativa “Train and Equip” en el Contexto de la Union Europea’ (2015) 94 Instituto Español de Estudios Estratégicos Opinión accessed 20 April 2017. 39 European Peacebuilding Liaison Office, ‘Capacity Building in Support of Security and Development: Gathering Civil Society Input – Minutes’ (2015) 4 accessed 19 April 2017.

122 Research handbook on the EU’s common foreign and security policy (Case C-91/05) of 20 May 2008, for example, raised important questions about whether EU financial assistance to support the Economic Community of West African States’ (ECOWAS) steps to control the proliferation of small arms and light weapons (SALW) should have been provided on the basis of a Council Joint Action (Article 14 TEU on the CFSP) or on the basis of the EC Treaty (Article 179 TEC). Through its support to ECOWAS, the Commission had argued that the issue of SALW was an economic and social development issue rather than just an issue of security and proliferation. The CJEU has ruled that there was an incompatibility between the CFSP and EC legal bases and, in line with the pre-Lisbon ‘non-affect’ clause of Article 47 TEU, gave precedence to the latter.40 Post Lisbon, a similar issue of legal demarcation can be observed in the CJEU’s judgment in European Parliament v Council of the EU (Case C-263/14) of 14 June 2016 concerning the transfer of suspected pirates from the EUNAVFOR Atalanta operation to the Republic of Tanzania. In this case, the European Parliament called for the annulment of Decision 2014/198/CFSP as it contested the legal basis on which it was taken.41 The Council had based its Decision on Article 37 TEU, which did not require the consent of the Parliament, but the Parliament claimed that Article 218(10) TFEU should have provided the legal basis for the EU–Tanzania Agreement (signed 1 April 2014). In making this case, the European Parliament was essentially arguing that the transfer of suspected pirates was as much an issue of police and judicial cooperation as it was one of military affairs (enshrined in the TEU). This symmetry is to be found back in the new ‘non-affect’ clause of Article 40 TEU, which states, on the one hand, that the implementation of the CFSP is not to affect the procedures and institutional powers established for the exercise of TFEU-based competences; on the other hand, nor is the implementation of those policies to affect the procedures and the institutional balance of power within the CFSP/CSDP. As noted by Cremona in Chapter 1 in this volume, ‘the Court did not … espouse this distinction between internal and international security, but preferred instead to link the objectives of the agreement at issue with those of the CSDP mission which it was designed to facilitate; the CSDP character of the agreement was thus made clear’.42 While these concerns are recognized by EU policy planners, this has not stopped EU efforts to find an adequate balance between security and development.43 As the High Representative/Vice-President (HR/VP) recognizes, the EU’s comprehensive approach might be tested when training provided by the CSDP is ‘hampered by a lack of basic partner country equipment’.44 For example, even though the EUTMs in Mali and Somalia were funded through Member States’ contributions and the Athena mechanism, policy planners soon discovered that the lack of communication systems, 40 S Blockmans and M Spernbauer, ‘Legal Obstacles to Comprehensive EU External Security Action’ (2013) 18 European Foreign Affairs Review 7; and R van Ooik, ‘Cross-Pillar Litigation Before the ECJ: Demarcation of Community and Union Competences’ (2008) 4 European Constitutional Law Review 399. 41 P Koutrakos, EU International Relations Law (2nd edn, Bloomsbury 2015) 538. 42 See Chapters 1 and 4 in this volume. 43 S Keukeleire and K Raube, ‘The Security-Development Nexus and Securitization in the EU’s Policies Towards Developing Countries’ (2013) 26 Cambridge Review of International Affairs 556. 44 Commission/EEAS (n 32) 2.

Military CSDP operations 123 protective equipment, medical equipment and accommodation stymied the ability of Malian and Somali forces to become operationally autonomous. In essence, the lack of capacity in Mali and Somali would hamper a more comprehensive approach for these two countries and the regions they inhabit as a whole. This is of particular concern where research shows that small and under-resourced ‘train and equip’ missions have a tendency towards small security pay-offs.45 For example, it was recognized that the Somali National Army (SNA) were finding it difficult to access key locations in Mogadishu and surrounding regions because it lacked the equipment needed to do so.46 Recognizing these challenges, the European Commission proposed an amendment to the IcSP Regulation in 2014 (see No. 230/2014) to allow for exceptional funding of third-country security and military services. However, while the resulting proposal (see COM(2016) 447 final) did foresee the funding of infrastructure and non-lethal equipment (i.e. communications hardware and medical support) it did not propose any amendments to allow the EU to provide lethal equipment to third-country services during CSDP operations.47 For the time being, therefore, EUTMs will continue to face financial pressures when it relates to equipping third-country forces, although the recently proposed European Peace Facility (EPF) could ease these financial pressures. The EPF would be a fund worth €10.5 billion that would exist outside of the EU budget by bringing together contributions under the Athena mechanism and the APF. The proposed EPF is designed to financially support EU partners with their military training capacities, among other things. Despite the well-founded reservations about ‘train and equip’, certain CSDP military operations face substantial operational constraints because training is only as useful as the equipment (both lethal and non-lethal) armed forces can use. Therefore, somewhere in the ‘security–development nexus’ there are EU military trainers who are currently facing severe tactical and operational restrictions. Calibrating the security–development nexus along EU lines is a particular challenge associated with ensuring that the ‘comprehensive approach’ works effectively. This challenge will likely become even more relevant as the EU builds on the ‘comprehensive approach’ with a new security concept that can be found in the EU Global Strategy: the ‘integrated approach’. Indeed, whereas the comprehensive approach focuses on the tools required to respond to crisis in all of its forms (development, economic, diplomatic and military), the integrated approach expands this by focusing not simply on a crisis but on ‘all stages of the conflict cycle, acting promptly on prevention, responding responsibly and decisively to crises, investing in stabilization, and avoiding premature disengagement when a new crisis erupts’.48 The EU also wants to use the integrated approach as a way to respond to conflicts at the local, national, 45 S Biddle, J Macdonald and R Baker, ‘Small Footprint, Small Payoff: The Military Effectiveness of Security Force Assistance’, Journal of Strategic Studies (early online version) accessed 4 May 2017. 46 European Scrutiny Committee, Capacity Building in the Horn of Africa (2015–16, HC 25)

accessed 16 April 2017. 47 Commission/EEAS (n 32) 4. 48 European External Action Service, ‘A Global Strategy for the European Union’s Foreign and Security Policy’ (2016) accessed 7 May 2017.

124 Research handbook on the EU’s common foreign and security policy regional and global levels. The integrated approach is therefore much more ambitious in scope than the comprehensive approach, but either way the EU will still have to iron out many of the existing creases that affect the EU’s ability to bring together development and security tools effectively.

4. THE QUEST FOR OPERATIONAL EFFECTIVENESS 4.1 Institutional Responses to Global Change Another relevant aspect of the EU’s military operations relates to the shifting context of the EU’s institutional make-up and broader geopolitical shifts. It is possible to point to a number of instances where shifts in the global geopolitical context have directly fed into institutional changes at the EU level. One of the recent examples that should be borne in mind is the establishment of the EEAS. Of course, the EEAS was born of the institutional changes that were deemed necessary in the first decade of the 2000s, which culminated in the Treaty of Lisbon. Along with ‘double hatting’ the High Representative as one of the European Commission’s Vice-Presidents, Article 26(2) TEU made clear that by establishing the EEAS the aim was to ensure unity, consistency and effectiveness of the EU’s approach to external action. The quest for a more coherent and effective EU in the field of foreign, security and defence had long been a stated ambition of EU Member States. The TEU explicitly recognizes that ‘Member States shall support the Union’s external and security policy actively and unreservedly in a spirit of loyalty and mutual solidarity’ (see Article 11(3) TEU). Yet coherence and effectiveness were not just about institutional streamlining by bridging the Council of the EU and European Commission’s competencies and tools. Indeed, during the establishment of the EEAS in 2010 a number of international crises emerged to drive home the necessity for more effective EU action. Not only did the global financial crisis of 2008 mean that resources for external action and security and defence were increasingly strained (e.g. defence spending in 27 EU Member States49 fell by approximately €6 billion from 2008 to 201050), but events occurring throughout the establishment of the EEAS (2010–2011), such as the 12 January 2010 earthquake in Haiti (resulting in over 316,000 deaths), the 18 February 2010 coup in Niger and the beginning of the ‘Arab Spring’ in Tunisia in December 2010, seemed to confirm the need for more credible crisis response mechanisms at the EU level. At a time when the EU appeared to be developing the types of institutions required to deal with security crises in its neighbourhood and further afield, the momentous crises that erupted through 2010 and 2011 meant that the EU needed more than just institutional

49

This is 27 rather than 28 because Denmark has an opt-out from CSDP and so its defence spending is not included in European Defence Agency figures. 50 European Defence Agency, ‘Defence Data 2011’ accessed 6 July 2017.

Military CSDP operations 125 streamlining. Indeed, the EU had to mobilize its institutions to craft new responses to crises in the southern and eastern neighbourhoods.51 More specifically on military operations, the EU did attempt to use the transition towards the EEAS to improve its military crisis response mechanisms. Despite being accused of serious divisions preceding the 2011 Libya intervention,52 the EU continued to adapt its institutional structures to support the emerging logic of the ‘comprehensive approach’. This approach, which was defined in clearer terms in a 2013 Joint Communication by the EEAS and European Commission, was supposed to lead ‘to the joined-up deployment of EU instruments and resources’.53 This ethos had already featured during the establishment of the EEAS. For example, the EU Military Staff – which was initially established under the Cologne European Council (1999) as a body within the Council General Secretariat – was moved to the EEAS. The logic here was that the HR/VP would not be able to conduct their duties without having at their disposal all the relevant crisis management institutions.54 Furthermore, by moving the EUMS within the EEAS it would be possible for the EU to focus on the closer civil–military coordination eventually called for as part of the EU’s comprehensive approach to crises.55 A more recent example of where institutional responses emerge out of strategic change can be found in the aftermath of the UK’s decision to leave the EU in 2016. Indeed, the UK’s decision has spurred a number of policy initiatives designed to improve the effectiveness of EU military operations. Once-dormant ideas, such as having a single ‘EU Operation Headquarters’ (EU OHQ),56 re-emerged following the UK referendum result in June 2016. As outlined in the Council conclusions of 14 November 2016, the EU28 have decided to create a ‘Military Planning and Conduct Capability’ (MPCC). The MPCC is being created within the EEAS with a view to providing a command and control facility for all non-executive CSDP operations – rather than relying on individual operation commanders for each operation, one commander in the form of the Director General of the EUMS will steer the operations. The British had for many years effectively blocked the idea of an EU OHQ on the understanding that the establishment of an operational headquarters would somehow duplicate NATO’s SHAPE.57 Intriguingly, however, despite (or indeed because of) the 51

R Whitman and AE Juncos, ‘The Arab Spring, the Eurozone Crisis and the Neighbourhood: A Region in Flux’ (2012) 50 Journal of Common Market Studies 147. 52 N Koenig, ‘The EU and the Libyan Crisis: In Quest of Coherence?’ (2011) 46 The International Spectator: Italian Journal of International Affairs 11; and A Menon, ‘European Defence Policy from Lisbon to Libya’ (2011) 53 Survival: Global Politics and Strategy 75. 53 Commission/EEAS, ‘The EU’s Comprehensive Approach to External Conflict and Crises’ (Joint Communication) JOIN (2013) 30 final 3; and ME Smith, ‘Institutionalising the “Comprehensive Approach” to EU Security’ (2013) 18 European Foreign Affairs Review 25. 54 L Erkelens and S Blockmans, Setting up the European External Action Service: An Institutional Act of Balance (CLEER Working Paper No. 1, TMC Asser 2012) 16. 55 N Hynek, ‘EU Crisis Management after the Lisbon Treaty: Civil-Military Coordination and the Future of the EU OHQ’ (2011) 20 European Security 81. 56 ibid. 57 S Biscop, ‘The UK and European Defence: Leading or Leaving?’ (2012) 88 International Affairs 1297.

126 Research handbook on the EU’s common foreign and security policy UK’s impending departure from the EU, it did not block the creation of the MPCC in the Council of the EU, even though it had some reservations about some of the wording surrounding the MPCC. London’s post-referendum messaging on security and defence relations with the EU has centred on a constructive and close relationship, and it does not appear that the UK wanted to spoil any chances of building healthy security and defence arrangements with the EU by blocking the MPCC.58 4.2 Political Responses to Global Change Even though the EUMS and EEAS have had the time to develop new working practices since 2011, the interplay between geopolitical changes and institutional responses remains a key factor in the evolving nature of the EU’s military response to crises. One of the more recent examples that has had an effect on European defence cooperation was the decision by France to invoke Article 42(7) TEU (otherwise known as the ‘mutual assistance clause’) on 17 November 2015. France’s decision was taken in the wake of the Paris terrorist attacks on 13 November 2015. Article 42(7) stipulates that EU Member States shall have an obligation of aid and assistance to a Member State that is the victim of armed aggression on its territory. While this specific article of the Treaties is vague in what precisely is meant by ‘armed aggression’ or ‘aid and assistance’, the invocation of the Treaty took on a symbolic meaning for European solidarity – even though it would not technically involve the EU institutions or lead to a decision to mount a military CSDP operation (especially not on the territory of the EU).59 It had practical implications too. Indeed, the severity of the Paris attacks and France’s need to respond to homeland security ‘meant that EU Member States could simply not ignore France’s requests’.60 As France would have to concentrate even more on homeland security by enhancing Opération Sentinelle, and as this would mean allocating more troops for deployment in France, the required EU solidarity from other Member States would come in the form of strengthening CSDP operations in sub-Saharan Africa61 and/or coalition operations in Syria and Iraq.62 Germany responded to the mutual assistance clause by looking into increasing its deployments in places such as Mali. Indeed, although the Bundeswehr has been present in Mali since 2013 under the UN MINUSMA mission, the aftermath 58

J Barigazzi, ‘Boris Johnson Plays Down Spat Over EU Military Unit’ Politico Europe (15 May 2017) accessed 6 June 2017. 59 C Hillion and S Blockmans, ‘Europe’s Self-Defence: Tous Pour Un et Un Pour Tous?’ (CEPS Commentary, 20 November 2015) 2 accessed 2 May 2017. 60 T Tardy, ‘Mutual Defence – One Month On’ (2015) (EU-ISS Policy Alert, 17 December 2015) accessed 16 April 2017. 61 ibid. 62 Hillion and Blockmans (n 59) 3; and A Bakker, ‘Defending Europe: Translating Mutual Assistance into Action’ (Clingendael Policy Brief, January 2017) accessed 19 April 2017.

Military CSDP operations 127 of the Paris attacks saw Berlin move to increase its troop, transport and Tiger and NH90 helicopter contingents in Mali.63 Following France’s invocation of Article 42(7), Germany also decided to deploy additional forces in the fight against Islamic State in Syria. In December 2015 Germany ‘decided that up to 1,200 Bundeswehr soldiers would take part in the international coalition’.64 Other countries would also respond to France’s call for assistance65 but the fact that Germany responded is of particular importance given how crucial France and Germany are to European defence cooperation. Significantly, while EU Member State responses to France’s invocation of Article 42(7) TEU were not strictly of relevance to the CSDP, they nonetheless flagged the important dynamic that exists not only between internal and external security challenges, but also between the EU institutions and Member States. More specifically, despite the fact that Article 42(7) TEU presupposes intergovernmental cooperation between EU Member States, it would become evident that any longer-term response to the so-called internal–external security nexus would require the involvement of supranational institutions at some point. As some commentators have observed, counter-terrorism and radicalization strategies, migration, and hybrid threats such as cyber security cannot be dealt with on an entirely intergovernmental basis.66 In fact, the security and defence implementation plan that followed on from the EU Global Strategy recognized this by remarking that if European states want to protect Europe, then attention needs to be focused on those security ‘grey areas’ such as border protection, critical infrastructure and disaster response that require both a defence and a security dimension.67 Thus, while the invocation of the mutual assistance clause had little direct bearing on CSDP, it did force the EU institutions to think rather more seriously about how the EU can support Member States in security and defence. Another instance where geopolitical realities have affected the nature of CSDP military operations relates to the security and defence implementation plan that was agreed in 2016 and which follows on from the EU Global Strategy. Indeed, on 14 November 2016 the Council of the EU elaborated a new level of ambition for the CSDP that is supposed to build on the crisis management tasks laid down in Articles 42(1) and 43(1) TEU (known as the ‘Petersberg Tasks’). The ‘Petersberg Tasks’ stated that the EU’s military level of ambition should include tasks such as humanitarian and 63

German Federal Government, ‘More Soldiers for Mali’ accessed 14 May 2017. 64 J Gotkowska and K Frymark, ‘Germany’s Engagement in the Resolution of the Syrian Conflict’ (OSW Commentary, 25 January 2016) accessed 20 April 2017. 65 For an update on responses to France’s invocation of Article 42(7) read European Parliament, ‘Activation of Article 42(7) TEU: France’s Request for Assistance and Member States’ Responses’ (Briefing, July 2016) accessed 6 June 2017. 66 S Biscop, ‘The European Union and Mutual Assistance: More than Defence’ (2016) 51 The International Spectator: Italian Journal of International Affairs 119. 67 Council of the EU (n 29) 5; and N Tocci, ‘The Making of the EU Global Strategy’ (2016) 37 Contemporary Security Policy 461.

128 Research handbook on the EU’s common foreign and security policy rescue tasks, peacemaking, disarmament, military advice and post-conflict stabilization. The Council conclusions of 14 November 2016, however, build considerably on these tasks. If, for example, one reads the ‘Annex to the Annex’ to the conclusions, it is possible to learn of the EU’s newly found desire to conduct air and special operations, close air support and air surveillance, response to hybrid threats and maritime security or surveillance operations.68 This last task is particularly interesting given the conclusions’ reference to the need to ensure ‘stable access to and use of the global commons, including the high seas and space’.69 It is no coincidence that such ideas have emerged. After all, one of France’s former defence ministers recently asked whether we could ‘not have European navies coordinate in order to ensure a regular and as visible a presence as possible in the maritime areas of Asia?’.70 Furthermore, much greater attention needs to be paid to the shifting operational context in which CSDP military operations will be deployed in the future. For example, the issue of hybrid threats is particularly important here – and not just because the EU sees itself as an ideal organization to deal with such threats (an assessment shared by NATO, incidentally71). Indeed, capabilities such as cyber defence and autonomous weapons systems are already being used by actual and potential adversaries in the field of operations. Here, one should not just think about the hybrid tactics employed by Russia in Crimea in 2014. There is, for example, now evidence to suggest that even in asymmetric situations, groups such as Islamic State and Boko Haram use unmanned aerial systems72 and/or cyber capabilities.73 It will take some time before the EU can fully conceptually and materially accommodate hybrid threats when it deploys CSDP military operations (the EU is not alone here for even the US military is adjusting to the hybrid paradigm74). Nevertheless, the EU has begun to adjust to the changing strategic environment through bodies such as the European Defence Agency (EDA), which is studying and organizing table-top exercises for hybrid operational scenarios.75 A key future opportunity to really prove the EU’s added value for tackling hybrid threats will emerge in 2018, when the EDA will revise the Capability Development Plan and therefore reconfigure the EU’s doctrine and capability requirements for CSDP military operations. 68

Council of the EU (n 29) 15. ibid, 5. 70 French Ministry of Defence, ‘Jean-Yves Le Drian in Singapore for the 15th Shangri-la Dialogue’, speech by the Minister of Defence (5 June 2016) accessed 14 April 2017. 71 P Pindjak, ‘Deterring Hybrid Warfare: A Chance for NATO and the EU to Work Together?’ (2014) NATO Review accessed 7 July 2017. 72 M Gault, ‘Islamic State’s Drone Fleet’ Reuters (31 March 2017) accessed 7 July 2017. 73 T Aladenusi, ‘Cyberharam: Can Nigeria Prepare for the Next Generation of Terrorists?’ (Deloitte June 2015) accessed 7 July 2017. 74 FG Hoffman, ‘Hybrid Warfare and Challenges’ (2009) 52 Small Wars Journal 34. 75 European Defence Agency, ‘Hybrid Warfare’ (16 July 2015) accessed 7 July 2017. 69

Military CSDP operations 129 Of course, the perennial problem for the EU when it conceives of new strategic visions is capabilities (see Chapter 8 in this volume). It is all very well to talk about a potential EU maritime role in the Asia-Pacific but this requires a serious commitment to pooling existing, and (even) developing new, defence capabilities. Although many European states have increased their defence spending in light of the multiple threats on Europe’s borders,76 European forces are increasingly stretched because of responsibilities related to NATO deterrence in Eastern Europe and coalition operations in places such as Iraq and Syria. In a number of countries the armed forces are also undertaking homeland security tasks. CSDP military operations are therefore just one type of tasking that can be handed to forces in individual European countries. European states have but a single set of forces after all, although any cursory glance at the footprint of CSDP operations since the policy’s operational inception in 2003 reveals that there is an emphasis on civilian rather than military operations. This is, of course, not to make the argument that more military operations are required to justify CSDP – that would be absurd – but rather that there is an intimate relationship between strategic objectives and expectations and capability development. Recent initiatives such as the European Defence Fund (EDIF)77 and the Co-ordinated Annual Review on Defence (CARD)78 may in time help with European military capability development, but there remains some way to go before the EU turns what the EU Global Strategy sees as a need to move from a shared vision to common action. In this respect, it will be interesting to see what role Permanent Structured Cooperation (PESCO) could play in enhancing the EU’s military effectiveness. Following the European Council’s launch of PESCO in December 2017,79 this is no longer just a theoretical question. Indeed, in moving forward with PESCO there will be a debate about what such cooperation should be for. One of the perennial issues confronting PESCO has been to agree on what (capability and industrial) projects or (operational) areas to focus on. All one can be certain of is that PESCO is a legal framework that should lead to binding commitments on defence. While PESCO could foreseeably ameliorate some of the financial constraints of EU military operations by encouraging Member States to develop a more efficient funding line for operations,80 structured 76 J Bund, D Fiott, T Tardy and Z Stanley-Lockman, EUISS Yearbook of European Security 2017 (EU Institute for Security Studies 2017) 97. 77 On 30 November 2016 the European Commission unveiled its ‘European Defence Action Plan’. The idea was to create a European Defence Fund. The fund would be divided between defence research investment (directly supported by the EU) and joint capability development projects (based on pooled investment by willing EU Member States). 78 Contained in the Council conclusions of 14 November 2016 is the idea to have a regular – albeit voluntary – common review of defence planning, defence budgets and capability plans at the ministerial level within the European Defence Agency. The idea is to ensure greater ex ante and ex post synchronization of EU Member State defence spending and capability development plans. 79 Council Decision (CFSP) 2017/2315 of 11 December 2017 establishing permanent structured cooperation (PESCO) and determining the list of participating Member States [2017] OJ L331/57. 80 A Bakker, M Drent and D Zandee, ‘European Defence Core Groups: The Why, What and How of Permanent Structured Cooperation’ (Clingendael Policy Brief, November 2016) 2.

130 Research handbook on the EU’s common foreign and security policy cooperation cannot realistically overcome many of the politico-strategic issues discussed at the beginning of this chapter.

5. CONCLUSION This chapter began with the task of analysing the intersection of the legal and operational parameters in which military operations under the CSDP are deployed. Accordingly, this chapter has focused on three broad challenges for CSDP military operations. First, the chapter focused on the interplay between legal parameters and operational needs/resources when CSDP military operations are deployed. Here, the chapter drew on the experiences that have emerged since the deployment of Operation Sophia. It was shown how, although the operation is designed to debilitate smuggling and human trafficking networks, the strategic landscape in the Mediterranean has led to tactical constraints and to questions about capacity in dealing with a range of – non-forceful – security challenges. In this regard, legitimate questions about the strategic rationale for the operation have arisen and on this basis it has been possible to provide a critique of the ‘comprehensive approach’. Indeed, the case of Operation Sophia clearly shows that there are limitations on how far the EU can draw on lessons learned from other military operations when the politico-strategic circumstances are unique to each deployment. Second, the chapter analysed the interaction between operational and development imperatives when the EU deploys certain types of military operations in third countries. Drawing on the experiences of EUTMs to countries such as the Central African Republic, Mali and Somalia, the chapter showed how any ‘train and equip’ operation must navigate the delicate line between security capacity-building and humanitarian/ development principles. Indeed, by looking specifically at the issue of how to fund EUTM exercises, this chapter has shown how tactical necessity – paying for the equipment that forces in third countries require in order to train – meets legal and political realities, such as restrictions on the use of EU funds for military purposes. Such a dilemma has arisen because of the shifting nature of the EU’s military engagement in third countries. In recent years less emphasis has been placed on military deployments for peacekeeping or peacemaking purposes and more on security sector reform and capacity-building. Yet, the less the EU deploys for traditional military deployments, the more the EU will have to craft innovative responses to how it uses development and military tools together. Third, the chapter explained how geopolitical and strategic shifts directly weigh on the rationale and form of EU institutions involved in the deployment of CSDP military operations. The examples of France’s invocation of Article 42(7) and the UK’s decision to leave the EU have shown how institutions are responsive to major political events, especially when they have a direct or indirect bearing on the CSDP. The chapter showed that there still exists a delicate relationship between the EU Member States and EU institutions as providers of security and defence. In the case of the invocation of the mutual assistance clause, despite the fact that EU institutions were not formally included in the call to provide assistance it soon became clear that these institutions would bring added value to Member State efforts. In the case of the UK’s

Military CSDP operations 131 decision to leave the EU, once forbidden initiatives re-emerged with renewed vigour and the EU is now on course to establish a single command and control centre for some of its military operations. Time will tell how far these various initiatives could enhance CSDP military operations. What is apparent is that CSDP military operations do not occur in a political or strategic vacuum. Every military strategist will be familiar with Clausewitz’s term ‘friction’ – the term describes those contingencies that emerge in the battlefield or back home that directly affect the commanders’ ability to wage a successful campaign. In an EU context, ‘friction’ not only relates to tactical challenges such as the need to augment operational objectives or having to fend off harsh weather conditions during operations. The very fact that the EU embodies both Member States and EU institutions means the chances of new and challenging forms of friction emerging is relatively high. The future development of CSDP military operations will occur in a context in which the security landscape will arguably be more demanding and where the EU will need to continue to assess how it financially supports these operations. Of course, layered on top of these challenges are the mores and needs of individual Member States. As the cases that relate to France and the UK in this chapter indicate, CSDP military operations will continue to be largely shaped by intergovernmental dynamics.

7. Responsibility and liability for CSDP operations Joni Heliskoski*

1. INTRODUCTION: THE SCOPE, PURPOSES AND CONCEPTS This chapter addresses questions of responsibility and liability arising in the context of civilian and military crisis management operations conducted within the framework of the Common Security and Defence Policy (CSDP).1 The specific set of legal problems relating to questions of responsibility or liability for such operations stems from their nature as an instance of multi-level governance involving both the European Union and its Member States and, possibly, third states and/or international organizations. While the launch and conduct of a CSDP operation is always based on a decision taken by the EU Council pursuant to the Treaty on European Union (TEU), the operations are, to a considerable degree, carried out with resources provided by the Member States.2 In some cases, third states may also provide personnel to an operation.3 The basis for the authorization of an operation may be found in a resolution of the UN Security Council and sometimes an operation may be conducted having recourse to assets provided by NATO. Hence the specific set of legal problems relating to the topic of responsibility or liability in this context. This brief introduction demonstrates that the present topic not only covers a multiplicity of different actors, but also necessarily involves several different legal orders. At the very least, it stands, so to speak, at the tri-frontier of public international law, EU law and the domestic law of the Member States. These various legal orders, however, intertwine in the sense that a substantive rule or principle of one legal order may be recognized by or applied within another.4 Given this state of affairs, it would * The author is grateful to Frederik Naert for comments and suggestions on an earlier draft. The usual disclaimer applies. 1 See esp. F Naert, ‘The International Responsibility of the Union in the Context of Its CSDP Operations’ in P Koutrakos and M Evans (eds), The International Responsibility of the European Union (Hart 2013) and F Naert, ‘Shared Responsibility in the Framework of the EU’s CSDP Operations’ in A Nollkaemper and I Plakokefalos (eds), The Practice of Shared Responsibility in International Law, Vol. III (CUP 2017). 2 The standard legal bases for establishing an operation are Articles 42(4) and 43(2) TEU. 3 As a rule, the participation of third states in CSDP operations or missions is organized through the conclusion of an agreement between the EU and the third state concerned. See, e.g., the Agreement between the European Union and the Swiss Confederation on the participation of the Swiss Confederation in the European Union military mission to contribute to the training of the Malian Armed Forces (EUTM Mali) [2014] OJ L151/18. 4 By way of illustration, reference could be made to the case law of the Court of Justice of the European Union (CJEU) recognizing that international law is binding on the institutions as a matter of EU law. See e.g. Case C-366/10 Air Transport Association of America and Others [2011] ECLI:EU:C:2011:864, para 101.

132

Responsibility and liability for CSDP operations 133 not seem appropriate to seek to address the substantive law of each of the above systems of law in isolation from one another. Rather, it may be more appropriate to look at various different jurisdictions and to seek to map out how they have dealt with questions of responsibility or liability, whatever the origin of the substantive law they are deemed to have applied in a particular context. Accordingly, the structure of this chapter is based on the notion of jurisdiction, beginning with the public international law jurisdiction (Section 2), followed by EU law jurisdiction (Section 3) and concluding with domestic law jurisdiction of the Member States (Section 4). The conclusions from these different approaches will then be brought together and examined by way of a concluding assessment (Section 5). The following introductory remarks are needed. First, for reasons of space, no systematic account will be provided of the legal framework of CSDP operations, let alone a description of the particular features of the individual operations established in practice.5 Secondly, the issue concerning the limits of jurisdiction of the CJEU in the field of the CFSP is specifically covered by another chapter in the volume and therefore will not be addressed in depth by the present chapter.6 Finally, a terminological issue: in accordance with the standard practice, the term ‘responsibility’ is mainly used in the context of rules and principles of public international law, while, for the purposes of EU law as well as the domestic law of the Member States, reference is made to the notion of ‘liability’.

2. PUBLIC INTERNATIONAL LAW JURISDICTION 2.1 The Challenge of Defining the Substantive Law on the Attribution of Conduct Under the orthodox position of general international law, there are two standard elements of an internationally wrongful act capable of entailing the international responsibility of a state or an international organization: first, the conduct concerned must be attributable to the given state or international organization and, secondly, that conduct must constitute a breach of an international obligation of that state or international organization.7 In the context of the EU, including the conduct of the Union’s CSDP operations, it is the question of attribution of conduct to either the EU or one or more of its Member States that has become the centre of attention of the academic scholarship on the matter. In particular, the discussion has focused on 5 See e.g. P Koutrakos, The EU Common Security and Defence Policy (OUP 2013) 64–68 and chapters 5 and 6. See also https://eeas.europa.eu/headquarters/headquarters-homepage/area/ security-and-defence_en. 6 See Chapter 4 in this volume. 7 See Draft Articles on Responsibility of States for Internationally Wrongful Acts, with commentaries, adopted by the International Law Commission at its 53rd session in 2001 and submitted to the General Assembly (UN Doc. A/56/10) and Draft Articles on the Responsibility of International Organizations, with commentaries, adopted by the International Law Commission at its 63rd session in 2011 and submitted to the General Assembly (UN Doc. A/66/10) (hereafter ‘the DARIO’).

134 Research handbook on the EU’s common foreign and security policy whether the test of ‘effective control’ exercised by an international organization over the conduct of an organ of state placed at the disposal of that organization8 adequately reflects the specificities of the EU in that regard, or whether CSDP operations should rather be conceived of as de jure or de facto organs of the EU.9 In the view of this author no conclusive appraisal of the above question may be given by way of a deduction from general principles of public international law as described, most notably, in the DARIO or the case law of the European Court of Human Rights.10 Indeed, Draft Article 64 of the DARIO (‘Lex specialis’) recognizes that the Draft Articles do not apply where and to the extent that the conditions for the existence of an internationally wrongful act are governed by ‘special rules of international law’, the commentary drawing specific attention to the possibility of the situation of the EU being subject to such ‘special rule’.11 Moreover, as there exists, for the time being, virtually no international dispute settlement practice regarding international responsibility of the EU and its Member States in the context of conduct of CSDP operations,12 one could argue until the Greek calends about the possible features of an appropriate rule of attribution applicable to the Union and its Member States. Therefore, as any attempt to approach the matter on the basis of general principles of public international law would necessarily remain speculative, there appears to be no good reason for taking that debate any further at this stage. At most, one could make some observations on the state of the public international law aspects of responsibility on the basis of the more recent case law of the CJEU concerning the question of attribution under EU law. On the assumption that this case law qualifies as being indicative of the ‘rules of the organization’ envisaged in the DARIO, one could possibly attempt to proceed with the definition of an international law rule of attribution applicable to CSDP operations or missions. This would, however, need to wait until the concluding observations and assessment of the chapter. The realm of public international law governing the international responsibility of the EU and its Member States is not exhausted by the above considerations. At the level of public international law, the EU has entered into contractual commitments aimed at governing, inter alia, the question of third-party claims arising in the context of CSDP operations or missions in agreements concluded between the EU and a given host state on the status of forces (hereafter ‘SOFAs’) or mission (‘SOMAs’). While the question 8

Established by Draft Art. 7 of the DARIO as the central criterion for the conduct concerned being attributable to the organization in question. 9 See Draft Art. 6 of the DARIO and, for scholarly discussion of the issue, esp. A Sari and RA Wessel, ‘International Responsibility for EU Military Operations: Finding the EU’s Place in the Global Accountability Regime’ in B Van Vooren, S Blockmans and J Wouters (eds), The Legal Dimension of Global Governance: What Role for the EU? (OUP 2013) and PJ Kuijper and E Paasivirta, ‘EU International Responsibility and its Attribution: From the Inside Looking Out’ in P Koutrakos and M Evans (eds), The International Responsibility of the European Union (Hart 2013). 10 See Behrami v France and Saramati v France, Germany and Norway (2007) 45 EHRR SE10; Al-Jedda v United Kingdom (2011) 53 EHRR 23. 11 See para 2 of the commentary to Draft Art. 64. 12 The absence of such practice is noted most recently in Naert, ‘Shared Responsibility in the Framework of the EU’s CSDP Operations’ (n 1) 689.

Responsibility and liability for CSDP operations 135 of substantive law governing the question of responsibility may have to be put aside, at least for the time being, for the reasons set out above, an account of the relevant provisions of the SOFAs and SOMAs could nonetheless contribute to a more realistic understanding of the mechanisms through which claims arising from the conduct of CSDP operations may be resolved in practice.13 2.2 The Regime of Responsibility or Liability under EU-SOFAs and EU-SOMAs The purpose of SOFAs and SOMAs is to define the legal position of military forces or civilian personnel deployed by one or more states or by an international organization in the territory of another state with the consent of the latter. In respect of most CSDP operations and missions, the EU has either concluded SOFAs or SOMAs with the third states concerned or extended pre-existing status arrangements to cover EU operations.14 While no comprehensive account of the practice may be provided in the present context,15 the aspects of the SOFAs and SOMAs aimed at governing the question of international responsibility of the EU and its Member States clearly merit attention in the present context. In this respect, the SOFAs and SOMAs operate essentially through the following logic: on the one hand, they exempt the personnel of an operation or a mission from the jurisdiction of the host state through privileges and immunities granted by the latter and, on the other hand, they create alternative procedures by means of which private claimants may seek compensation for acts or omissions attributable to the operation or mission. While there is some diversity in the arrangements in place for various CSDP operations or missions, the principal features of the arrangements governing the question of responsibility or liability may be described through the generic model agreements (hereafter referred to as ‘Model SOFA’ and ‘Model SOMA’, respectively) which nowadays constitute the basis for negotiating SOFAs and SOMAs for individual crisis management operations.16 In respect of the immunity of the personnel from local jurisdiction, there is a distinction between criminal jurisdiction on the one hand and civil and administrative jurisdiction on the other. While the former applies ‘under all circumstances’,17 the latter only applies to ‘words spoken or written and all acts performed … in the exercise of their official functions’, together with a procedure for determining whether or not a given act had been performed in the exercise of the official functions of the 13 A similar approach is followed in respect of settlement of claims relating to military operations in D Fleck (ed.), The Handbook of the Law of Visiting Forces (OUP 2001) 159–186. 14 In the case of participation of third states in CSDP operations or missions the SOFAs and SOMAs are made applicable to the personnel contributed to a given operation or mission by a third state by virtue of an agreement concluded between the EU and the third state concerned. See e.g. Art. 2(1) of the Agreement between the European Union and the Swiss Confederation on the participation of the Swiss Confederation in EUTM Mali (n 3). 15 For a treatment of the practice up to 2007, see A Sari, ‘Status of Forces and Status of Mission Agreements under the ESDP: The EU’s Evolving Practice’ (2008) European Journal of International Law 67. 16 For the EU Model SOFA, see Council document 11894/07 (20 July 2007) and the EU Model SOMA, Council document 17141/08 (15 December 2008). 17 See Art. 6(3) of the EU Model SOFA and Art. 6(4) of the EU Model SOMA.

136 Research handbook on the EU’s common foreign and security policy personnel.18 Regarding the arrangement concerning the treatment of claims, both the Model SOFA and the Model SOMA exclude the liability of the personnel for ‘any damage to or loss of civilian or government property which is related to operational necessities or caused by activities in connection with civil disturbances or the protection of the [EU force or mission]’.19 For other claims for damage to, or loss of, civilian or government property, and for any claims for death of, or injury to, persons as well as claims for damage to, or loss of property of the EU force or mission, the Model SOFA and the Model SOMA put in place a three-step settlement procedure as follows. In the first instance, these claims are forwarded to the EU force or mission via the competent authorities of the host state, with regard to claims brought by a legal or natural person from the host state, and to the competent authorities of the host state, with regard to claims brought by the EU force or mission. Where no amicable settlement can be found, the claim shall be submitted to a claims commission composed on an equal basis of representatives of the EU force or mission and representatives of the host state. Settlement of claims shall be reached by common agreement. Finally, where no settlement can be reached within the claims commission, the dispute shall be settled by diplomatic means between the host state and EU representatives for claims up to and including EUR 40,000. For claims exceeding that amount, the dispute shall be submitted to an arbitration tribunal,20 whose decisions shall be binding. The primary objective of the above provisions of the SOFAs and SOMAs is to bring about an amicable settlement of claims through the internal mechanisms of the CSDP operation.21 In the light of the existing practice, it appears that this objective has been largely achieved: virtually all claims lodged under SOFAs and SOMAs in the context of CSDP operations are resolved by means of amicable settlement.22 18

Art. 6(4) of the EU Model SOFA and Art. 6(5) of the EU Model SOMA. Art. 15(1) of the EU Model SOFA and Art. 16(1) of the EU Model SOMA. 20 The arbitration tribunal shall be composed of three arbitrators, one being appointed by the host state, one being appointed by the EU force or mission, and the third being appointed jointly by the host state and the EU force or mission. Where one of the parties does not appoint an arbitrator within two months or where no agreement can be found between the host state and the EU force or mission on the appointment of the third arbitrator, the arbitrator in question shall be appointed by the President of the CJEU. The EU force or mission and administrative authorities of the host state shall conclude an administrative arrangement in order to determine the terms of reference of the claims commission and the arbitration tribunal, the procedure applicable within these bodies, and the conditions under which claims are to be lodged. 21 Sari (n 15) 95. 22 Email of 27 April 2016 from E Chaboureau, Legal Adviser, Legal Affairs Division, European External Action Service. Naert, ‘Shared Responsibility in the Framework of the EU’s CSDP Operations’, above n 1, reports that ‘no claims commission or arbitral tribunal has actually been set up, with the possible exception of Althea and two arbitrations in relation to staff in a civilian mission’. According to the same author, ‘most claims have been settled amicably [and] … rather few claims have been brought at all’ (689, footnotes omitted). He notes that Althea (EU military operation in Bosnia and Herzegovina, established by Council Joint Action 2004/570/CFSP [2004] OJ L252/10), applies the SOFA that applied to NATO’s Implementation Force (IFOR) and Stabilisation Force (SFOR) operations. 19

Responsibility and liability for CSDP operations 137 In the rather unlikely event that no amicable settlement of a claim is achieved, which may then give rise to the question as to whom – the operation or mission concerned, the ‘European Union’ or a Member State – a given conduct (and, as a corollary, the international responsibility) is to be attributed, considerations of general public international law could be supplemented by the following observations that are more specific to the situation of the EU and its Member States. Notwithstanding the fact that individual CSDP operations or missions do not possess international legal personality of their own,23 the above standard provisions of the Model SOFA and Model SOMA on claims are designed in a way that identifies the given ‘EU force’ or ‘EU mission’ – that is, an entity the legal basis of which is the TEU – as the ‘party’ to the procedure of dispute settlement. In other words, there are no cases where a ‘Member State’ (or ‘Member States’) would play any role in the establishment of responsibility or liability under a SOFA or a SOMA. Indeed, all of the status agreements relating to CSDP operations or missions have been concluded in the name of the ‘European Union’ without the participation of the Member States in their individual capacity. Therefore, unless substantial grounds to the contrary could be identified from the corpus of general international law of responsibility, the design of the SOFAs and SOMAs would appear to warrant, if not a conclusion, a presumption that responsibility under public international law for the conduct of CSDP operation or missions is borne by the EU in its own right rather than by the Member States.

3. EUROPEAN UNION JURISDICTION 3.1 On the Concepts of Attribution, Representation and Jurisdiction As regards EU law, there are two principal branches governing the question of liability: first, the contractual and non-contractual liability of the Union under the first and second paragraphs of Article 340 TFEU; and secondly, the liability of a Member State for breaches of EU law under the case law of the Court in Francovich24 and Brasserie du Pêcheur/Factortame,25 as further defined in subsequent cases.26 As far as the discharge of liability is concerned, the above two branches fall, in principle, within two distinct realms of jurisdiction. Under Articles 268 and 272 TFEU it is the CJEU that has jurisdiction in disputes relating to compensation for damage arising under the contractual and non-contractual liability of the Union, as provided for in the first and second paragraphs of Article 340 TFEU.27 Secondly, as regards the liability of a 23 It should, however, be noted that civilian CSDP missions now have a legal capacity in EU law. See further Section 3.3 below. 24 Joined Cases C-6/90 and C-9/90 Francovich and Bonifaci v Italy ECLI:EU:C:1991:42. 25 Joined Cases C-46/93 and C-48/93 Brasserie du Pêcheur v Bundesrepublik Deutschland and The Queen/Secretary of State for Transport, ex parte Factortame and Others ECLI:EU: C:1996:79. 26 See P Aalto, Public Liability in EU Law – Brasserie, Bergaderm and Beyond (Hart 2011). 27 In light of Art. 274 TFEU it cannot be ruled out that proceedings might be brought against the EU or its bodies, offices or agencies before domestic courts of a Member State.

138 Research handbook on the EU’s common foreign and security policy Member State for breaches of EU law, the relevant remedies exist before the domestic courts of the Member State concerned.28 This section will address the first branch while the second one, together with the question of the liability of a Member State under domestic law, will be dealt with in Section 4 below. Before embarking on a more detailed analysis of the liability of the EU under Article 340 TFEU, a few important conceptual issues will first have to be clarified. To begin with, it follows both from the combined reading of Articles 268, 272 and 340 TFEU and from the Court’s case law that the Court has no jurisdiction to rule on liability arising from the unlawfulness of conduct by a Member State; therefore, a basic precondition for the jurisdiction of the Court to rule on the liability of the Union is the attribution of the conduct concerned to the EU.29 Secondly, once the relevant conduct has adequately been attributed to the Union, it needs to be established whether the action for damages has been brought against the correct defendant.30 Finally, it will have to be ascertained that the Court’s jurisdiction to examine the action is not affected by the limitations of its jurisdiction set out in paragraph 1 of Article 24 TEU and the first paragraph of Article 275 TFEU. While a more general treatment of each of the above issues is beyond the scope of this chapter, they will be addressed in light of the recent case law specifically concerning the liability of the Union for CSDP operations.31 3.2 The Attribution of Conduct to the EU As already noted, the CJEU has no jurisdiction to rule on an action concerning liability arising from the unlawfulness of conduct by a Member State. Therefore, the precondition for the establishment of liability of the EU under Article 340 TFEU is the attribution of the given conduct to the Union. In that regard, a distinction should be drawn between, on the one hand, contractual liability of the Union governed by Article 272 and the first paragraph of Article 340 TFEU and, on the other hand, the non-contractual liability falling within the scope of Article 268 and the second paragraph of Article 340 TFEU. As far as the contractual liability of the Union is concerned – that is, liability arising out of a contract entered into ‘by or on behalf of the Union’ – it suffices to point out that under Article 272 TFEU the jurisdiction of the Court may only be based upon an arbitration clause contained in such a contract. In the absence of an arbitration clause, 28

Unless jurisdiction of the CJEU is excluded by virtue of Art. 24(1) second subparagraph TEU and Art. 275(1) TFEU, the domestic courts of Member States exercise their jurisdiction subject to their obligations deriving from Art. 267 TFEU. See Case C-72/15 Rosneft ECLI: EU:C:2017:236 and further Chapter 4 in this volume. 29 See e.g. Case C-72/90 Asia Motor France v Commission ECLI:EU:C:1990:230, para 14 and Case T-277/97 Ismeri Europa v Court of Auditors ECLI:EU:T:1999:124, para 49. 30 See e.g. Joined Cases C-63-69/72 Werhahn Hansamuehle and Others v Council ECLI: EU:C:1973:121, para 7. 31 It must be noted, for the sake of clarity, that this chapter does not examine heads of jurisdiction other than Arts 268 and 272 TFEU and, in respect of the question of jurisdiction of the CJEU in the context of proceedings brought before domestic courts of Member States, Art. 267 TFEU. In particular, issues specific to actions for annulment (Art. 263 TFEU) are excluded.

Responsibility and liability for CSDP operations 139 the Court cannot adjudicate on what in reality is an action for the performance of a contract entered into by the Union as to do so would be to extend its jurisdiction beyond the limits imposed by Article 274 TFEU – a provision specifically giving national courts or tribunals of the Member States general jurisdiction over disputes to which the Union is a party.32 In the context of civilian CSDP missions, an arbitral clause granting the Court jurisdiction on the basis of Article 272 TFEU may be contained, for instance, in an employment contract concluded between a mission and a member of its personnel employed on a contractual basis.33 In the case of operations with military or defence implications – the operating expenditure of which are as a rule charged to the budgets of Member States34 – contractual liability of the Union based on Article 272 TFEU and on the first paragraph of Article 340 TFEU would, however, appear to be excluded by virtue of the relevant provisions of Council Decision (2015/528/CFSP) of 27 March 2015 establishing a mechanism to administer the financing of the common costs of EU operations having military or defence implications (Athena) and repealing Decision 2011/871/CFSP (hereafter the ‘Athena Decision’).35 According to the Decision, contracts of employment (relating, for example, to staff employed on a contractual basis) are entered into by the operation commander ‘on behalf of Athena’36 and the contractual liability is ‘covered through Athena by the contributing States or third parties’.37 Hence, contractual liability of the EU governed by Articles 272 and 340 TFEU would not seem to arise. As regards, secondly, the non-contractual liability of the Union based on the second paragraph of Article 340 TFEU, the question of attribution of a given conduct to either the Union or one its Member States (or, indeed, a third state participating in an operation or a mission) may prove more problematic. This may be illustrated through H v Commission and Council – an action, first, for annulment of a decision signed by the Chief of Personnel of the European Union Police Mission in Bosnia and Herzegovina (EUPM) and, if needed, of a decision signed by the Head of Mission38 32 See e.g. Case T-186/96 Mutual Aid Administration Services v Commission ECLI:EU: T:1997:149, paras 46–47. 33 See the Order of the General Court of 9 November 2016 in Case T-602/15 Jenkinson v Council and Others ECLI:EU:T:2016:660, para 38. Depending on the contract in question, the situation of a person employed by a mission on a contractual basis may also be governed by the national law of a Member State and fall within the jurisdiction of national courts of the Member State concerned. See e.g. Case T-410/13 Burim Bitigi and Others v Commission ECLI:EU: T:2014:871. 34 According to Art. 41(2), first subparagraph TEU, ‘[o]perating expenditure to which the implementation of this Chapter [on the CFPS] gives rise shall also be charged to the Union budget, except for such expenditure arising from operations having military or defence implications and cases where the Council acting unanimously decides otherwise’. 35 [2015] OJ L84/39. 36 See Art. 8(2)(b), Annex II and para 1.2(e) of the Athena Decision. 37 See Art. 44(3) of the Athena Decision. Art. 3 of the Decision provides that ‘[w]ith a view to the administrative management of the financing of Union operations with military or defence implications, Athena shall have the necessary legal capacity, in particular, to hold bank accounts, acquire, hold or dispose of property, enter into contracts and administrative arrangements and be a party to legal proceedings’. 38 Referred to in Art. 6 of Council Decision 2009/906/CFSP of 8 December 2009 on the EUPM in Bosnia and Herzegovina (BiH) [2009] OJ L322/22.

140 Research handbook on the EU’s common foreign and security policy confirming the above decision of the Chief of Personnel, entailing the redeployment of the applicant and, secondly, for damages.39 While in H v Commission and Council the judgments of both the General Court and the Court of Justice mainly focus on the issue of jurisdiction of the Court to rule on the action in the light of the limitations set out in paragraph 1 of Article 24 TEU and Article 275 TFEU, the case is also illustrative of the question whether the measures (of the Head of Mission) in question should be attributed to the Union or to Member State (that is, Italian) authorities. Indeed, starting from a premise that they were attributable to the Union, the applicant argued for the need to recognize the jurisdiction of the CJEU: her argument was that in the absence of such jurisdiction, she would be denied the right to an effective remedy given that the national courts could neither annul those decisions nor order the institutions of the EU to compensate her for the harm that they have caused.40 The General Court, however, dismissed that argument by affirming that the contested decisions could ‘in principle be attributed to the Italian authorities’41 and that, accordingly, ‘the legality of those measures must be reviewed by the Italian court’.42 This conclusion was essentially based on the assessment that, in adopting the decisions in question, the Head of Mission had acted pursuant to the powers delegated to him by the Italian authorities as regards seconded staff as well as on the principle that measures adopted pursuant to delegated powers are normally attributed to the delegating institution. According to the General Court, it was for the Italian courts to review the legality of the contested decision and to rule on possible damages.43 On appeal, Advocate General Wahl, while agreeing with the General Court that it was (in the absence of jurisdiction of the CJEU) for the Italian courts to examine the lawfulness of the contested decisions and to rule on the claim for damages, took a different view on the question of attribution of the decisions in question.44 In his submission, the fact that the contested measure had not been taken on behalf or in the name of a Member State (Italy) but on behalf of the Union was evident given the chain of command of the EUPM as set out in Article 9 of the Decision 2009/906/CFSP:45 while the political control and strategic direction of the mission is exercised by the Political Committee of the Council (PSC) under the responsibility of the Council and the High Representative, the mission is, for the purposes of its actual conduct and day-to-day management, headed by the Civilian Operation Commander at the ‘strategic level’ and by the Head of Mission who exercises command and control ‘at theatre level’. Given that both are appointed, and entrusted with their respective powers, by the Council and/or the PSC, to whom they also report, the Head of Mission, in adopting the contested decisions, had acted as an EU body on the basis of provisions of EU law. Accordingly, the Advocate General concluded the applicant should have brought proceedings against the EU before the national courts, requesting a declaration of 39 Case T-271/10 H v Council ECLI:EU:T:2014:702 and, on appeal, Case C-455/14 P H v Council and Commission ECLI:EU:C:2016:569. 40 Case T-271/10 H v Council (n 39) para 30. 41 ibid, para 50. 42 ibid, para 52. 43 ibid, paras 50–53. 44 Case C-455/14 P H v Council and Commission (n 39), Opinion of AG Wahl. 45 ibid, para 97.

Responsibility and liability for CSDP operations 141 inapplicability of the contested decisions and/or reparation for damages.46 The Court of Justice essentially agreed with the Advocate General on the question of attribution of the contested decisions: having regard to the relevant provisions of Decision 2009/906/ CFSP on the chain of command of the EUPM, the Court concluded, those decisions were ‘imputable’ to the Council.47 It disagreed, however, with both the General Court and the Advocate General on the question of jurisdiction and concluded that the acts in question did not fall within the scope of application of the second subparagraph of Article 24(1) TEU and the first paragraph of Article 275 TFEU and the Court therefore had jurisdiction to rule on the action. It is submitted that the conclusion reached by the Advocate General and the Court of Justice on the question of attribution of the contested decisions is probably a correct one. The claim concerning the redeployment of the applicant should not be conceived as ‘linked to the secondment’ of the staff member concerned, for which the Member State having seconded the applicant would remain responsible under Article 8(2) of Decision 2009/906/CFSP. The contested decisions on the reassignment of the applicant within the EUPM in Bosnia and Herzegovina rather constituted, as the Court put it, ‘acts of staff management whose purpose is the redeployment of members of the mission at theatre level’.48 According to the Decision, such decisions fall within the scope of operational control of the personnel, a power transferred by the national authorities to the Civilian Operation Commander and exercised by the Head of Mission at ‘theatre level’ under the instructions of the Commander. Thus, the redeployment of the applicant constituted an action taken by the Head of Mission in strict compliance with the command and control structure laid down by the act of the Council pursuant to which the mission had been established, that is, Decision 2009/906/CFSP.49 Therefore, the authority exercised by the Head of Mission had not been delegated to him by the Italian authorities but rather by the Council. Another matter is that according to Decision 2009/906/CFSP it is the sending Member State that retains control of the seconded staff outside ‘theatre level’. Therefore, in principle, only the sending Member State is empowered to take acts that affect the conditions of employment of the seconded staff in such a way which may alter the legal situation of that member of staff. 46

ibid, para 99. Case C-455/14 P H v Council and Commission (n 39), para 68 of the judgment. 48 ibid, para 59. As the Advocate General pointed out in para 85 of his opinion, the legal or economic status of the appellant was, in substance, not altered by the decision to relocate her to the regional office of Banja Luka. In particular, neither her occupational grade nor her remuneration was affected by that decision. The appellant had explicitly agreed to serve in a position in the EUPM other than that she had originally applied for. 49 In that regard, see also the judgment of the General Court, to which the case was referred back for the judgment on the substance of the action, in Case T-271/10 RENV, H v Council, ECLI:EU:T:2018:180, paras 57–72. The General Court concluded that, in adopting the contested decisions, the Head of Mission had acted in conformity with Council Decision 2009/906, as complemented by the Operation Plan (OPLAN) of the EUPM and the Guidelines for Command and Control Structure for EU Civilian Operations in Crises Management. An appeal (Case C-413/18 P, H v Council) is currently pending before the Court of Justice. 47

142 Research handbook on the EU’s common foreign and security policy While H v Commission and Council was about attribution, as between the Union and one of its Member States, of conduct in the context of a civilian CSDP mission, another question would be whether the EU could incur non-contractual liability in the sense of the second paragraph of Article 340 TFEU on the basis of conduct related to the CSDP missions having military or defence implications. It has already been pointed out that the administrative expenditure arising from military operations is not charged to the EU budget but to the budgets of the relevant Member States and administered through the mechanisms set out in the Athena Decision.50 Athena is endowed with the legal capacity, inter alia, to ‘be a party to legal proceedings’.51 As far as non-contractual liability is concerned, Article 44(4) of the Decision provides that any damage caused by the operation headquarters, force headquarters and component headquarters of the crisis structure, the composition of which shall be approved by the operation commander, or by their staff in the course of their duties shall be covered through Athena by the contributing states, ‘in accordance with the general principles common to the laws of the Member States and the staff regulations of the forces, applicable in the theatre of operations’. Article 44(5) of the Athena Decision then explicitly rules out (subsidiary) liability of the Union or the Member States (for both contractual and non-contractual liability) for damage caused by the units and departments of the crisis structure, the composition of which shall be approved by the operation commander, or by their staff in the course of their duties. These specific arrangements would appear to indicate that non-contractual liability of the Union under Article 340 TFEU is excluded as far as military operations are concerned. Be that as it may, a puzzling aspect of H v Commission and Council concerns the manner in which the General Court, the Advocate General and the Court of Justice first address the question of the limits of the Court’s jurisdiction in the light of the second subparagraph of Article 24(1) TEU and the first paragraph of Article 275 TFEU and only thereafter consider the issue of attribution of the decision in question to either the Union or the Member State concerned. By so doing they sideline the principle that, logically, the question of attribution always arises before there is a need to consider the scope of the exception to the Court’s jurisdiction in the field of the CFSP. As already noted, this is the case because the CJEU does not, in any event, have jurisdiction to review conduct attributable to a Member State, nor to award damages on the basis of such conduct. 3.3 The Question of the Proper Defendant before the CJEU Once it has been established that a given conduct forming the basis for the alleged contractual or non-contractual liability is attributable to the Union, it will have to be determined against whom the action should be brought before the General Court. As a general rule, where proceedings are brought in respect of its non-contractual liability, the Union is represented before the CJEU by the institution or institutions responsible 50

Under Art. 15(1) of Annex I to the Athena Decision, the mechanism shall bear as ‘common costs’, inter alia, ‘indemnities for damages and costs resulting from claims and legal actions to be paid through Athena’. 51 See Art. 3 of the Athena Decision.

Responsibility and liability for CSDP operations 143 for the matters at the origin of the alleged damage.52 Should the action have been directed against a defendant other than the one deemed to be responsible for the damage, the Court would hold the action inadmissible.53 If, by mistake, a defendant other than the one responsible for the act in question has been identified and if it is clear, in the light of the application, against whom the action ought to have been directed, the Court would order the former to be replaced by the latter.54 A specific question arising in the present context is whether a claim for damages originating in the conduct of a CSDP mission may be brought against the mission or operation concerned in its own right or, alternatively, against one (or several) of the Union’s institutions. In that regard, a distinction should be drawn between, on the one hand, civilian CSDP missions and, on the other hand, operations having military or defence implications. As regards, first, civilian CSDP missions, it now follows from an express provision in their constituent instruments that such missions are endowed with legal personality, including ‘the capacity to … be a party to legal proceedings, as required in order to implement [the] Joint Action [concerned]’.55 Accordingly, it has been confirmed by the General Court that such missions may act as a defendant in legal proceedings before Union courts.56 Therefore, actions aimed at discharging, on the one hand, contractual liability of the Union under Article 272 and the first paragraph of Article 340 TFEU and, on the other hand, non-contractual liability under Article 268 and the second paragraph of Article 340 TFEU, for conduct attributable to a civilian CSDP mission should be directed against the relevant mission, rather than the institution that may have delegated powers to that mission. The position of civilian missions is therefore similar to Union bodies, offices and agencies having legal personality and, as such, capable of incurring contractual and non-contractual liability.57 52

See e.g. Case T-383/00 Beamglow v Parliament and Others ECLI:EU:T:2005:453, para

68. 53

Case T-162/89 Mommer v Parliament ECLI:EU:T:1990:72, paras 19–20. See e.g. the order of the Tribunal of 16 October 2006 in Case T-173/06 Aisne and Nature v Commission ECLI:EU:T:2006:320, paras 17–18 and the order of the General Court of 6 January 2015 in Case T-479/14 Kendrion v European Union ECLI:EU:T:2015:2, para 13. Cf, however, Opinion of AG Lenz in Case C-62/83 Eximo v Commission ECLI:EU:C:1984:197. Indeed, in H v Council and Commission, the Court held that, since the Commission was not involved in the chain of command of the EUPM in Bosnia and Herzegovina, and the contested decisions did not concern the implementation of the EUPM’s budget, those decisions could not be ‘imputed’ to the Commission. Consequently, the action was dismissed as inadmissible in so far as it had been directed against the Commission. See Case C-455/14 P H v Council and Commission (n 39), para 65. 55 See e.g. Art. 15 bis of Council Joint Action 2008/124/CFSP on the European Union Rule of Law Mission in Kosovo, EULEX KOSOVO [2008] OJ L340/73, as amended by Council Decision 2014/349/CFSP [2014] OJ L174/42. A similar provision has been introduced into the constituent instruments of other civilian CSDP missions. 56 Order of the General Court of 9 November 2016 in Case T-602/15 Jenkinson v Council and Others (n 33) para 35. 57 The current legal situation is therefore different from the one at issue in Case T-213/12 Elitaliana v Eulex Kosovo ECLI:EU:T:2013:292, and, on appeal, Case C-439/13 P Elitaliana v Eulex Kosovo ECLI:EU:C:2015:753 – a dispute relating to the award of a public service contract 54

144 Research handbook on the EU’s common foreign and security policy As regards, secondly, operations with military or defence implications, no grant of legal personality has been provided. Therefore, any action aimed at discharging contractual or non-contractual liability before the Union courts should in all likelihood be directed at either Athena58 or the Council as the institution which has delegated to the EU Operation Commander and/or EU Force Commander the authority to conduct the operation. Given that military operations are charged to the budgets of the Member States, claims against the Commission would clearly be excluded. Insofar as military operations are concerned, the question of the representation of the Union before Union courts would, in any event, appear as a theoretical one given the limitations of the jurisdiction of the CJEU based on paragraph 1 of Article 24 TEU and Article 275 TFEU.59 Jurisdiction for these claims would in all likelihood fall upon domestic courts of the Member States.60 3.4 Limitation of Jurisdiction of the CJEU in the Field of the CFSP Once it has been established that any given conduct in the course of a CSDP mission or operation is attributable to the Union and once the claim is deemed to have been brought against the correct defendant, it will still have to be confirmed that the Court’s jurisdiction to examine the action is not affected by the limitations of its jurisdiction set out in paragraph 1 of Article 24 TEU61 and the first paragraph of Article 275 TFEU.62 concerning a project involving helicopter support for EULEX Kosovo. The General Court dismissed the action as inadmissible on the ground that Eulex Kosovo did not (at the time) have legal capacity to be a defendant. On appeal, the Court of Justice reached the same conclusion, drawing attention to the fact that, having regard to Arts 8(5) and 16(4) of the Joint Action 2008/124/CFSP, the Head of Mission was, when signing the procurement contract, exercising his powers on the basis of a delegation from and under the supervision and authority of the Commission. Hence, the contested measures being attributable to the Commission as the delegating authority, the action (for annulment and for damages) should have been brought against that institution rather than Eulex Kosovo. 58 See Art. 3 of the Athena Decision. 59 In his opinion of 21 May 2015 in Case C-439/13 P Elitaliana v Eulex Kosovo (n 57), AG Jääskinen pointed out that the award of contracts similar to the one at issue in the case could nonetheless fall outside the jurisdiction of the EU Courts if those contracts relate to military actions (para 60). 60 Naert, ‘Shared Responsibility in the Framework of the EU’s CSDP Operations’ (n 1) 695, reports a judgment of the Brussels Court of First Instance of 27 November 2015, ruling that the court had jurisdiction over a claim against Athena relating to the award of a contract for the EU military operation in the Central African Republic (EUFOR RCA). 61 The final sentence of paragraph 1 of Art. 24 TEU provides that ‘[t]he Court of Justice of the European Union shall not have jurisdiction with respect to these provisions, with the exception of its jurisdiction to monitor compliance with Article 40 of this Treaty and to review the legality of certain decisions as provided for by the second paragraph of Article 275 of the Treaty on the Functioning of the European Union’. 62 According to the first paragraph of art 275 TFEU, ‘[t]he Court of Justice of the European Union shall not have jurisdiction with respect to the provisions relating to the common foreign and security policy nor with respect to acts adopted on the basis of those provisions’. The second paragraph, however, ‘re-introduces’ the Court’s jurisdiction by providing that ‘[h]owever, the Court shall have jurisdiction to monitor compliance with Article 40 of the Treaty on European

Responsibility and liability for CSDP operations 145 Should the Court’s jurisdiction be excluded on that basis, the liability of the Union cannot be discharged before Union courts.63 A more general analysis of the limitations of the jurisdiction of the CJEU in the field of the CFSP exceeds the scope of the present chapter. It is enough to note that the question concerning the limitations to the jurisdiction of the Court has been treated in two recent judgments specifically concerning the liability of the Union for conduct attributable to the Union in the context of CSDP operations, that is, Elitaliana SpA v Eulex Kosovo64 and H v Council and Commission.65 The jurisdiction aspect of this case law is addressed in another chapter.66

4. DOMESTIC JURISDICTION OF THE MEMBER STATES In some cases, proceedings aimed at discharging liability arising from the conduct of CSDP operations may also fall within the jurisdiction of domestic courts of a Member State.67 While in most cases the defendant before a domestic court would be the Member State concerned, the Member State typically having contributed troops or personnel to an operation, it cannot be ruled out that proceedings before a domestic court might be brought either against the EU in its own right or against bodies, offices or agencies of the Union endowed with a capacity to be a party to legal proceedings.68

Union and to rule on proceedings, brought in accordance with the conditions laid down in the fourth paragraph of Article 263 of this Treaty, reviewing the legality of decisions providing for restrictive measures against natural or legal persons adopted by the Council on the basis of Chapter 2 of Title V of the Treaty on European Union’. 63 As the General Court has confirmed in Case T-328/14 Jannatian v Council ECLI: EU:T:2016:86 para 31, ‘a claim seeking compensation for the damage allegedly suffered as a result of the adoption of an act relating to the CFSP falls outside the jurisdiction of the Court’. 64 Case T-213/12 Elitaliana v Eulex Kosovo and, on appeal, Case C-439/13 P Elitaliana v Eulex Kosovo (n 57). 65 Case T-271/10 H v Council and, on appeal, Case C-455/14 P H v Council and Commission (n 39). As regards the Court’s jurisdiction to review CFSP acts post Lisbon, see also Case C-658/11 Parliament v Council ECLI:EU:C:2014:2025, concerning the legality of Council Decision 2011/640/CFSP of 12 July 2011 on the signing and conclusion of the Agreement between the European Union and the Republic of Mauritius on the conditions of transfer of suspected pirates and associated seized property from the European Union-led naval force to the Republic of Mauritius and on the conditions of suspected pirates after transfer [2011] OJ L254/1. 66 See Chapter 4 in this volume. 67 Or, as the case may be, a third state that has provided personnel to a CFSP mission or operation. 68 Proceedings might be brought against civilian CSDP missions or the Athena mechanism, for example.

146 Research handbook on the EU’s common foreign and security policy It follows from Article 274 TFEU69 that the Union does not have immunity before domestic courts of the Member States.70 In relation to questions of contractual liability, the jurisdiction of a domestic court may be based on an arbitration clause contained in a contract of employment between the mission and local or international staff recruited on a contractual basis.71 While the CJEU has, in principle, exclusive jurisdiction to adjudicate actions concerning the Union’s non-contractual liability under Articles 268 TFEU and the second paragraph of Article 340 TFEU,72 it has been recently suggested that – regard being had to the limitations to the jurisdiction of the CJEU in the field of the CFSP73 – Article 274 TFEU could provide an individual with access to domestic courts in those cases where the CFSP is implemented by the EU’s institutions, bodies, offices or agencies in such a way as to be of direct and individual concern to an applicant; those courts could then review the CFSP acts in question as to their compatibility with higher-ranking EU law and, as the case may be, suspend their application in the given case74 and, possibly, award damages.75 In that regard, the state of the law is uncertain given that the Court has not yet had an occasion to rule on the issue76 and given that no proceedings have been brought against the Union before Member State courts aimed at discharging the liability of the Union arising from the conduct of the CFSP in general or CSDP operations in particular. In the submission of the present author, that question is not decided by the Rosneft case: in that judgment, the Court merely confirmed that it also has (an exclusive) jurisdiction under Article 267 TFEU to review the legality of decisions providing for restrictive measures against natural or legal persons adopted by the Council on the basis of Chapter 2 of Title V of the TEU.77 The judgment therefore has no bearing on the issue whether or not domestic courts of Member States might be entitled to suspend the application of CFSP acts 69 According to Art. 274 TFEU, ‘[s]ave where jurisdiction is conferred on the Court of Justice of the European Union by the Treaties, disputes to which the Union is a party shall not on that ground be excluded from the jurisdiction of the courts or tribunals of the Member States’. 70 It may observed that the immunity of officials and other servants of the Union from legal proceedings in Member State courts, based on Art. 11 of the Protocol (No. 7) on the privileges and immunities of the European Union [2012] OJ C326/266, has no bearing on the liability of the Union. See Case 5/68 Sayag and others ECLI:EU:C:1968:42. 71 See e.g. the order of the General Court of 30 September 2014 in Case T-410/13 Burim Bitigi and others v Commission and others (n 33). 72 See e.g. K Lenaerts, I Maselis and K Gutman, EU Procedural Law (OUP 2014) 691. 73 See Case T-328/14 Jannatian v Council (n 63), paras 30–32. 74 See the view of AG Kokott in Opinion 2/13 regarding accession of the European Union to the European Convention for the Protection of Human Rights and Fundamental Freedoms ECLI:EU:C:2014:2475, paras 99–100. 75 See Case C-455/14 P H v Council and Commission (n 39), Opinion of AG Wahl, paras 41–44 and 101–103. 76 Notably, while not contesting the jurisdiction of domestic courts as such, in his Opinion in Case C-72/15 Rosneft (n 28), AG Wathelet, arguing in favour of acknowledging the jurisdiction of the CJEU under Article 267 TFEU in matters of the CFSP, seems to distance himself from the position expressed by AG Kokott in Opinion 2/13. See especially footnote 15 of the Opinion AG Wathelet. 77 Case C-72/15 Rosneft (n 28), paras 77–80.

Responsibility and liability for CSDP operations 147 other than restrictive measures against natural or legal persons, that is, those CFSP acts that do not fall within the jurisdiction of the CJEU or, as the case may be, award damages on the basis of breaches of EU law falling within the scope of the CFSP, as proposed by AG Kokott in Opinion 2/13 and AG Wahl in H v Commission and Council. Another matter is that neither Advocate General articulates in any detail the conditions for, or the implications of, a decision by a domestic court to suspend the operation of a CFSP act without the possibility of referring the question of its validity to the CJEU. What is clear is that the conditions for the suspension of the implementation of an act adopted within the ‘communitarised’ EU policies set out in Zuckerfabrik78 would not apply. Presumably, the suspension of the implementation of a CFSP act would only apply vis-à-vis the applicant in the given case and, in the absence of the jurisdiction of the CJEU, ‘it would then be for the EU institution responsible for the act to draw the necessary inferences from the decision of the national court; by repealing or amending the act whose application vis-à-vis the applicant has been suspended’.79 As far as the liability of a Member State is concerned, that could be based on the principle of liability of a Member State for breaches of EU law. In the field of the CFSP, however, the reliance on the Court’s case law governing such liability might prove challenging, if only in the light of the requirement to establish a breach of a rule of EU law of a kind that is intended to grant rights to individuals.80 Given the inherently inter-governmental nature of CFSP acts, such a prospect would seem merely theoretical and, to the knowledge of the present author, there is no case law on the matter. In a much more realistic scenario, a domestic court of a Member State (or a third state participating in an operation or a mission) could be required to decide on claims aimed at discharging liability incurred in the context of a CSDP operation as a matter of domestic law (or international law applied by such courts). In this regard, relevant claims would fall into two main categories. First, domestic courts of a Member State have (civil, criminal or administrative) jurisdiction in respect of the personnel transferred or seconded to an operation by the Member State concerned.81 In the case of civilian missions, the act establishing a mission usually provides in express terms for the responsibility of the Member State having seconded a member of staff for any claims ‘linked to the secondment’, from or concerning the member of staff, as well as for the responsibility of that state for bringing any action against the seconded person.82 Likewise, responsibility for any disciplinary action or the exercise of criminal jurisdiction rests with the Member State having seconded the member of personnel 78 Joined Cases C-143/88 and C-92/89 Zuckerfabrik Süderdithmarschen and Others ECLI: EU:C:1991:65, para 33. 79 Case C-455/14 P H v Council and Commission (n 39), Opinion of AG Wahl, para 103. 80 See e.g. Joined Cases C-6/90 and C-9/90 Francovich and Bonifaci v Italy (n 24), para 40. 81 SOFAs and SOMAs provide that the personnel of an EU force or mission are not exempted from the jurisdiction of the respective sending states. See Art. 6(7) of the EU Model SOFA and Art. 6(8) of the EU Model SOMA. The provisions of the relevant SOFA or SOMA are made applicable to the personnel contributed to a given operation or mission by a third state by virtue of an agreement concluded between the EU and the third state concerned. 82 See e.g. Art. 10(2) of the Council Joint Action 2008/124/CFSP on the European Union Rule of Law Mission in Kosovo (n 55).

148 Research handbook on the EU’s common foreign and security policy concerned.83 As an example from actual practice, Naert reports of a French soldier who served in the EU military operation in the Republic of Chad and in the Central African Republic (EUFOR Chad/CAR)84 and killed two fellow French soldiers, a UN peacekeeper and a local peasant, and who was prosecuted in France.85 More recently, cases of sexual exploitation and abuse of minors by Georgian troops participating in the EU military operation in the Central African Republic (EUFOR RCA)86 have also been reported87 and investigated.88 Secondly, domestic courts of Member States may also be requested to entertain claims against Member State authorities by third parties alleging to have suffered damage from actions taken in the context of a CSDP operation. While the powers of review of those courts might in certain Member States be circumscribed by doctrines such as ‘non-justiciability’89 or ‘acte de gouvernement’,90 reflecting the principle that certain activities of the government relating to foreign policy remain outside the jurisdiction of courts,91 one finds no evidence of the application of such considerations to acts relating to the day-to-day conduct of CSDP operations of the EU. On the contrary, in a rare instance of domestic case law concerning a third-party action against Member State authorities relating to acts committed by German forces in the context of 83 ibid, Art. 8(6). The Model SOFAs and SOMAs further provide that ‘the competent authorities of a Sending State shall have the right to exercise on the territory of the Host State all the criminal jurisdiction and disciplinary powers conferred on them by the law of the Sending State with regard to … personnel’. 84 See e.g. Council Joint Action 2007/677/CFSP on the European Union military operation in the Republic of Chad and in the Central African Republic [2007] OJ L279/21. 85 Naert, ‘The International Responsibility of the Union in the Context of its CSDP Operations’ (n 1) 323. 86 See Council Decision 2014/73/CFSP on a European Union military operation in the Central African Republic (EUFOR RCA) [2014] OJ L40/59. 87 See press release of the UN High Commissioner for Human Rights (29 January 2016) http://www.ohchr.org/EN/NewsEvents/Pages/DisplayNews.aspx?NewsID=16995. 88 See the report of the UN Secretary General, Special measures for protection from sexual exploitation and abuse: a new approach, UN Doc. A/71/818 (28 February 2017) 63. 89 For the United Kingdom, see e.g. R (Abbasi and Anor) v Secretary of State for Foreign & Commonwealth Affairs and Anor [2002] EWCA Civ 1598, paras 99 and 106. 90 For France, see e.g. Conseil d’Etat (23 juillet 2010, Société Touax, n° 328757), for the holding that ‘les opérations militaires ne sont, par nature, pas susceptibles d’engager la responsabilité de l’Etat, y compris sur le fondement de la rupture de l’égalité devant les charges publiques [i.e.: responsabilité sans faute]; que les préjudices résultant d’opérations ayant ce caractère ne sauraient ainsi ouvrir aux victimes droit à réparation à la charge de l’Etat que sur le fondement de dispositions législatives expresses’. The question, however, is what constitutes a ‘military operation’ in the sense of this case law. As the rapporteur public (equivalent to an Advocate General of the CJEU) held in Société Touax: ‘L’îlot d’irresponsabilité lié à la conduite des opérations militaires nous semble pouvoir, encore aujourd’hui, trouver une justification … à la condition d’user de ce régime avec modération, à la fois dans l’identification des opérations militaires mais aussi dans l’appréciation du lien avec la conduite de ces opérations. C’est dire qu’il convient de s’en tenir à une application stricte du principe d’irresponsabilité à raison des opérations militaires, comme pour tout régime dérogatoire.’ 91 See J Crawford, Brownlie’s Principles of Public International Law (8th edn, OUP 2012) 72ff.

Responsibility and liability for CSDP operations 149 a CSDP military operation, there is no indication of limitations of the above kind to the jurisdiction of the courts in question. The case at hand concerned the review – by, first, the Administrative Court of Cologne92 and, on appeal, the Oberverwaltungsgericht Nordrhein-Westfalen93 – of measures taken by the German authorities in the course of Operation Atalanta,94 involving the capture, detention and transfer of nine suspected pirates of the Somalian nationality to the authorities of Kenya for criminal proceedings and sentencing in March 2009. According to the claimants, the right to a fair trial enshrined in Article 6 of the European Convention on Human Rights had been breached by the contested measures, which did not meet the basic standards of the rule of law relating to the rights of defence, the duration of the procedure and the conditions of detention. In their defence, the German authorities submitted that the proceedings had been brought against the wrong defendant in the light of the fact that both the capture and transfer of the claimants were based on Joint Action 2008/851/CFSP and, as such, attributable to the EU. In respect of the contested measures the German naval forces had not acted in their national capacity but as part of the EUNAVFOR (EU Naval Force) given that the command and control over Operation Atalanta resided with the Union. In relation to the capture and detention of the suspected pirates, it could, according to the Administrative Court, be left open whether the measures concerned should be attributed to either the German or the EU authorities; in any event, there existed no basis in substantive law for their challenge. As regards, however, the transfer of the persons concerned to the Kenyan authorities, the Court admitted the claim, considering that the act of transfer was an act of the German State (‘Akt deutscher Staatgewalt’) attributable to the German authorities.95 Notwithstanding the fact that it was, fundamentally, the EU institutions that had had the command and control of the forces in the operation area, the concrete orders in respect of the transfer of the claimants had been made by the German authorities. First, immediately following the capture of the suspected pirates an inter-ministerial decision-making forum had been created in Germany to examine the further measures concerning the captured individuals. In that context, the transfer of the persons concerned to Germany for trial had been cancelled and replaced by the objective of their surrender to Kenya. That latter objective was then communicated by the Ministry of Defence to the captain of the vessel on board which the suspected pirates were held, together with an order to take a course to Mombasa. In the view of the Court, the German authorities had had the choice to cancel the transfer of the suspects for trial to either Kenya or another third state and to have them put on trial in Germany instead. Even if those authorities might appear also to have taken 92 VG Köln, Urteil vom 11. November 2011, Az. 25 K 4280/09, openJur 2012, 83059, available in German at http://openjur.de/u/451905.html. 93 Oberverwaltungsgericht Nordrhein-Westfalen, Urteil vom 18. September 2014, available in German at http://openjur.de/u/731026.html. 94 Council Joint Action 2008/851/CFSP on a European Union military operation to contribute to the deterrence, prevention and repression of acts of piracy and armed robbery off the Somali coast [2008] OJ L301/33. 95 The central reasoning is at paras 60–61 of the judgment of the Administrative Court.

150 Research handbook on the EU’s common foreign and security policy directions from the EU Operation Commander (EUOPC),96 this would not, according to the Administrative Court, have led to a different conclusion as far as the attribution of the contested measures is concerned since those directions appeared as nothing more than a (subsequent) approval of the decisions previously taken solely by the German authorities. Finally, the central role of the authorities in the transfer was also illustrated by the fact that the preparations for the transfer of the persons concerned with the Kenyan authorities had been undertaken essentially by the embassy of the Federal Republic and not by the representatives of the EU. On appeal, the Oberverwaltungsgericht Nordrhein-Westfalen agreed with the Administrative Court on the question of attribution of the contested measures. While the Oberverwaltungsgericht first addressed at length the question whether the lack of international personality of the Union prior to the entry into force of the Treaty of Lisbon as such precluded the attribution of the contested measures to the Union,97 that circumstance was not, according to the Appeal Court, decisive for the dismissal of the appeal: even on the assumption that the EU had been considered to possess international legal personality at the time, the contested measures were to be attributed to the German authorities in the light of the fact that the transfer of the suspected pirates had been both initiated and implemented solely by those authorities. In that regard, it could also be left open whether the vessel in question had been under the operative command and control of the EU Operational Headquarters (OHG) of the EUNAFOR.98 According to the Appeal Court it had already become clear at the meeting of the inter-ministerial decision-making forum on 4 March 2009 that a transfer of the suspects to Kenya was to be favoured by the German authorities, who then, on the following days, prepared the concrete framework for the transfer together with the Kenyan authorities. While on 6 March 2009 an Exchange of Letters was made between the EU and Kenya with a view to defining the conditions and modalities for the transfer of persons suspected of having committed acts of piracy on the high seas and detained by EUNAFOR,99 there apparently was – in the context of the concrete case – no willingness on the part of the German authorities to involve the Presidency of the Council in the negotiations on the preparation of the transfer of the suspects in question.100 The Appeal Court also placed emphasis on the fact that, within the German inter-ministerial forum, several alternative courses of action (including a criminal trial in Germany) had stood open to the authorities, who had, however, preferred a transfer

96 In this regard, the Court makes reference to the Decision of the EUOPC of 8 March 2009 reading as follows: ‘Decision to transfer to any third country is made by OPCDR [Operation Commander] based on available evidence. OPCDR has made this decision today, 08.03.2009, and approved transfer.’ 97 See paras 80–92 of the judgment. 98 ibid, para 93. 99 [2009] OJ L79/49. 100 Para 94 of the judgment.

Responsibility and liability for CSDP operations 151 of the suspects to Kenya.101 Moreover, there was, according to the Oberverwaltungsgericht, no evidence that an order concerning the transfer had been given by the OHQ:102 on the contrary, the Court agreed with the assessment of the Administrative Court that the Decision of the EUOPC of 8 March 2009 on the ‘approval’ of the transfer could be considered as nothing other than a (subsequent) approval of a decision on the transfer made by the German authorities.103 Finally, even if those authorities had been considered to have acted on the basis of an authorization from Union institutions – that is, Joint Action 2008/851/CFSP – the Appeal Court concluded that it fell to the domestic courts of the Member States to grant legal protection in regard to the measures of those authorities as, in any event, the CJEU would have no jurisdiction over measures falling within the CFSP.104 The judgments of both the Administrative Court and the Appeal Court leave open whether the assessment of the question of attribution would have been different had the German authorities acted upon orders of the OHQ or the EUOPC, or had the input of the EUNAVFOR command structure been greater in some other way. On the one hand, the judgments emphasize that no orders were given by the OHQ. On the other hand, the Appeal Court notes that it would not have made any difference in regard to the question of attribution even if the EU had made an ‘autonomous decision’ on the transfer.105 Be that as it may, the reasoning of both courts appears to be based on the understanding that the implementation of CFSP measures by Member State authorities in any event entails the liability of the Member State in question before a domestic court, irrespective of whether EU institutions might also be held liable as a matter of EU law. Insofar as the Appeal Court in particular is concerned, this conclusion appears to have been motivated by the desire to ensure that adequate legal protection is provided by a court of law, something the CJEU might not be in a position to do in light of the limitations to its jurisdiction in the field of the CFSP. As to the substance, both the Administrative Court and the Oberverwaltungsgericht held that the surrender of the suspected pirates to Kenya had been unlawful, including in light of the fact that the conditions of detention in the said country amounted to inhuman or degrading treatment in the sense of Article 3 of the European Convention.

5. CONCLUSIONS AND EVALUATION Any attempt to assess the question of responsibility or (for the purposes of EU law as well as domestic law of the Member States) liability in the context of CSDP operations is still marked by the fact that the relevant case law and other judicial practice remain rather limited. As regards international responsibility and non-contractual liability 101

ibid, para 95. As the Appeal Court put it at para 140, ‘die Übergabe was jedenfalls zwangsläufige Folge ihrer Entscheidung, ihrer Vorrecht auf eine Strafverfolgung in Deutschland nicht wahrzunehmen’. 102 ibid, para 109. 103 Ibid, para 111. 104 ibid, paras 115–119. 105 ibid, para 140.

152 Research handbook on the EU’s common foreign and security policy vis-à-vis third parties in particular, the claims are as a rule settled amicably through ex gratia payments. So far, relatively few cases have ended up either before a claims commission, the CJEU (Elitaliana v Council and Commission; H v Council and Commission as well as a number of staff cases) or domestic courts of a Member State (the German case concerning the surrender of pirates). In spite of the limited scope of the existing practice, some conclusions on the question of the distribution of responsibility or liability as between the EU and its Member States may nonetheless be drawn from both the institutional practice of the CSDP and the case law of the European and domestic courts. First, having regard to the chain of command designed for CSDP operations, running from the level of the Council right down to the Head of Mission or the Operation Commander, the CJEU is inclined to treat the conduct of such operation as being attributable to the EU in its own right (H v Commission and Council). Coupled with the Court’s narrow understanding of the limits of its jurisdiction in the field of the CFSP (Elitaliana; H v Commission and Council; Rosneft), the case law suggests that there would be at least some legal remedies available before the EU Courts even in matters relating to operational activity of the CFSP. Secondly, for those cases where no remedies before the CJEU exist (either as a consequence of the attribution of a given conduct to a Member State or in view of the limitations of the Court’s jurisdiction specific to the CFSP), the Court seems to acknowledge that remedies may still exist before the domestic courts of the Member States. The German practice provides an illustration of how the liability of a Member State for an action related to the conduct of a CFSP operation may effectively be charged before domestic courts. In that regard, one should also recall the possibility of an individual bringing a case against the given Member State before the ECHR. As regards, thirdly, the question of an appropriate rule of attribution under public international law in general and the DARIO of the International Law Commission in particular, the manner in which that question is treated in the case law of the CJEU appears to place less weight upon the actual command and control over a given conduct, the focus being more on the formal structure of the chain of command laid down in generalized fashion for all CSDP operations. This, it is submitted, could be interpreted as speaking for the treatment of such operations as de facto organs of the EU. While the question of attribution of conduct as between the EU and its Member States would seem to remain relevant at the level of public international law, if only in light of the fact that CSDP operations do not possess international legal personality of their own (and, as a corollary, may not be regarded as being internationally responsible in their own right), the same may not be true for the level of EU law or the domestic laws of the Member States. This is the fourth concluding point: this chapter suggests that, within those legal orders, there may be other entities capable of charging the requisite liability. Indeed, civilian CSDP missions have recently been endowed with legal personality of their own, including the capacity to be a party to legal proceedings before courts of law. In relation to military operations, claims for compensation may be directed against the Athena mechanism in its own right, possibly before a domestic court. Consequently, redress may be obtained without necessarily having to address the

Responsibility and liability for CSDP operations 153 question of whether any given conduct undertaken within the framework of a CSDP operation should, in the final analysis, be attributed to the EU or (one or more of) the Member States.

8. Capabilities and CSDP: resourcing political will or paper armies Simon Duke

1. INTRODUCTION AND TERMINOLOGY The term ‘capabilities’ is used variously in CSDP to denote resources that can be used collectively to attain a given goal. Resources, in turn, can be considered a reference to physical assets, which may take the form of either equipment or trained personnel. There is, however, no generally agreed definition for capabilities, resources or other associated terms like competences. In a useful discussion on terminology Galavan defines capabilities as ‘the capacity to deploy a combination of resources through collective organizational routines to achieve goals’. Resources, in this context, are ‘tangible and intangible assets under the effective control of the organization’.1 There is the risk that generic definitions of capabilities may lead to rather static ideas or bean counter exercises to establish physical assets, whereas it should be considered to be a dynamic concept that includes not only identifiable and quantifiable ‘assets’, but also the ability to improve the use of assets (and thus upgrade them in a sense) through training and the incorporation of lessons learned from field use. A further definitional quandary arises in the case of CSDP since the capabilities employed for missions or operations are, with a few minor exceptions, not those of the EU but those of the Member States.2 Capabilities in this context cannot therefore be divorced from the idea of competences and must be understood as those capabilities that might be available to CSDP. Any discussion of ‘capabilities’ is therefore notional in the sense that assumptions have to be made about their availability although, in reality, they may not be. The Member States are nevertheless to operate in accordance with the principle of a ‘single set of forces’ which can be used nationally or in multilateral frameworks.3 This implies that duplications should be avoided within the ‘single set of forces’ although there is ample evidence of duplications of systems across the EU’s members.4 It is this essential uncertainty that has led some to advocate more predictable forms of ‘on-call’ capabilities or even EU-owned resources. There is nevertheless an inherent 1 RJ Galavan, ‘Understanding Resources, Competences, and Capabilities in EU Common Security and Defence Policy’ (2015) IECEU Working Paper at . 2 Generally, the EU has used the term ‘mission’ for civilian activities and ‘operations’ for military ones. 3 Council of the European Union, ‘Implementation Plan on Security and Defence’ (14 November 2016) 1, 4. 4 European Commission, ‘Reflection Paper on the Future of European Defence’ (7 June 2017) 9.

154

Capabilities and CSDP 155 ambiguity about when the EU should act as the EU and the extent to which subsidiarity should apply to crisis scenarios. In particular, the presence of mindsets that are conditioned to think in terms of national security and defence, as well as legal barriers and conditions applying to the use of force, have proven significant impediments since it is often only after resource sufficiency at the national level is satisfied that thoughts (might) turn to the provision of collective capabilities at the European level. This is of course an imperfect state of affairs since it leads, in circular motion, back to the very surpluses, duplication and shortcomings that CSDP has been battling since inception. It remains to be seen whether the newly established European Defence Fund, which offers up to €5.5 billion to facilitate research, as well as development and acquisition, will attain its core goal of helping the Member States to ‘spend money more efficiently, reduce duplications in spending, and get better value for money’.5 The existence of high levels of duplication of assets when it comes to larger-scale and more expensive items, like destroyers or fighter aircraft, should not be read as an indictment of the more recent initiatives since the development of capabilities, or changes to existing ones, demands a longer-term perspective of a decade or more. This, typically, represents the horizon for the development and production of the higherticket capabilities. Given the relatively recent development of CSDP, many of the capabilities shortfalls, to use the jargon, are still with us. This applies in particular to the military aspects of CSDP, whereas capabilities in the civilian realm tend to put more emphasis on training, common standards and dynamism (see Chapter 5 in this volume). The presence, or otherwise, of capabilities has been one of the fundamental concerns of CSDP, even avant la lettre. For instance, it was an integral part of Hill’s ‘capabilities-expectations gap’. Indeed, Hill commented that in the absence of an effective military capability the Community (as it then was) ‘would have to face the dilemma of either trusting to other forms of security and/or leaving in place the individual Member State’s armed forces and rights to use them’.6 The EU has in effect tried to do elements of both by developing CSDP as a ‘common’ policy in a highly intergovernmental space. This has created an often-awkward duality in the policy whereby consensus is required for a political decision at the European level to use military force or to launch other types of mission or operation, but this may have little bearing on the national decision on whether to make available the required personnel and resources for a CSDP mission or operation. The capabilities issue is therefore at the heart of the perceived effectiveness of CSDP and is subject to political contention. Reliance upon one of the bigger Member States (a so-called ‘framework nation’) is open to charges by smaller Member States that those missions or operations that are successfully adopted often reflect not only the capabilities, but also the will of the larger members, as was the case with the French 5

European Commission ‘A European Defence Fund’ (Press Release, 7 June 2017). C Hill, ‘The Capability-Expectations Gap, or Conceptualizing Europe’s International Role’ (1993) 31(3) Journal of Common Market Studies 305. It should be recalled that Hill’s remarks came the year after the adoption of the Petersberg tasks by the Western European Union (WEU) and well before these tasks were incorporated into the Treaty on European Union in 1997 during the European Council in Amsterdam. 6

156 Research handbook on the EU’s common foreign and security policy lead in military operations in the Democratic Republic of the Congo and Mali. From a planning perspective, it is obviously difficult to plan for timely interventions when there is a fundamental uncertainty about what and who is available (this, incidentally, is not uniquely an EU issue since it also applies to UN peacekeeping operations). The response to this has been to advocate various forms of standing or on-call capabilities, as well as various types of sharing, pooling and joint development of platforms. This chapter will commence by considering CSDP capabilities from a legal perspective, especially whether there are implied or stronger commitments on the part of the Member States to provide capabilities. I shall then consider the manner in which capabilities have been developed in the military and civilian spheres respectively, including the policies and strategies that frame the capabilities debate. Finally, I shall assess the more recent initiatives mentioned above to develop capabilities, especially in the military domain. It should, however, be noted that the defence industrial aspects of capability development are of considerable and growing importance. The European Defence Action Plan (EDAP), in particular, refers to the defence industrial aspects which, to do them justice, merit separate consideration.

2. CAPABILITIES AND THE LEGAL DIMENSION Questions of capabilities pre-date CSDP.7 The introduction of stipulations on security and defence were first specified in the new Common Foreign and Security Policy (CFSP), which formed a new ‘title’ of the 1993 Maastricht Treaty. The early days of CFSP were, in some ways literally, a baptism by fire since they coincided with the disintegration of federal Yugoslavia and the ensuing conflicts fought over a decade or so. The EU itself had no military capabilities at its disposal and had to rely upon the Western European Union (WEU) to ‘elaborate and implement decisions and actions of the Union which have defence implications’ (Article 17(3) TEU at Maastricht). The types of mission for which an ‘operational capacity’ might be necessary are framed by the 1992 Petersberg tasks, which now appear, in updated form, in Article 43(1) TEU. The terminology is general (including, for example, reference to ‘tasks of combat forces in crisis management’) and does not indicate in any detail what kind of capacities might be necessary since this falls beyond the scope of the Treaties. The role of the WEU as a capacity provider was also hampered by the fact that only ten of the EU’s members were full members. Frustration turned to embarrassment as the EU and WEU did little to stabilize the chaotic situation in former Yugoslavia or the unrest in Albania in 1997 following the unravelling of a government-backed pyramid investment scheme. Frustration at the EU’s obvious inadequacies led France and the UK to agree that the EU needs to be in a position to ‘play its full role on the international stage’ and that this should include the development of ‘the capacity for autonomous action, backed up by credible military forces, the means to decide to use them, and a readiness 7

Historically CSDP first emerged as the European Security and Defence Policy (ESDP) and, for a short time, was even called the Common European Security and Defence Policy (CESDP), before becoming CSDP. For the sake of simplicity, I shall refer to CSDP throughout although this is not strictly accurate historically.

Capabilities and CSDP 157 to do so, in order to respond to international crises’.8 Although the precise interpretation of autonomy was open to debate, the push to create what eventually became CSDP had to come from these two sources since France and the UK were, and remain, the EU’s two main military powers and are also the only members with a global horizon to their foreign policies.9 The Anglo-French political push, provided by the St Malo Declaration, was subsequently developed by the European Council and, in time, by specialist military and civilian bodies within the EU. This included the capabilities aspects which are covered in more detail in the following sections. Notwithstanding the development of what became CSDP and the demise of the WEU, the EU remained reliant upon its members. The TEU is clear about this when it states that the Petersberg tasks ‘shall be undertaken using capabilities provided by the Member States’ (Art. 42(1) TEU). This raises the question of whether there is any legal obligation under EU law on the Member States to place ‘capabilities’ at the disposal of the EU for the execution of (Petersberg) tasks. The simple answer is negative in the sense that there is no specific obligation, but there are nevertheless some general obligations relating to capabilities that are worth noting. Since CSDP is an integral part of CFSP, there are a number of stipulations that apply per extensionem to security and defence. For instance, there is the need to ‘support the Union’s external and security policy actively and unreservedly in a spirit of loyalty and mutual solidarity’ (Art. 24(3) TEU). More specifically, the purpose of CSDP is to ‘provide the Union with an operational capacity drawing on civilian and military assets. The Union may use them on missions outside the Union for peacekeeping, conflict prevention and strengthening international security in accordance with the principles of the United Nations Charter’ (Art. 42(1) TEU). Member States are also expected to ‘progressively improve their military capabilities’ (while not specifically mentioning civilian capabilities) (Art. 42(3) TEU).10 The stipulation that any operational capability may be used outside the Union is obvious enough but it should be noted that Article 42(7) TEU could imply the use of assets on the territory of a Member State, and the ‘Solidarity Clause’, or Article 222 TFEU, relates to assistance to a Member State ‘in its territory’.11 Article 42(7) TEU notes that ‘[i]f a Member State is a victim of armed aggression on its territory, the other Member States shall have towards it an obligation of aid and assistance by all means within their power, in accordance with Article 51 of the United Nations Charter’ 8

Saint-Malo, Joint Declaration Issued at the British-French Summit, Saint-Malo, France, 3–4 December 1998. 9 This chapter was written at a time when the UK was still a full member of the European Union. Although somewhat conjectural, the potential impact of Brexit upon capabilities is discussed throughout the text. 10 It should be noted that the initial emphasis was upon the military dimensions of crisis management, as a reaction to the unrest in the Western Balkans in the early 1990s. The civilian aspects of crisis management were included later following the 2000 European Council in Feira. Other parts of the Lisbon Treaty (like Article 42(1)) clarified the scope of EU crisis management to include both the military and civilian dimensions. 11 See P Koutrakos, The EU Common Security and Defence Policy (OUP 2013) 68–72; European Parliament, ‘The EU’s mutual assistance clause’ (Briefing, November 2015).

158 Research handbook on the EU’s common foreign and security policy (emphasis added). Although the Treaty is careful to note that this shall not ‘prejudice the specific character of the security and defence policy of certain Member States’ (such as the six neutral or non-aligned EU members), the general remit of ‘all means’ could extend to various forms of security and defence assistance. This particular part of the Treaty is stronger than the Article 5 counterpart found in the 1949 Washington Treaty, which obliges NATO members to take ‘individually and in concert with the other Parties, such action as it deems necessary, including the use of armed force …’ (emphasis added). The only example of the invocation of Article 42(7), following the November 2015 Paris attacks, led rather surprisingly not to direct support for the security of France per se, but to the relief of French military commitments overseas so that French security assets could be concentrated on national security tasks following the bombings. Although there was no obligation on other Member States to provide military assistance, there was nevertheless a bilateral obligation upon the EU’s members rather than one that applies specifically to the EU context.12 With the sole and rather rarefied exception of Article 42(7) TEU, any obligation to provide capabilities to CSDP missions or operations should be considered as a general commitment but subject to decisions by the relevant national authorities. There is, in other words, no automatic right of release that would be required to constitute a ‘European army’. The other associated issue is who should provide capabilities. As has been observed, there are varying levels of capabilities and preparedness among the EU Member States and it is unlikely that all EU members will be able, or willing, to contribute. This was foreseen in the Treaties through the provision for flexible forms of CSDP engagement in missions and operations. An issue that has resurfaced in current debates is permanent structured cooperation (PESCO), which is open to those Member States ‘whose military capabilities fulfil higher criteria and which have made more binding commitments to one another in this area with a view to the most demanding missions’ (Art. 42(6) TEU). Article 46 TEU refers to a protocol on PESCO which lays down some of the requirements, albeit in somewhat non-specific terms (see Protocol 10 on PESCO). The emphasis is, however, on exclusivity with reference to those who proceed to develop their defence capacities ‘more intensively’, or to those who have the capacity to supply ‘targeted combat units for missions planned’ and those who will bring their defence apparatus ‘into line with each other as far as possible’ and ‘take concrete measures to enhance the availability, interoperability, flexibility and deployability of their forces’ (Protocol 10: Arts 1–2). PESCO has generated considerable interest as a way to boost the EU’s military capabilities post Brexit. I shall discuss this in more detail later. Article 42(5) TEU makes it possible to entrust the ‘execution of a task, within the Union framework, to a group of Member States in order to protect the Union’s values and serve its interests’. This rather pragmatic provision recognizes the disparities in capabilities between the Members and foresaw that it would be impossible to insist on every member contributing to a task, in much the same way that PESCO implicitly 12 See C Hillion and S Blockmans, ‘Europe’s Self-defence: Tous pour un et un pour tous?’ CEPS Commentary, 25 November 2015; European Parliament, ‘The EU’s mutual assistance clause’ (n 11).

Capabilities and CSDP 159 recognizes the desire of EU members to move at different speeds. How then are we to understand this provision, along with Article 46, when set against Article 42(3) TEU which obliges Member States to ‘make civilian and military capabilities available to the Union for the implementation of the common security and defence policy’? Article 42(4) TEU requires that decisions relating to CSDP shall be adopted by the Council acting unanimously. Such decisions will typically lay out the mandate and objectives of a CSDP mission, the general requirements, as well as details such as the operational headquarters and commander. Any associated force generation is, however, a separate process that relies in most cases upon capabilities that fall under national command structures and thus the relevant national decision-making structures (which may, in some cases, even include the need for an affirmative decision by the national parliament). In other cases, participation in a CSDP mission by a given Member State may hinge upon a prior UN Security Council resolution (this became a bone of contention in the case of the EU’s Rule of Law mission to Kosovo where the anticipated UN Security Council resolution never materialized). The political-level decision, which requires unanimity, does not therefore extend to an obligation on the Member States to provide the necessary military or civilian capabilities. This logic is consonant with the inclusion of PESCO and the ability to entrust tasks to groups of Member States in the Treaties since it was recognized that not only are there differences in capabilities between the Union’s members, but that not all may wish to be involved in contributing to a CSDP mission or operation since decisions on nearly all resources and personnel are made on a national basis. The role of the European Defence Agency (EDA) is a further indication of the restricted role that the EU can play in capabilities issues. The powers of the Agency are set out in Article 45 TEU and are generally advisory and, at best, rely upon persuasion. The general lack of authority of the European Court of Justice in CFSP and CSDP also means that the commitments in the articles mentioned above have to be viewed as essentially political in nature (for an in-depth analysis of this point see Chapter 4 in this volume). Security and defence priorities continue to be decided upon at the national level subject to strategic defence reviews that are largely uncoordinated between the Member States. As we shall see later, the EU’s members have (so far) been resistant to most of the arguments put forward by the EDA, which often appeal to economic rationale in a time of austerity. The arguments so far have suggested that there is no binding obligation for the Member States to provide capabilities, with the possible exceptions of Article 42(7) TEU and Article 222 TFEU. But in both of these cases the precise nature of any assistance, and thus capabilities, lies beyond the Treaties. Generally, the EU continues to rely upon the willingness of its members to make the necessary capabilities available, but this is subject to national determination and the aforementioned principle of a single set of forces. Having said this, there have been numerous efforts on the EU side to make the generation of the required forces and assets more predictable since the effectiveness of the Union will ultimately depend upon the application of the required expertise and capabilities at the right time and in the right place. Since the capability development processes for military and civilian operations and missions have developed in a rather distinct manner, they will be treated separately in the following

160 Research handbook on the EU’s common foreign and security policy sections, notwithstanding the many efforts to harmonize more closely the civilian and military aspects of crisis management.

3. MILITARY CAPABILITY DEVELOPMENT Until recently it was normal to speak of capabilities as falling under the EU’s Comprehensive Approach to external conflicts and crises.13 This has now been supplemented by an ‘integrated approach to conflicts’, stressing the need for coherent use of all of the instruments at the EU’s disposal.14 It is also framed by proposals for a ‘new level of ambition’ in the EU’s security and defence, as expressed in the 2016 Implementation Plan on Security and Defence (IPSD).15 Part of the IPSD involves using the full potential of the Lisbon Treaty (hence the renewed interest in PESCO and entrusting tasks to groups of Member States). In this document the EDA and the Member States have agreed to ‘specify and complement capability priorities based on the Level of Ambition and the EUGS, as part of the revision process of the Capability Development Plan’.16 The capability requirements therefore stem from strategic perspectives (the European Security Strategy of 2003, its update in 2008 and, more recently, the EUGS) as well as the ‘level of ambition’. The adoption of the Petersberg tasks, mentioned above, framed the initial level of ambition in terms of capacity. Following the St Malo Declaration, also mentioned above, the first of a number of ‘Headline Goals’ (HLGs) was established at the Helsinki European Council in 1999 and were supposed to be implemented by 2003 (they are therefore often referred to as HLG 2003). The aim set in Helsinki was described as follows: To develop European capabilities, Member States have set themselves the headline goal: by the year 2003, cooperating together voluntarily, they will be able to deploy rapidly and then sustain forces capable of the full range of Petersberg tasks as set out in the Amsterdam Treaty, including the most demanding, in operations up to corps level (up to 15 brigades or 50,000–60,000 persons). These forces should be militarily self-sustaining with the necessary command, control and intelligence capabilities, logistics, other combat support services and additionally, as appropriate, air and naval elements. Member States should be able to deploy in full at this level within 60 days, and within this to provide smaller rapid response elements available and deployable at very high readiness. They must be able to sustain such a deployment for at least one year. This will require an 13

Joint Communication to the European Parliament and the Council from the European Commission and the High Representative of the European Union for Foreign Affairs and Security Policy, ‘The EU’s comprehensive approach to external conflict and crises’ JOIN (2013) 30 Final. 14 European External Action Service, ‘Shared Vision, Common Action: A Stronger Europe. A Global Strategy for the European Union’s Foreign and Security Policy’ (28 June 2016) 9. 15 European Commission, ‘European Defence Action Plan: Towards a European Defence Fund’ (30 November 2016) 2–4. 16 Council of the European Union, ‘Implementation Plan on Security and Defence’ (n 3) 5.

Capabilities and CSDP 161 additional pool of deployable units (and supporting elements) at lower readiness to provide replacements for the initial forces.17

A year later a Capabilities Commitment Conference was held in Brussels where Member States took it upon themselves to make national contributions to the HLGs on a voluntary basis.18 But no sooner had the deadline arrived for the implementation of the HLG 2003 than the Council approved new HLGs in 2004 to support the recently adopted European Security Strategy (ESS). The December 2003 ESS called for a ‘more capable’ Europe, which included the exhortation to use pooled or shared assets to avoid duplication and overheads’.19 The military capabilities for the new HLGs were calculated on the basis of five illustrative scenarios ranging from separation of forces; stabilization, reconstruction and military advice to third countries; conflict prevention; evacuation operation; and assistance to humanitarian operations.20 The new HLGs envisaged 2010 as the adoption date. HLG 2010 opened the possibility of either a major operation or the ability to conduct a series of operations and missions of varying scope. The EDA was created in 2004 and its role was therefore included in the HLG 2010, with an emphasis on interoperability at the technical, procedural and conceptual levels. The Battlegroup concept was also woven into the new HLGs.21 The actual development of capabilities is carried out through three catalogues. The first is the Requirements Catalogue (RC), which is a compilation of the capabilities the EU would be likely to need, based on the five illustrative scenarios outlined above. These estimates were then fed into generic force packages and reference units, which formed the basis for the RC. It should, however, be noted that any capabilities in the catalogue ‘are voluntary and non-binding and cannot be used for Force Generation processes’.22 A Force Catalogue (FC) lists actual qualitative and quantitative capabilities which the Member States could make available to the EU on a case-by-case basis.23 The FC is regularly updated by the EU Military Committee (EUMC) and the EUMC Working Group/Headline Goal Task Force. The FC is then compared to the HLGs and in those capability areas where shortages are identified as ‘shortfalls’ a third 17 Council of the European Union, Annex iv of the Presidency Conclusions, Helsinki European Council, 10–11 December 1999 2–3 at . 18 Denmark did not participate, having obtained an opt-out on all defence-related provisions of the Treaties prior to its second referendum on the Maastricht Treaty. 19 ESS, ‘European Security Strategy: A Secure Europe in a better world’ (12 December 2003) 12. 20 Council of the European Union, ‘Development of European Military Capabilities’ (July 2009) at . 21 Council of the European Union, ‘Headline Goal 2010’ (17 May 2004) . 22 See P. van der Heijden, ‘Military Capability Development’ in CSDP Handbook: Missions and Operations (Federal Ministry of Defence and Sports, Republic of Austria 2015) 238. 23 The initial Force Catalogue has been updated to reflect not only new EU members, but also voluntary contributions by non-EU states.

162 Research handbook on the EU’s common foreign and security policy catalogue, the Progress Catalogue (PC), comes into play. The PC ‘identifies quantitative and qualitative military capability shortfalls on the basis of requirements set out in the Requirements Catalogue 2005 and the contributions compiled in the Force Catalogue 2007’.24, 25 An initial cross-referencing of the catalogues indicated that around 104 of 144 capabilities had been ‘filled’, leaving 40 or so shortfalls, but of these 21 were deemed ‘significant’.26 A European Capability Action Plan (ECAP) was launched in 2002 to address these shortfalls. Nineteen panels of national experts developed possible solutions in the main shortfall areas.27 Addressing the shortfalls was, however, reliant upon a ‘bottom-up’ approach where the Member States were expected to make voluntary national commitments.28 Some commitments might therefore depend upon Member States making available national capabilities that had not previously been offered. The more difficult problem was to address shortfalls where no capabilities exist at the European level. Although some shortfalls could be rectified with short-term solutions (such as leasing large transport aircraft), others were more likely to require major investment. ECAP had its limitations, stemming from its voluntary nature, often vague timelines, difficulties for the EU Military Staff to follow the deliberations and the ad hoc nature of the exercise.29 Responsibility for monitoring shortfalls was transferred to the EDA, as were some of the flaws inherent in ECAP. ECAP’s successor, the Capability Development Plan (CDP), was developed in close collaboration with the EDA, the EUMC and the Member States. CDP is built around four chapters addressing: short-term capability shortfall analysis against HLG requirements; long-term challenges and risks based on a Long-Term Vision 2025; the identification of potential cooperation at the national level; and lessons learned from CSDP operations and missions. It is therefore not a ‘plan’ as such, but a basic planning element showing the likely short- and longer-term capability needs. CDP is regularly updated at the biannual meetings of the EDA Steering Board in Defence Minister format, based upon longer-term strategic perspectives (2030 and beyond). It differs 24

Council of the EU, ‘Development of European Military Capabilities’ (n 20). It has become increasingly difficult for the external analyst to monitor progress on shortfalls since the EU stopped publishing its shortfalls in public in 2006 (presumably for well-founded security reasons). The basic shortfalls have, however, remained more or less consistent over the last decade. 26 AJK Shepherd, ‘EU Military Capabilities Development and the EDA: Ideas, Interests and Institutions’ in N Karampekios and I Oikonomou I (eds), The European Defence Agency: Arming Europe (Routledge 2015) 68. 27 These were: attack/support helicopters; nuclear, biological and chemical protection; unmanned aerial vehicles; medical role and protection role; special operations forces; suppression of enemy air defences; air-to-air refuelling; combat search and rescue; cruise missile/ precision guided munitions; theatre ballistic missile defences; deployable communications modules; headquarters; theatre surveillance and reconnaissance air picture; strategic image intelligence collection; early warning and distant detection strategic level; strategic air mobility/ outsize transport aircraft; roll-on-roll-off vessels. 28 Council of the European Union (2001), General Affairs, 2386th Council meeting, Press Release 13802/01, Brussels, 19–20 November, Section III. 29 See B Schmitt, European Capabilities Action Plan (EU Institute for Security Studies 2003). 25

Capabilities and CSDP 163 from ECAP in the sense that its primary focus is on the future development of research, technology, armaments and industry, which are at the centre of the EDA’s work. The CDP was able to incorporate the consequences of HLG 2010, estimates of capability requirements in 2025 (see below), plans and programmes announced by Member States and the ‘lessons learned’ from CSDP operations.30 Since a number of non-EU NATO members have been involved in CSDP missions or operations (such as Canada and Norway) complementarity between the EU and NATO capability plans was essential and led to the development of common Capability Codes and Statements, which apply in either context. Timelines for reporting and longer-term development are coordinated between the EU and NATO. The EDA also placed more emphasis on pooling and sharing of military capabilities, as well on the education and training of military staff. A 2010 Pooling and Sharing initiative, based upon a German-Swedish food for thought paper (also known as the Ghent initiative), was launched the following year with the adoption of a list of priorities.31 This, in turn, was followed by a ‘Code of Conduct on Pooling and Sharing’ in 2012 in an attempt to incorporate pooling and sharing into national planning and decision-making processes.32 Thereafter the Council regularly called for further development of military capabilities ‘for sustaining and enhancing CSDP’.33 The December 2013 European Council held its first thematic debate on defence since the entry into force of the Lisbon Treaty. The European Council identified three axes or priority areas: increasing the effectiveness, visibility and impact of CSDP; enhancing the development of capabilities; and strengthening Europe’s defence industry.34 With regard to the second priority, the European Council identified several capability developments in particular: + the development of Remotely Piloted Aircraft Systems (RPAS) in the 2020–2025 timeframe; a programme for the next-generation European Medium-Altitude Long Endurance MALE RPAS; a RPAS user’s community; synergies with the European Commission on regulation; and appropriate funding for RPAS activities; + air-to-air refuelling capacity: progress in establishing a Multi-Role Tanker Transport capacity, with synergies in certification, qualification, in-service support and training; + satellite communications: preparations for the next-generation Government Satellite Communication through close cooperation between Member States, the Commission and the European Space Agency; + cyber defence: the development of a roadmap and concrete training and exercises, improving civilian/military cooperation on the basis of the EU Cybersecurity Strategy as well as the protection of assets in EU missions and operations.35 30

Council of the EU, ‘Development of European Military Capabilities’ (n 20). The list currently concentrates on air-to-air refuelling, medical support, helicopter initiatives, counter improvised explosive devices and the European Air Transport Fleet. 32 European Defence Agency, ‘Code of Conduct on Pooling and Sharing’ (2012) at . 33 Foreign Affairs Council, 19 November 2012, Conclusions. 34 European Council, European Council Conclusions EUCO 217/13, 19–20 December 2013, 2. 35 ibid 5–6. 31

164 Research handbook on the EU’s common foreign and security policy The fact that many of the major capability programmes have a horizon of beyond 2025 tells us something about the research, development and manufacturing cycles of major defence assets. Nevertheless, the EDA’s provisional assessment is positive, with ‘good progress’ in the four key capability programmes.36 To summarize, military capability requirements are derived from the HLGs and more recently the EUGS and its ‘Level of Ambition’. The strategic level will, in turn, inform the illustrative scenarios and the strategic planning assumptions, which will then lead to the identification of shortfalls as input into the CDP. The EDA is the primary body with responsibility for working with the Member States to develop military capabilities, although progress thus far has depended heavily on voluntary actions and national calculations.

4. CIVILIAN CAPABILITY DEVELOPMENT The civilian aspects of crisis management developed after the initial emphasis on the military aspects, with the 2000 Feira European Council often being marked as the starting point. Despite this, most CSDP operations have been civilian in nature, as are the majority of the ongoing operations (which include police, rule of law, training and advice missions). As of 2015, the EU deployed 1,500 international experts on three continents for CSDP civilian missions.37 Unlike their military counterparts, which were able to draw upon a long history of collaboration prior to CSDP through the WEU, NATO or even the UN, much of the Union’s civilian crisis management capacities had to be built up from scratch. Most of the EU’s civilian crisis management experts are volunteers from the Member States who are usually available for a year at a time. As with military experts, the Member States have to make an often-difficult cost-benefit calculation about the relative merits and demerits of seconding a national expert to EU missions. Based upon evidence from the Annual CSDP Lessons from 2013–2015, the EU has been beset by shortages of adequately trained and experienced staff for its civilian missions, especially when it comes to niche specializations and senior experienced staff. The issue is compounded by different national procedures for assessing the range and quality of civilian personnel; an issue that will hopefully be addressed through Goalkeeper’s ‘Registrar’ module (see below). Further problems with availability have arisen due to legal, administrative and financial conditions applying to secondment to CSDP missions. Legal issues have arisen as a result of variations in contract terms for seconded personnel from one seconding agency to the other, while the financial complications arise from the fact that the contributing states bear all personnel-related costs for seconded personnel. There have also been administrative issues due to the differences between the EU’s members on training, coaching (before and after deployment) and reintegration into the domestic labour markets. Other more general challenges arise 36

European Defence Agency, Annual Report 2015, 3 at . NA Tovornik, ‘Civilian Capability Development’ in CSDP Handbook: Missions and Operations (n 22). 37

Capabilities and CSDP 165 from debates about what should constitute ‘best practice’ at the European level for civilian missions.38 Generating the required civilian capabilities and expertise for CSDP missions has proven challenging due, in large part, to the scarcity of trained expertise at the Member State level or, where it exists, the reluctance of members to spare such expertise. Capability development in the civilian sphere is broadly modelled on that in the military sphere, outlined above. As has been noted, this is logical enough given the prior development of the military aspects. The civilian aspects were first clearly enunciated in 2000 at the Feira European Council, which also adopted ‘priority areas for targets in civilian aspects of crisis management and of specific targets for civilian policy capabilities’.39 These priorities were listed in an annex to the Presidency Conclusions as police, strengthening the rule of law, strengthening civilian administration and civil protection. A further annex laid down specific targets for the development of police capabilities, which are sometimes referred to as the ‘Civilian Headline Goals’, whereby the Member States undertook to provide up to 5,000 police officers for international missions ‘across the range of conflict prevention and crisis management operations’ by 2003. In addition, 1,000 police officers should be deployable within 30 days. This necessitated the preidentification and training of a large pool of police staff. Their identification, ‘levels of expertise’ (DEL and expertise) were fed into a police database originally housed in the Council Secretariat. The ESS provided the political-level justification for the development of civilian crisis management capabilities, having noted the value added of developing ‘operations involving both military and civilian capabilities’.40 Subsequently two civilian HLGs established the level of ambition, tasks and thus capabilities. The first, adopted in 2008, had as the level of ambition approximately a dozen CSDP civilian missions of varying types, alongside a major mission, which could involve up to 3,000 experts for several years.41 The 2008 goals formulated the capability requirements around the four priority areas identified at Feira. But they also noted the importance of contributing to monitoring missions, providing support to Special Representatives as well as, inter alia, contributing to ‘activities such as security sector reform and support disarmament, demobilisation/reintegration processes’.42 The emphasis was not upon substitution of local forces or authorities, but to strengthen local institutions, as well as upon preventative activities. Since few crises are uni-dimensional, emphasis was also placed on the development of ‘integrated civilian crisis management packages’ and providing ‘an effective response across the full range of tasks in conflict prevention and crisis 38

European Parliament, ‘Civilian and military personnel in CSDP missions and operations’ PE 578.035 (February 2015); and ME Smith, Europe’s Common Security and Defence Policy: Capacity-Building, Experiential Learning and Institutional Change (CUP 2017) 128–173. 39 European Council, Conclusions of the Presidency, Santa Maria da Feira, 19–20 June 2003. 40 ESS, ‘European Security Strategy: A Secure Europe in a better world’ (12 December 2003) 11. 41 Council of the European Union, ‘Civilian Headline Goal 2008’ 15863/4 (7 December 2003). 42 ibid, 2.

166 Research handbook on the EU’s common foreign and security policy management’.43 The emphasis of the 2008 Headline Goals was very much upon personnel and establishing what and who might potentially be available. A further set of Civilian Headline Goals (2010) were approved by the Council in 2007. The main purpose of the second set was to help the EU ‘establish a clear illustrative framework for civilian capability planning and development, drawing on civilian European Security and Defence Policy (ESDP) mission experience and informed assessment of the expected and most urgent threats and challenges, and set capability targets accordingly’.44 A Civilian ESDP Capability Planning Process was agreed to in 2007 and work began the following year to review illustrative scenarios (as in the military case), assess required capabilities and survey civilian capabilities. A report on civilian preparedness would then be fed into national ministerial guidelines and civilian capability targets. As in the case of their military counterparts, a series of ‘conferences’ was envisaged from 2009 onwards to assess the state of play, monitor progress and guide future efforts. The 2010 Civilian Headline Goals were more sophisticated in the sense that the latter could benefit from the lessons learned from the first civilian missions, as well as being able to consider other important aspects such as the emerging synergies between the Area of Freedom, Security and Justice and CSDP and other relevant actors. In December 2010 the Council extended the implementation of the civilian (and military) HLGs beyond 2010, thus offering the opportunity to explore in greater depth the synergies mentioned above, the chance to build in ‘lessons learned’ (as the result of CSDP missions), national strategies and other changes in the strategic environment (such as the 2016 EUGS). The extension also permitted more effective use of support instruments (such as the ‘Goalkeeper’ software mentioned below) and other technological innovations (including those developed through or notified by the EDA). The Civilian HLGs, like their military counterparts, are scenario driven. Based on various scenarios, Member State capabilities will be identified, the availability of resources will be based on a questionnaire and the subsequent responses will then lead to the identification of shortfalls. This process is supported by the ‘Goalkeeper’ information hub in the EEAS, which is based upon the Civilian HLGs, mission outcomes, agreed concepts for the conduct of civilian CSDP missions and training offered to support civilian CSDP.45 The ‘Goalkeeper’ software is built around training (schoolmaster), standard job descriptions (head-hunter), rosters in the Member States and electronic responses (registrar) and EU Concepts and national measures (governor). Each of the components of ‘Goalkeeper’ has specific access protocols, with the intention of allowing appropriate access to each of its components. In the post-Lisbon context, the HLGs continued to provide the general level of ambition for capability development. In December 2011 the Council called for a multi-annual capability development approach, which led the following year in July to the Civilian Capability Development Plan (CCDP). The CCDP adopts a multi-annual 43

ibid, 3. Council of the European Union, ‘Civilian Headline Goal 2010’, approved by the Civilian Capabilities Improvement Conference and noted by the General Affairs and External Relations Council (19 November 2007). 45 EEAS, Goalkeeper, at . 44

Capabilities and CSDP 167 approach, incorporating the software environment, with the overall aim of making better use of existing resources as well as efforts to facilitate rapid deployment of personnel and assets. The objective of the CCDP, based on the HLGs, was to establish a ‘list of generic CSDP tasks’ based on abstractions from the (civilian) scenarios.46 A permanent CSDP warehouse for civilian assets was established in 2012 to support the launch of operations with around 200 personnel within 30 days of the approval of the Crisis Management Concept.47 The May 2015 Foreign Affairs Council noted that ‘[r]ecognising the continuous high demand for rapidly deployable, well trained civilian experts, including specialised profiles, the Council underlines the need to further improve and expedite the development of civilian capabilities’, while noting that the CCDP was still not fully implemented.48 It is, however, worth noting that the formulation of a number of generic civilian CSDP tasks in 2015 was an important step towards the implementation of the CCDP. In terms of the EEAS’s support structures the most relevant in this domain are the Committee for the Civilian Aspects of Crisis Management (CivCom), an advisory body composed of Member State representatives, and the Crisis Management and Planning Directorate (CMPD). CivCom was established in 2000 and it advises the Political and Security Committee (PSC), an ambassadorial-level committee, on the civilian aspects of crisis management. It also prepares planning documents for new missions, makes recommendations to the PSC and helps develop strategies for the civilian crisis management and capabilities. CMPD also reports to the PSC but was only created in 2009 and is also composed of representatives of the Member States. Among its mandates is the development of the EU’s civilian and military capabilities, with special attention being paid to the synergies between the two aspects of crisis management. It is worth noting several contrasts between the civilian and military capability processes. The former suffered from having far less information on availability of assets and levels of training and expertise than their military counterparts. As noted, the latter could benefit from decades of coordination and information stemming from collaboration through other organizations (such as the WEU or NATO). There was no such civilian crisis management collective memory to build upon from the outset. The lack of a ‘one-stop shop’ in the civilian arena was a further source of frustration although, unlike the military domain, civilian missions can at least draw upon a modest but permanent CSDP warehouse.49 The CSDP permanent warehouse for civilian assets was established in 2012 and became operational in June 2013. The warehouse is, however, limited to providing storage for strategic equipment for the rapid deployment of up to 200 personnel of newly launched missions within 30 days. It was used to provide equipment for EUBAM Libya. The European Parliament has since urged the 46

Council of the European Union, ‘Multi-annual Civilian Capability Development Plan: Action Lines for 2012–2013’ 12111/12 (6 July 2012). 47 Council of the European Union, ‘Council conclusions on Common Security and Defence Policy’, 3130th Foreign Affairs Council (1 December 2011) 4. 48 Council of the European Union (2015), ‘Council conclusions on CSDP’ 8971/15 (18 May 2015) 12. 49 ‘Council conclusions on Common Security and Defence Policy’ (n 47).

168 Research handbook on the EU’s common foreign and security policy expansion of its stocks and mission service under a new Shared Services Centre. Currently, The involvement of different national ministries, personnel and contacts complicates the collection of basic information and liaison. From the perspective of many of the national ministries, the primary purpose of personnel and resources that could be of potential interest for civilian CSDP missions remains, in the first place, subject to national priorities. The first sections addressed the nature of the obligations of the Member States to provide capabilities for CSDP operations and missions. At the political level, provision of the necessary capabilities has been driven by a series of ‘Headline Goals’. This led the European Council, with a carefully parsed statement, to claim at Laeken in 2001 that ‘[t]hrough the continuing development of ESDP, the strengthening of its capabilities, both civil and military, and the creation of the appropriate EU structures, the EU is now ready to conduct some crisis-management operations’ (emphasis added).50 By 2003 the European Council was ready to declare that ‘[b]ased on the Forces contributed to the Helsinki Force Catalogue 2003, the current military assessment of the EU military capabilities is that the EU now has operational capability across the full range of Petersberg tasks’.51 In retrospect, both declarations seem optimistic since the EU remains reliant upon the willingness of its members to provide the political will and capabilities, notwithstanding efforts to develop common understanding of capability shortfalls and development, or pooling and sharing, options. The voluntary nature of capability development is an obvious pitfall, which explains some of the enthusiasm for pooling and sharing, but this has only produced ‘marginal results’ that are ‘not yet an adequate response to the size of the problems’.52 The procedures to identify military capability shortfalls were developed reasonably rapidly, largely due to the legacy of similar exercises carried out in the WEU or NATO context. By way of contrast, there was little legacy to build upon when it came to civilian capabilities. But the issues encountered with civilian CSDP missions were more related to the availability of personnel, their training and compatibility of standards and procedures. Those relating to the military operations were in many ways more difficult to address, due, in large part, to the challenges associated with the transition from Cold War contingencies, which stressed defensive postures and equipment, to those that required expeditionary-type forces for out-of-area operations. Other issues, such as austerity measures across much of the EU following the financial crisis, also had an impact upon capabilities for CSDP missions and operations. The previous sections prompt obvious questions about the future. Will there be more initiatives at the EU level which will be frustrated by decisions made at the national level predicated upon calculations of national interest or perhaps even the lack of political will to back up the rhetoric? Or, are we at a watershed where the economics 50 Laeken Declaration, ‘Declaration on the Operational Capability of the Common European Security and Defence Policy’ (2001). 51 Council of the European Union, ‘Declaration on EU Military Capabilities’ Press Release 9379/03 (19–20 May 2003). 52 European Parliament, ‘State of play of the implementation of EDA’s pooling and sharing initiatives and its impact on the European Defence Industry’ PE 534988 (6 June 2015).

Capabilities and CSDP 169 associated with affording comprehensive national civilian and military capabilities has become simply unaffordable, thus making joint development, pooling and sharing necessities? If so, are there any initiatives that are likely to upset the trends outlined so far that will not only bolster the Union’s security capabilities but might also lead to a common defence? These are the questions that underpin the following section.

5. FUTURE CAPABILITY REQUIREMENTS This section will discuss future capability requirements based primarily on the EUGS and the IPSD, both of which were mentioned briefly in an earlier section. A third document, the Commission’s European Defence Action Plan, which, among other things, proposed the establishment of a European Defence Fund (EDF) to support investment in joint research and the joint development of defence equipment and technologies, SMEs and start-ups, and a strengthening of the Single Market for defence. The EDF has two windows, one for ‘research’ and the other for ‘capabilities’, although it is not yet entirely clear how complementarity between the two will be ensured. This, as Fiott has observed, marks ‘a radical shift in the way the EU thinks about and supports defence’.53 The EUGS makes a number of striking comments with potentially far-reaching implications for CSDP capabilities. To start with, it nicely encapsulates almost 35 years of debate on capability issues when it states that ‘Member States remain sovereign in their defence decisions: nevertheless, to acquire and maintain many of these capabilities, defence cooperation must become the norm’. The strategy then maintains that the EU needs to be strengthened as a security community: European security and defence efforts should enable the EU to act autonomously while also contributing to and undertaking actions in cooperation with NATO. A more credible European defence is essential also for the sake of a healthy transatlantic partnership with the United States.54

The EUGS introduces three important notions that may both have important implications for capabilities issues. First, it introduces the notion of autonomy under the general heading of Europe taking more responsibility for its own security. The EUGS acknowledges that [w]hile NATO exists to defend its members – most of which are European – from external attack, Europeans must be better equipped, trained and organised to contribute decisively to such collective efforts, as well as to act autonomously if and when necessary. An appropriate level of ambition and strategic autonomy is important for Europe’s ability to foster peace and safeguard security within and beyond its borders.55

53

D Fiott, ‘Promoting European Defence Cooperation and the Promise of Financial Incentives’ (Comment No. 16, Armament Industry Research Group 2017). 54 ‘Shared Vision, Common Action’ (n 14) 20. 55 ibid, 22.

170 Research handbook on the EU’s common foreign and security policy What exactly does this mean? As explained elsewhere in the EUGS, it implies that ‘European security and defence efforts should enable the EU to act autonomously while also contributing to and undertaking actions in cooperation with NATO’.56 Indeed, ‘strategic autonomy’ informs the underpinning level of ambition of the strategy. The phraseology employed is reminiscent of that employed in the 1998 St Malo Declaration, which also mentioned ‘the capacity for autonomous action’, largely to pander to French political wishes, but also referred to the ability of the EU to ‘take decisions and approve military action where the Alliance as a whole is not engaged’, presumably to placate British anxieties. The EUGS introduces a similar dilemma of what might happen when, for whatever reason, NATO (which usually means the US) does not come to the assistance of its European allies. A second term which is hard to ignore is ‘resilience’ since this word and ‘resilient’ are used no fewer than 41 times in the EUGS. This suggests that CSDP has a responsibility to bear for the security of its own citizens and societies and a wider (not entirely altruistic) duty to enhance different types of resilience, including those pertaining to a wide variety of security challenges.57 Part of resilience, which is seen as something that encompasses all individuals and the whole of society, involves intensifying EU–NATO cooperation in order to ‘bolster resilience as part of their work on countering hybrid threats’.58 In practical terms this involves, inter alia, greater coherence between the EU CDP and NATO’s Defence Planning Process (NDPP). While they are broadly compatible, there are also notable differences.59 The CDP is analytical but it is still based upon five illustrative scenarios developed as part of the 2003 HLG (which tend to reflect the EU’s experience in the Western Balkans at that time). This leaves significant gaps in the overall capability assessments. Traditionally, the bigger-ticket items, at the higher end of the combat spectrum, have been left to NATO’s NDPP. Unlike CDP, which is based on overall shortfalls, NDPP assigns national targets but it is worth noting that US capabilities often have a distorting effect (which, if unavailable for a non-Article 5 contingency, could leave NATO with significant shortfalls). The focus of NDPP also tends to be far shorter, with a horizon of around 4–5 years, in contrast to the longer-term horizon of CDP. The assignation of national targets by NDPP also moves the focus away from multilateral research, development and procurement, which tends to be the focus of CDP. The largely overlapping membership of NATO and the EU also poses the question of whether a common defence planning system might not be preferable to further align the two organizations and to avoid duplicative reporting. Since NDPP is older and in many ways better established, this might be the default choice. There is, however, no consensus on this point and both CDP and NDPP are at the mercy of the sovereign members of the respective organizations who make the ultimate decisions on which capabilities are required and what to develop. 56

ibid, 20. Joint Communication to the European Parliament and the Council, ‘A Strategic Approach to Resilience in the EU’s external action’ JOIN (2017) 21 Final. 58 ibid, 17. 59 European Parliament, ‘State of play of the implementation of EDA’s pooling and sharing initiatives and its impact on the European Defence Industry’ (n 52). 57

Capabilities and CSDP 171 More broadly, EU–NATO cooperation has been frequently discussed, even if it has been circumscribed in practice by differences over Cyprus. Most recently this resulted in a declaration designed to give ‘new impetus and substance’ to mutual relations.60 The capabilities implications are not entirely clear since they involve boosting abilities to counter hybrid threats (that is, hostile acts that may employ irregular warfare, terrorism, indiscriminate violence, criminal acts and propaganda) and bolstering ‘resilience’, including that of the Union’s neighbours. Cooperation at sea and on migration, cyber security and defence is also advocated. Defence capabilities should be ‘coherent, complementary and interoperable’ while defence industries, defence research and industrial cooperation will also need strengthening. If such declarations are to be taken seriously, the need for full-spectrum capabilities is evident, albeit subject to the inevitable political questions of who should do what (which will be complicated by uncertainties over US political will and commitments to the defence of NATO allies, as well as by the ambiguities surrounding the UK’s future relations with the EU). It is far from clear that the type of ‘step-change’ that this would necessitate is anywhere on the horizon since it would imply far more accountability and commitment on the part of the Member States.61 This brings us to the third notion, the frequent use of the term ‘defence’. This could be explained by the tendency in this and other EU documents to use security and defence as interchangeable terms. For instance, the EUGS exhorts members to take ‘greater responsibility for our security’ but then states that ‘[w]e must be ready and able to deter, respond to, and protect ourselves against external threats’ – deterrence and defence, in other words.62 It may also be explained by the historical reservations and political reluctance, at least until the EUGS, to clearly define the ‘D’ in CSDP. Yet the use of both terms in the EUGS has elements that go beyond random word choice. Take the example above where strategic autonomy implies the need for the EU to have the requisite capabilities for security (addressing challenges external to the EU) as well as defence (deterring or meeting challenges to the territory of the EU’s members). While the EUGS acknowledges that NATO is the ‘primary’ collective defence framework, the Alliance does not extend to all of the EU’s members. For these countries, whether they are then left to rely on national defence efforts, or if there is some implied EU responsibility for their collective defence, remains an open question. Care is taken to make it clear that the EU is not presenting itself as an alternative to NATO, but that ‘a more credible European defence is essential also for the sake of a healthy transatlantic partnership with the United States’.63 This resonates with the demands made by Donald Trump during his presidency campaign that US allies shoulder more equal shares of the collective defence burden, demands repeated by Mr Trump when he visited Brussels as President in May 2017. There are nevertheless 60

NATO Declaration, ‘Joint Declaration by the President of the European Council, the President of the European Commission and the Secretary-General of the North Atlantic Treaty Organization’ (8 July 2016). 61 M Drent, L Landman and D Zandee, ‘A New Strategy: Implications for CSDP’ (Clingendael 2016). 62 ‘Shared Vision, Common Action’ (n 14) 19. 63 ibid, 20.

172 Research handbook on the EU’s common foreign and security policy residual doubts surrounding the Trump administration’s observance of the collective defence guarantees under Article 5 of NATO’s founding treaty, which may make the quest for strategic autonomy in security and defence less of a choice for the EU and its members and more a matter of a necessity. This led Chancellor Merkel to observe two days after President Trump’s visit that ‘the times in which we can fully count on others are somewhat over’.64 Either way, the development of a genuine ‘D’ for CSDP will have widespread implications for capabilities and it will be a matter of diplomatic finesse whether this is perceived as an alternative to NATO or the emergence of a more credible European partner. The overall rise in defence expenditure in the EU during 2015–2017 may in part be a result of hectoring by the Trump administration but it is more likely to be the result of support for local arms industries, in the Italian case, and heightened threat perceptions of Russia, in Central Europe.65 The difference between ‘security’ and ‘defence’ could be dismissed as semantic quibbling but in capability terms the importance is crucial. Currently the types of capabilities required for missions outside the EU will tend to emphasize ‘expeditionary force’ military structures for low-intensity operations; in other words, forces that put more emphasis on the ability to move forces and their associated hardware from A to B, that are sustainable, networked and nimble. Those tasks, however, that are more associated with defence are likely to emphasize less mobility, more emphasis on air, land and sea (where applicable) interdiction, static defence installations, as well as quite different implications for logistics and intelligence support. As Luis Simón has observed, if the EU and its members are to rely less upon NATO, or the US specifically, it will imply serious thought and investment into anti-access and area denial (A2/AD) capabilities, including precision-guided munitions, short-range guided rockets, guided artillery and mortars, direct-energy or rail guns, stealth aircraft and greater numbers of submarines.66 The EU’s members are evidently torn on this issue, with some favouring ‘rebuilding the credibility of conventional deterrence and collective defence in light of a resurgent Russia’, while others ‘remain keen to engage in multi-faceted crisis management operations, and are prepared to ramp up efforts in defence capacity building to train and equip missions for which the EU seems to be a more natural fit’.67 As Schilde has aptly commented, a succession of crises surrounding the EU (in Georgia in 2008 and in Ukraine in 2014) has changed aspects of force structure in some, but not all, EU Member States with indications of a shift from the dominant 2000s pattern of investing in expeditionary capabilities (those of a discretionary nature for managing conflict or projecting power outside the continent) back 64

P McGee and G Parker, ‘Europe cannot rely on US and faces life without UK, says Merkel’ Financial Times (London, 28 May 2017). 65 See SIPRI Yearbook 2017, chapter 9, ‘Armaments, Disarmament and International Security, “Military Expenditure”’ (OUP 2018). 66 L Simón, ‘The “Third” Offset Strategy and Europe’s “Anti-access” Challenge’ (2016) 39(3) Journal of Strategic Studies 417. 67 B Giegrich, ‘European military capabilities and future conflict’ Geostrategy (5 March 2016) at .

Capabilities and CSDP 173 towards territorial capabilities (resources towards defending territory or projecting power within Europe).68

The threat from the south is less obvious, but the sale of sophisticated weapons systems by Russia to Syria, or the arming of terrorists or rebels in Libya or Yemen with precision-guided munitions give pause for thought. Until then direct territorial aggression against an EU member had been a rather abstract notion, especially since most of the Union’s members are also NATO members and could therefore count on US security guarantees. The advent of the Trump administration and allied concerns about the reliability of these guarantees has renewed debates about the desirable extent of European autonomy from NATO and its principal benefactor. The possible twin demands of providing for both expeditionary-style capabilities as well as those of a more defensive nature will not only lead to intense political debate (notably among the six neutral and non-aligned EU members), but it will also require far more serious thought about the type of capabilities required, including the question of who should provide for them. These are not entirely new concerns, since similar debates erupted in the 1950s and 1960s, but they have taken CSDP into strange terrain that considers not only defence but its adjunct deterrence. The very thought of a nuclear dimension to CFSP would until very recently have been dismissed as completely outré. Yet, although still rather unlikely, the very fact that the notion of a European nuclear deterrent was introduced by Roderich Kiesewetter, a senior member of Merkel’s CDU, into a debate in the Bundestag is a remarkable, if surreal, development.69 Suffice it to say that the possibility of a common defence policy and a common defence in the EU context has yet to be linked up to any purposive capabilities considerations. If the uncertainties introduced in the early stages of Donald Trump’s presidency can be overcome, it is possible that EU–NATO relations will follow the path suggested by the July 2016 EU-NATO declaration and that these debates will become less relevant (but are still worth having). If they do not, it is important to recognize that the types of investments and capabilities that the EU institutions are currently advocating are linked to the Petersberg tasks and are generally speaking not those associated with defence and deterrence. It could plausibly be argued that the capability aspects of security and defence are tangentially linked in the sense that A2/AD capabilities may be required to access operational theatres in Africa or elsewhere and that some capabilities may serve dual purposes, but the overall capability requirements for successful A2/AD in Europe would imply a radically different debate on capabilities and a change in investment and industrial perspectives from those currently under way. There is little indication that the 68 K Schilde, ‘European Military Capabilities: Enablers and Constraints on EU Power?’ (2016) Journal of Common Market Studies 1, 3 (pre-publication version DOI: 10.1111/ jcms.12444). 69 A Shalal, ‘German lawmaker says Europe must consider own nuclear deterrence plan’ Reuters (16 November 2016) at ; and B Kohler, ‘Das ganz und gar Undenkbare’ Frankfurter Allgemeine Zeitung (27 November 2016) at .

174 Research handbook on the EU’s common foreign and security policy EU’s members are ready for this debate, unless they are forced to this point by an increasingly recalcitrant Trump administration and a hollowed-out NATO. The ambiguous language regarding security and defence has also crept into the CSDP lexicon as a reaction to the UK ‘Brexit’ referendum result of 23 June 2016 after which the French and German Foreign Ministers advocated a European Security Compact encompassing ‘all aspects of security and defence dealt with at the European level’, promoting the EU as an independent actor.70 Their Italian counterpart, Paolo Gentiloni, called for a ‘Schengen for defence’ and the development of ‘the defence capabilities needed to be a prominent player on the international scene’.71 In both cases the foreign ministers recognized that core groups may wish to move ahead more swiftly with defence integration. An informal meeting of 27 EU defence ministers (minus the UK) in Bratislava in September led to a declaration and the adoption of a ‘roadmap’.72 The ‘roadmap’ was high on ambition but less specific when it came to capabilities, which was left to the IPSD. The IPSD, adopted by the Council in November 2016, makes the case for a new Capability Development Plan, a review of military requirements, revisiting the Feira priority areas for civilian missions in light of the changing security environment, and emphasizes the key role of the EDA.73 On the development of civilian capabilities, the plan recommends: + establishing a list of generic civilian CSDP tasks common to all missions and identifying the required capabilities; + ensuring more effective and rapid force generation, including by deploying specialized teams of experts; + strengthening capacities available for generic functions common to all missions; + improved training of mission staff, including through the new CSDP Training Policy. On the military side the plan wants the Member States to ‘collectively retain and further develop full-spectrum military land, air, space and maritime capabilities’ (emphasis added).74 The areas that need investment and collaborative approaches are those identified in the earlier HLGs, namely: + intelligence, surveillance and reconnaissance, RPAS, satellite communications, and autonomous access to space and permanent earth observation; + high-end military capabilities, including strategic enablers; + cyber and maritime security. 70 J-M Ayrault and F-W Steinmeier, ‘A Strong Europe in a World of Uncertainties’ (Voltaire Network, 27 June 2016) 3 at . 71 P Gentiloni, ‘Europe needs “Schengen for Defence”’ Politico.eu (15 September 2016). 72 Council of the European Union Presidency, ‘The Bratislava Declaration’ (16 September 2016). 73 Council of the European Union, ‘Implementation Plan on Security and Defence’ (n 3) 2–4. 74 ibid, 20.

Capabilities and CSDP 175 Precise military requirements, over and above the four priority areas identified by the European Council in 2013, will derive from the EUGS and the Level of Ambition, based on a review of the illustrative scenarios and strategic planning assumptions. The need to deepen defence cooperation is noted due to the rising cost of defence technology and the need to retain full-spectrum armed forces. Yet, as the IPSD notes, ‘80% of defence investment in Europe is still spent nationally and our collective output need to be increased substantially’.75 There remains the question of why so little seems to have changed since the EDA was founded in 2004 in terms of changes in defence expenditure habits and procurement of big-ticket items. There is the hope though that the injection of funding, in the form of a proposed EDF, may redirect defence expenditure towards joint research and the joint development of defence equipment and technologies.76 There is also the possibility that funding for the EDA may also increase since British objections to increasing funding for the agency will disappear. Recent progress has been made on building the EU’s capacity to respond to civilian and military contingencies through, for example, the introduction of a new Military Planning and Conduct Capability (MPCC) and a Joint Support Coordination Cell to strengthen civilian–military coordination.77 Without dismissing the potential benefits of streamlining and various other forms of institutional plumbing, it is evident that any real movement on the underlying capability issues will be far more difficult and less amenable to quick fixes. A Coordinated Annual Review on Defence (CARD), as advocated in the IPSD, is a good place to start, since it would allow coordination between EU (EDAP and the revised CAP), NATO (NDPP) and national planning processes and capability requirements. CARD could then ‘provide an overarching assessment on capability-related issues contributing to political guidance by the Council’.78 CARD has been promoted as a ‘more structured way to deliver the key capabilities needed in Europe, based on greater transparency, political visibility and commitment from Member States, while avoiding any unnecessary additional administrative effort by Member States and EU institutions’.79 CARD could also be of importance in terms of coherence with NATO’s NDPP since it would facilitate information on the contributions provided to NATO through NDPP (or the Defense Investment Pledge or the Partnership for Peace Planning and Review Process) which is currently not automatically released to the EDA.80 CARD could also complement the activities of the NATO-EU Capability Group, which was established in May 2003 to reinforce the respective capability development efforts. The Group, which comprises officials from the EDA and NATO, identifies and attempts to address common capability shortfalls, while promoting ‘Smart Defence’ in the NATO context and the EU’s Pooling and Sharing Initiative. However, it is worth noting that CARD relies entirely upon voluntary efforts in ‘full respect of Member State’s 75

ibid, 21. See European Commission, ‘European Defence Action Plan’ (n 15). 77 Council of the European Union, ‘Council conclusions on progress in implementing the EU Global Strategy in the area of Security and Defence’ Press Release 110/17 (6 March 2017). 78 ibid. 79 ibid, para 10. 80 Council of the European Union, ‘Council conclusions on Security and Defence in the context of the EU Global Strategy’ 9178/17 (18 May 2017) 13. 76

176 Research handbook on the EU’s common foreign and security policy prerogatives and commitments in defence, including, where it applies, in collective defence, and their defence planning processes, and taking into account external threats and security’.81 The EDA would support the biannual meeting of defence ministers through assessments and supporting analysis (they would, in essence, play the role of a secretariat). Progress on the civilian capability aspects should, in theory, be slightly easier with the emergence of ‘standing capacity, pre-configured specialised teams of experts and contingents of police and/or other professional categories …’.82 Even here though, tensions between the internal security demands on Member State human resources and budgets will be an important determinant of the willingness to create any standing capacity and other forms of investment. (The experience of the EU’s Battlegroups on the military side, which, having been created have never been used, offers a salutary lesson on the creation of standing capacities.) The extent of any capability development in the civilian or military realms will also depend upon wider political developments. A Commission follow-up to their ‘White Paper’ on the future of Europe, addressing security and defence, presents three possible scenarios.83 Each has implications for capabilities. First, a continuation of voluntary cooperation with no binding common direction for security and defence. While this scenario envisages greater exchange of information on external threats between the Member States, defence cooperation would be driven in a ‘bottom-up’ manner by economic and technological drivers. This could result in reliance on voluntary national contributions ‘leading to insufficient national cooperation in critical areas such as high-end capabilities’.84 A slightly rosier scenario sees the Member States move towards shared security and defence. Under this, ‘[n]ational defence planning would become far more aligned, facilitating Member States’ cooperation on the acquisition and maintenance of capabilities, thus improving interoperability’.85 The EDF would also come into its own by facilitating the development of multinational capabilities, supported by joint planning, command and logistics at the EU level. Finally, a move towards common security and defence (a Security and Defence Union) would involve fully integrated defence forces which are pre-positioned and permanently available for rapid deployment on behalf of the Union. Defence planning would be fully synchronized and national capability development would take into account European priorities. Capabilities in such areas as ‘space, air and maritime surveillance, communications, strategic airlift and cyber would be commonly procured by the Member States’ with the support of the EDF. This would be underpinned by

81

European Commission, ‘Reflection Paper on the Future of European Defence’ (n 4), para 11. 82 Council of the European Union, ‘Council conclusions on progress in implementing the EU Global Strategy in the area of Security and Defence’ Press Release 110/17 (6 March 2017). 83 ibid. 84 ibid, 13. 85 European Commission, ‘Reflection Paper on the Future of European Defence’ (n 4) 14.

Capabilities and CSDP 177 a genuine European defence market and a dedicated European Defence Research Agency.86

6. PESCO TO THE RESCUE? It is too soon to ascertain which of the three scenarios might transpire but there has never been so much political momentum behind European security and defence. The gap between political vision and actual capabilities remains wide but there is at least now acknowledgement of this fact. The Council’s conclusions of 18 May 2017 noted that there remains the need to ‘substantially update civilian capability development in order to address both identified new challenges and persisting gaps in the originally agreed capabilities’.87 PESCO has gained momentum as the vehicle to develop capabilities which will remain owned and operated by the Member States and which will remain a ‘single set of forces’ that can be employed in other frameworks, like NATO and the UN.88 The frequency of its mention, in the EU Global Strategy, in the EU-NATO declaration and the EDAP, all point to the political weight now attached to PESCO as a core means of addressing the EU’s capabilities and strategic shortcomings. The debates surrounding the extent to which PESCO would be exclusive or inclusive were eventually answered by the European Council, which states that it would be ‘inclusive and ambitious’.89 The European Council acknowledged that any common criteria and commitments that include the most demanding missions will apply to ‘those Member States which are in a position’ to enter into binding commitments based upon agreed criteria.90 Twenty-three EU members signed the joint notification on PESCO in November 2017 and they were soon joined by Portugal and Ireland (leaving only Malta, Denmark and the UK outside). Although PESCO is an intergovernmental programme, the European Commission saw the role of the EDF and the European Defence Industrial Development Programme as a way not only of supporting PESCO, but also of moving the EU and its members towards a European Defence Union. In a further significant development with implications for capabilities, the European Council also advocated that the deployment costs of Battlegroups, which have largely fallen to the Member States participating in the six-monthly rotating Battlegroup, should be borne as a common cost under the Athena mechanism on a permanent basis. This could also be a positive development but it skirts significant questions about the size and utility of the Battlegroups themselves, which have been on standby since 2005 but never used. 86

Council of the European Union, ‘Council conclusions on progress in implementing the EU Global Strategy in the area of Security and Defence’ Press Release 110/17 (6 March 2017) 15. 87 Council of the European Union, ‘Council conclusions on Security and Defence in the context of the EU Global Strategy’ (n 80). 88 ibid, 11. 89 European Council, ‘European Council conclusions on security and defence’ Press Release 403/17 (22 June 2017). 90 ibid.

178 Research handbook on the EU’s common foreign and security policy In order to be workable, PESCO will have to have stability in the form of either permanent or long-term arrangements and it will have to be linked to capability commitments including joint development and procurement, which, as we have seen, is the Commission’s preference. Aside from these important practical stipulations, some additional issues will need to be answered if PESCO is to fulfil its potential. Three issues stand out. First, the incentives for joint procurement, pooling and sharing are probably highest among the main defence industrial powers in the EU but any commitments via PESCO would inevitably create the expectation of returns in the defence industrial sector. The question arises whether PESCO participants would be expected to assume the main burden when it comes to RPAS, air-to-air refuelling, strategic surveillance and satellite imagery, as common assets that could feasibly be called upon by not only PESCO countries, but also those with lower capabilities. If so, the idea of some kind of return on investment, quite aside from the security provided, would be bound to surface.91 It could of course be argued that CSDP missions and operations have tended to be heavily influenced by the interests of the larger Member States and that PESCO would merely formalize what is already evident. But it is unclear whether a club of those who fulfil ‘higher criteria’ might not discourage those who do not meet the criteria for participation. It may even encourage free-rider behaviour and the eruption of intra-EU burden-sharing debates, which have been a feature of transatlantic security relations for almost five decades. Second, PESCO would have to be linked with existing initiatives and EU institutions. Fortunately, there is ample scope for the EDA to take a leading role in the capabilities dimensions of PESCO and other plans mentioned above, such as CARDS, EDAP, the CDP and the EDF, and the envisaged European Defence Research Programme could also be incorporated. The necessary linkages between PESCO, the High Representative, the EUMC and EU Military Staff would also have to be developed. Similar links would have to be established with NATO’s NDPP as well as between the respective military staffs. PESCO is, however, more ambiguous when it comes to civilian capabilities since it was conceptually designed with military missions in mind. It is therefore unclear how PESCO relates to civilian capabilities and, if included, whether the ‘higher criteria’ would be applied to different types of capabilities (this may be especially relevant if a Member State might qualify on the grounds of civilian criteria but not the military ones). Finally, PESCO mentions ‘defence’ at frequent intervals although, in line with the preceding section, there is a tendency to use security and defence as interchangeable terms. Protocol 10 on PESCO, attached to the Final Act of the Lisbon Treaty, clearly has in mind the Petersberg tasks (reference is made to Article 43 TEU) and not those of collective defence. If the current debates take the EU towards a common defence policy and common defence it is not entirely clear if PESCO reinforces or detracts from the ‘three musketeers’ principle that underpins any common defence (to paraphrase, ‘Some 91 A Marrone, N Pirozzi and P Sartori, PESCO: An Ace in the Hand for European Defence (Instituto Affari Internazionali 2017).

Capabilities and CSDP 179 for one, and one for some’ is counter to the original spirit). Certainly, potential resentment of free riders on the Petersberg tasks could sour the notion of a common EU defence.

7. CSDP CAPABILITIES AND BREXIT Little is known about the implications of Brexit for the ongoing CSDP capability discussions. One analyst has argued that Brexit will make little difference since the UK has not been particularly involved in CSDP for the last decade or so, with the important exception of the ongoing anti-piracy operations in the Indian Ocean and Gulf of Aden. He also observed that CSDP was not even mentioned in the UK’s 2015 Strategic Defence and Security Review.92 At the time of writing the nature of any agreement between the EU and the UK on security and defence cooperation is difficult to determine, beyond its presence as a potential bargaining chip (which logically implies that UK cooperation in security and defence could be used for potentially important concessions, but it also implies that the denial of such cooperation could be used punitively). Paradoxically, the departure of one of the EU’s most capable security and defence partners has not led to despondency or death knells. The UK and France account for around 40 per cent of public defence investments in the EU and the UK’s military expenditure is (just) above the 2 per cent of GDP threshold established by NATO in 2014 (in 2017 Estonia, Greece and Poland also met this basic target). The UK, alongside France, has the largest range of combat means, including high-intensity capabilities.93 Serious discussion about CSDP without the UK has been overshadowed by the charges from the Trump administration that Germany, Italy, Spain and other allies do not meet the basic GDP target, although such claims are often countered by the observation that the 2 per cent target is largely fictitious, that defence expenditure is actually rising and that the European allies contribute substantially to peace and security in other ways. More damagingly, it has sparked a debate about the reliability of American security guarantees and their apparently transactional nature at the same time as one of the EU’s most capable military powers prepares to leave.94 Brexit may imply that the EU will lose access to capabilities but it could also mean that the considerable experience of UK civilian and defence personnel would be missed. Operation Atalanta’s headquarters at Northwood will have to be relocated. Recent UK investments in big-ticket items (including two aircraft carriers and fighters) could also have been useful for more distance missions or operations and would have decreased over-reliance on the US. In the event that some kind of association between the EU and the UK is forthcoming, perhaps along the lines of the Framework 92 R Whitman, ‘Defence on the Brexit frontline’ The UK in a Changing Europe (26 February 2016) at . 93 G Faleg, ‘The Implications of Brexit for the EU’s Common Security and Defence Policy’ CEPS Commentary (26 July 2016). 94 A Rettman, ‘Merkel urges EU to take care of own security’ EU Observer (13 January 2017).

180 Research handbook on the EU’s common foreign and security policy Participation Agreements that currently exist with over 15 non-EU states, the dilemmas for CSDP capacities will not be as sharp as they might be. If there is no such accommodation, much of the capacity burden will fall upon the larger EU members, notably France and Germany. It is unclear whether either is willing to shoulder the responsibilities for the new Level of Ambition outlined in the EUGS at the very time when a major military power and co-founder of CSDP is leaving. The UK also has extensive bilateral military ties with, for example, France and the Netherlands, and these too may have to be reassessed. The UK could conceivably contribute to missions or operations as a third party on an ad hoc basis and join industrial defence cooperation programmes. Intelligence cooperation may pose greater hurdles but would presumably be on a reciprocal basis. President Emmanuel Macron’s suggestions for a European Intervention Initiative (EII) made in September 2017 may be one way of involving the UK through the Anglo-French Combined Joint Expeditionary Force. But EEI lacks specificity and it is unclear whether it is designed as an alternative to PESCO (based on the French belief that PESCO concentrates on capabilities to the detriment of important issues of common strategic outlook and culture). If EII is extended to other critical allies, such as Germany, Italy and Spain, as well as non-EU European NATO members such as Norway, it may provide a useful way of involving the UK in European security post Brexit. Fears that Brexit might heighten sensitivities about duplication of NATO assets with a UK in the Alliance, but out of the EU, can be ameliorated by the enhancement of cooperation between the organizations, as was suggested at the July 2016 Declaration on EU–NATO Cooperation. This argued for closer cooperation on hybrid threats, early detection, information sharing and strategic communication. It also advocated broadened maritime cooperation, including on migration. Cyber security is also likely to be another area of growing cooperation.95 In so far as capabilities are concerned, pledges were made to increase interoperability and complementarity with, wherever possible, multilateral projects.

8. CONCLUSIONS Capability issues have generated a modest but specialized literature, much of it stemming from official sources or policy-oriented think tanks. It has not generally penetrated the mainstream academic literature other than at a rather superficial level.96 But it is likely to become an area of more academic scrutiny as the high-level political pressure to address CSDP shortfalls escalates. Recent initiatives stemming from the EUGS and Brexit suggest, at least prima facie, renewed political determination to provide for Europe’s security and defence in an uncertain world. A flurry of national initiatives has ensured that the issue has remained at the top of the agenda. The IPSD is the first of its kind and more will follow to implement other aspects of the EUGS. There are, of course, risks associated with the presence of security and defence at the forefront of efforts to reinvigorate the European project. The most obvious is that it is 95 96

NATO Declaration (n 60). For an exception, see Shepherd (n 26).

Capabilities and CSDP 181 up to the Member States to deliver the capabilities to underpin the new Level of Ambition contained in the strategy. If the 27, without the UK’s sizeable military capabilities and know-how, fail to do so it will not only impact on Europe’s security and defence, but perhaps also upon the progression of European integration in other areas. It is also striking how little has actually changed since the first ‘headline goals’ were adopted. With a few exceptions, Member States have not spent more on defence and most calls for more pooling, sharing and joint development have gone unheeded. Like diplomacy, issues of national security and defence are often at the heart of notions of sovereignty and, in many cases, national pride. The necessary changes of mindset are not evident yet, unless of course an external shock nudges the collective consciousness of the Member States in an appropriate direction (which may be prompted by ongoing concerns about US defence guarantees to its allies or further acts of Russian military aggression) or perhaps by internal inducements (such as the EDF) that change the nature and content of national defence planning and expenditure. In any event, the first concrete attempt to implement the EUGS happens to be in the area of security and defence. It is for this reason that capabilities have become a (or even the) critical issue, not just for Europe’s security and defence, but for the future of the European project itself.

9. Non-proliferation of weapons of mass destruction Eileen Denza

1. INTRODUCTION Efforts by the European Union to coordinate policies of Member States and to take positive action to control the proliferation of weapons of mass destruction (WMD) are long standing, multifaceted and for the most part conducted without high-profile successes or failures. The primary purpose of the work is to reinforce the international agreements which control non-proliferation of WMD – both generally and by concentrating on states of significant concern. Non-proliferation work is integrated with many other aspects of the Union’s CFSP, given its importance in the context of security policy and conflict prevention activities conducted by the Union.

2. THE INTERNATIONAL FRAMEWORK The most important of the three key multilateral agreements which form the background to EU policy is the 1968 Treaty on Non-Proliferation of Nuclear Weapons (NPT).1 The nuclear weapon states (NWS) were at that time limited to the five Permanent Members of the Security Council and Article I of the Treaty imposed a total ban on the direct or indirect transfer of ownership and control of their nuclear weapons. The non-nuclear-weapon states (NNWS) undertook under in Article II ‘not to manufacture or otherwise acquire nuclear weapons or other nuclear explosive devices; and not to seek or receive any assistance in the manufacture of nuclear weapons or other nuclear explosive devices’. Verification of this obligation and control of any diversion was to be guaranteed by technical Comprehensive Safeguards Agreements concluded between each NNWS and the International Atomic Energy Agency (IAEA) – a UN Specialized Agency – and in the ensuing years most NNWS concluded and implemented safeguards agreements. Every year the IAEA issues a Safeguards Implementation Report with its conclusions for each NNWS as to the effectiveness of these agreements in ensuring that nuclear material is in use for peaceful purposes or otherwise accounted for.2 Verification is further enhanced by the conclusion between all NWS, many NNWS and the IAEA of 1 729 UN Treaty Series 161; UK Treaty Series No. 88 (1970). The Treaty was opened for signature and ratification in London, Moscow and Washington, in the hope of attracting universal participation, including entities such as North Korea and Taiwan not then generally recognized as states. 2 See ‘IAEA Safeguards Overview’, available at www.iaea.org/publications/factsheets/iaeasafeguards-overview.

182

Non-proliferation of weapons of mass destruction 183 additional bilateral protocols which permit IAEA inspectors not only to verify the use of declared nuclear material, but also to investigate possible unauthorized activities or undeclared nuclear material. In exchange for these commitments to abstention or renunciation, the NNWS are entitled under the NPT to an ‘inalienable right … to develop research, production and use of nuclear energy for peaceful purposes without discrimination’ and there is to be full exchange among all parties of equipment, materials and information for peaceful use of nuclear energy. The NWS, moreover, are required to pursue negotiations in good faith on effective measures towards nuclear disarmament and towards a treaty on complete nuclear disarmament under effective international control. These reciprocal commitments have been described as a ‘grand bargain’ based on balanced mutuality.3 Chemical weapons are very much older than nuclear weapons and can be traced back to the use of poisoned arrows by Neolithic peoples and to records of the fifth century BC Peloponnesian wars between Athens and Sparta. They were extensively used by both sides during the First World War notwithstanding early attempts to prohibit their use in warfare through the Hague Regulations of 1899 and 1907. Widespread experience and descriptions of the resulting horrors led to the conclusion in 1925 of the Geneva Protocol for the Prohibition of the Use in War of Asphyxiating, Poisonous or other Gases, and of Bacteriological Methods of Warfare – an instrument widely accepted among states but dependent on reciprocity and containing other defects of clarity and enforcement.4 A multilateral convention which bans outright the possession and use of chemical weapons, remedies the defects in earlier treaties and contains effective supervisory and enforcement mechanisms is the 1993 Convention on the Prohibition of the Development, Production, Stockpiling and Use of Chemical Weapons and on their Destruction (CWC).5 The Convention contains a precise definition of the term ‘chemical weapon’, a detailed non-exhaustive list of materials covered in an Annex on Chemicals, and a description of permitted purposes for the use of chemical weapons – for example industrial and pharmaceutical purposes, protection and law enforcement. There is an Annex on Implementation and Verification which establishes procedures for checking compliance and the Organisation for the Prohibition of Chemical Weapons (OPCW) is established and entrusted with the task of monitoring and assisting implementation. The third of the key international treaties is the 1972 Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on Their Destruction6 – also elaborating the prohibition in the 1925 Geneva Protocol but concluded earlier when it was realized that a multilateral prohibition on possession could be achieved more easily in the field of biological weapons than in that of chemical weapons. Parties to the Convention undertake ‘never 3

DH Joyner, International Law and the Proliferation of Weapons of Mass Destruction (OUP 2009) ch 1: ‘The Nuclear Non-proliferation Regime’, esp. 8–9. 4 LNTS vol. 94, 66. 140 states are now parties. 5 32 ILM 800. 192 states are now parties – almost all states are entitled to join. For detailed analysis of the Convention and practice in applying it, see W Krutsch, E Myjer and R Trapp (eds), The Chemical Weapons Convention (OUP 2014). 6 1015 UNTS 163. 175 states are now parties.

184 Research handbook on the EU’s common foreign and security policy in any circumstances to develop, produce or stockpile or otherwise acquire or retain microbial or other biological agents, or toxins whatever their origin or method of production, of types and in quantities that have no justification for prophylactic, protective or other peaceful purposes …’ and to prohibit ‘weapons, equipment or means of delivery designed to use such agents or toxins for hostile purposes or in armed conflict’. The purpose-related nature of the definition makes it difficult to apply given the high proportion of biological agents which have multiple purposes, and in contrast to the Chemical Weapons Convention there is no list of prohibited substances or methods of delivery. As the Convention was concluded during the Cold War, the verification and dispute resolution provisions are also markedly weaker than in the case of the later Convention.7 In addition to those three Conventions, other multilateral international agreements have been concluded or are under active negotiation – the most important being the 1996 Comprehensive Nuclear Test Ban Treaty (CTBT).8 This Treaty would complement the 1963 Treaty Banning Nuclear Weapon States in the Atmosphere, in Outer Space and Under Water9 by seeking to achieve the discontinuance of all test explosions of nuclear weapons, but it cannot come into force without ratification by states listed in Annex II, of which the most significant is the United States. These Treaties are supplemented by UN Security Council Resolution 1540,10 adopted in 2004 under Chapter VII of the Charter so as impose legal obligations on all UN Members, but expressly excluding authorization of the use of force. Its main focus is to prohibit States Parties from providing support to non-state actors attempting to manufacture, possess, transport or use WMD or their means of delivery, to require the parties to adopt domestic laws prohibiting support for such activities by non-state actors and to establish effective national border controls and laws to ensure physical protection of WMD-related materials. The use of a Security Council Resolution for what is essentially a multilateral legislative instrument is almost unprecedented, but was justified by the gaps in the treaty-based system and by the urgency of the problem, particularly the risk of terrorists acquiring or using WMD.11

3. EU CONTRIBUTIONS Although Euratom has extensive rules relating to the peaceful use of nuclear energy, neither the European Communities nor now the EU have internal rules regarding the possession of nuclear weapons or on nuclear disarmament. The Union was from the earliest years of European Political Cooperation intensely concerned with nonproliferation of nuclear weapons in particular, but despite the shared objectives among 7 8 9 10 11

2004.

Joyner (n 3) ch 2: ‘The Chemical and Biological Weapons Non-proliferation Régimes’. Cm 3665. UKTS No. 3 (1964), Cmnd. 2245. S/RES/1540 (2004). GH Oosthuizen and ES Wilmshurst, Chatham House Briefing Paper 04/01, September

Non-proliferation of weapons of mass destruction 185 the Member States there was always difficulty in formulating a common position on the interrelated issues involved between the UK and France as NWS and the others as NNWS. A key objective of the Union in the earliest years of CFSP was to encourage and support the newly independent former Soviet republics which held nuclear weapons on their territories (Belarus, Kazakhstan and Ukraine) to transfer these weapons to Russia to be decommissioned and decontaminated and then to accede to the NPT as NNWS. In return, the three post-Soviet states received security assurances from three nuclear powers – Russia, the US and the UK – in the form of three identical political agreements signed at the OSCE conference in Budapest in December 1994 (the ‘Budapest Memorandum’). China and France gave somewhat weaker individual assurances in separate documents. An extensive and coordinated lobbying exercise was carried out by the Member States using agreed arguments and incentives, which included the promise of full integration into the international community and closer relations with the EU. By 2000 it could be said that the EU had contributed to the achievement of this objective in that all three states had decommissioned their nuclear weapons or surrendered them to Russia and had acceded to the NPT as NNWS – a success from the early years of CFSP which could not have been confidently foreseen following the disintegration of the Soviet Union but which received little recognition.12 When Russian armed forces violated Ukrainian sovereignty and territorial integrity in March 2014, the EU institutions were quick to condemn these actions as clear breaches of the UN Charter and the OSCE Helsinki Final Act, as well as of Russia’s specific commitments to respect Ukraine’s sovereignty and territorial integrity under the Budapest Memorandum of 1994.13 At the time of writing the Council of the EU maintains a non-recognition policy of Russia’s annexation of Crimea, supported by CFSP decisions and sanctions regulations.14 Within the framework of the EU Common Strategy on Russia, the Council established a Cooperation Programme financed and administered by the Commission for non-proliferation and disarmament in Russia. The Programme assisted Russian efforts towards secure and environmentally sound dismantlement of infrastructure and equipment linked to WMD.15 Current actions continuing this support include assistance

12

For detail, see E Denza, The Intergovernmental Pillars of the European Union (OUP 2002) 98–100. 13 See European Commission and High Representative of the Union for Foreign Affairs and Security Policy, ‘Implementation of the European Neighbourhood Policy in Ukraine: Progress in 2013 and recommendations for action’(Joint Staff Working Document) SWD (2014) 96 final 2; and European Parliament resolution of 17 July 2014 on Ukraine (2014/2717(RSP) [2016] OJ C224). 14 Council Decision 2014/386/CFSP of 23 June 2014 concerning restrictive measures in response to the illegal annexation of Crimea and Sevastopol, and Council Regulation (EU) No. 692/2014 of 23 June 2014 concerning restrictive measures in response to the illegal annexation of Crimea and Sevastopol, both in OJ 2014 L 183 (as amended). 15 Joint Action 2003/472/CFSP of 24 June 2003, OJ 2003 L 157/69; Council Decision 2003/874 of 8 December 2003, OJ 2003 L 326/49; K Zwolski, ‘The External Dimension of the EU’s Non-proliferation Policy: Overcoming Inter-institutional Competition’ (2011) 16 European Foreign Affairs Review 325, 330 and 334.

186 Research handbook on the EU’s common foreign and security policy to a Russian non-profit organization to enhance expertise on export control of dual-use items, and assistance to Georgia with border monitoring and detection of exports.16 In December 2003 the European Council endorsed two related documents prepared by the then High Representative for the CFSP, Javier Solana – a European Security Strategy and an Action Plan for the Implementation of the Basic Principles for an EU Strategy against Proliferation of Weapons of Mass Destruction.17 Both strategies were formulated in response to the failure of the EU to articulate a common policy in regard to the invasion of Iraq and also to the US National Security Strategy of 2002 – widely seen as articulating the Bush doctrine authorizing pre-emptive war against a range of threats. While recognizing the need to tackle the sources of the desire for WMD such as instability and conflict situations, the European Strategy emphasized practical steps which the EU should take. These included increasing participation in, and strengthening compliance and verification of, multilateral treaty regimes, enforcing export controls, interdicting international movement of prohibited materials, improving safeguards of WMD-related material and interdicting their international movement, and integrating these objectives into EU negotiations with non-member states. In succeeding years, all these methods have been systematically applied using the various tools – legal instruments and financial resources – offered by the CFSP. Every six months, the Council issues a Progress Report on the Implementation of the Strategy.18 Five years later the Council supplemented the Strategy by adopting Conclusions on new lines for action by the EU on combating the proliferation of WMD and their delivery systems.19 This noted the emerging risk of WMD being developed and used by terrorists. It sought to raise awareness of the threats among scientific and academic circles, to make greater use of consular cooperation and to assist non-member states to improve their export control procedures. 3.1 Successes with High-risk States – Libya and Iran In addition to the assistance to the former Soviet republics mentioned above, it can be said that the EU has made substantial contributions to non-proliferation objectives in the cases of Libya and Iran. In the case of Libya, the EC and then the EU from the mid-1980s applied sanctions that were more extensive than required under the Security Council Resolution in solidarity with the UK and France whose nationals had been direct victims of Libyan terrorism. Although Libya was a party to the NPT, it actively 16

See Council Six-Monthly Progress Reports on the Implementation of the EU Strategy OJ C 54, 296. 17 Council Doc. 10352/03, Conclusions of European Council, December 2003. The Strategy is analysed in House of Lords EU Committee 13th Report 2004–05, Preventing Proliferation of Weapons of Mass Destruction: The EU Contribution, HL Paper 96. For extensive analysis of the Strategy, with particular emphasis on the constitutional aspects, see P Koutrakos, ‘The Non-proliferation Policy of the European Union’ in M Evans and P Koutrakos (eds), Beyond the Established Legal Orders: Policy Inter-connections between the EU and the Rest of the World (Hart Publishing 2011). 18 See, for example, Council Docs. 2014/C 54, 25.2.2014, 2015/C 41, 5.2.2015, 2015/C 296, 8.9.2015. 19 Council Doc. 17172/08, 17.12.2008.

Non-proliferation of weapons of mass destruction 187 pursued a nuclear programme throughout the 1980s and 1990s, confirmed by inspection by the IAEA. In 2003 a German-owned vessel, the BBC China, was intercepted under the Proliferation Security Initiative (described below), taken to Italy and found to be carrying nuclear material supplied through the A.Q. Khan network.20 Confronted by hard evidence of deception and following intensive negotiations with the UK, Colonel Gaddafi on 19 December 2003 announced a decision to disarm and agreed that US and UK experts would remove and destroy illegally held equipment. In January 2004 it was agreed that US and UK experts would remove and destroy weapons and that this would be verified by the IAEA. Libya ratified the Comprehensive Test Ban Treaty in January 2004 and in March 2004 ratified an Additional Protocol to the NPT opening the way to more intensive inspections by the IAEA.21 Responding also to other measures taken by Libya to end its support for terrorism, the EU revoked sanctions on Libya going beyond those required by the Security Council in October 2004.22 There has been much criticism of wider actions – and inaction – by the EU in the context of the later collapse of the Gaddafi regime but it can certainly be said that the ensuing chaos would have been much worse had the stockpiles of WMD held earlier in Libya been appropriated by the various factions competing for power. Iran is where the EU has made its most high-profile contribution to containing the risks arising from proliferation of nuclear capacity. Iran’s programme of enrichment was for a number of years suspected of being capable of turning Iran in a relatively short time into an NWS. Iran consistently claimed that its activities were justified by its right under the NPT ‘to develop research, production and use of nuclear energy for peaceful purposes without discrimination’ – pursuant to which it intended to construct nuclear power plants over a 20-year period. In 2003, however, the Director-General, Mohamed El-Baradei, reported to the Board of Governors of the IAEA the existence of uranium enrichment facilities at Natanz which had not been declared to the Agency as required under Iran’s Safeguards Agreement. Under sustained pressure from the EU, Iran agreed to negotiate an Additional Protocol to its Safeguards Agreement, and the Foreign Ministers of France, Germany and the UK visited Iran to try to resolve the issues. Under the informal settlement reached, Iran agreed to suspend further enrichment and reprocessing, and the EU Ministers stated that dialogue would open on wider cooperation, including security guarantees and the possibility of a nuclear-free zone in the Middle East. After a further visit by the EU High Representative for the CFSP, Javier Solana, Iran suspended enrichment to the provisional satisfaction of the IAEA and on 18 December 2003 signed an Additional Protocol which it implemented in advance of ratification.23 20

See D Albright and L Hinderstein, ‘Uncovering the Nuclear Black Market: Working Toward Closing Gaps in the International Nonproliferation Regime’, paper prepared for the Institution of Nuclear Materials Management 2004 Annual Meeting; Bronwen Maddox Briefing, The Times, 13 February 2004; G Corera, Shopping for Bombs (OUP 2006), esp. chs 5 and 8. 21 ‘Contemporary Practice of the United States’ (2004) 98 American Journal of International Law 195, (2005) 99 American Journal of International Law 195; The Times, 29 May 2004. 22 Council Common Position 2004/698/CFSP of 14 October 2004, OJ 2004 L 317/40. 23 This episode is described in detail in T Cronberg, Nuclear Multilateralism and Iran: Inside EU Negotiations (Routledge 2017). Iran has still not ratified the Additional Protocol.

188 Research handbook on the EU’s common foreign and security policy In September 2004, however, the IAEA Board of Governors concluded that Iran had failed to observe these commitments and called for immediate suspension by Iran of all enrichment-related activity. The EU joined the US in threatening to refer Iran’s activities to the UN Security Council but the Foreign Ministers of France, Germany and the UK expressed to Iran readiness to conclude a wider and more generous agreement in exchange for full transparency and a halt to enrichment. The Paris Agreement concluded in November 2004 between Iran on the one hand and France, Germany and the UK, with the support of the EU, on the other, implemented this offer and included commitments on the EU side to resume the suspended negotiations for a Trade and Cooperation Agreement, to support Iran’s accession to the World Trade Organization and to join forces in fighting terrorism and in supporting the establishment of a constitutionally elected government in Iraq.24 In June 2005, however, Iran elected President Ahmadinejad, who quickly rejected the Paris Agreement, resumed enrichment at the Uranium Conversion Facility in Isfahan, and described a further European offer as ‘an insult to the Iranian nation’.25 In September the IAEA Board agreed to refer Iran’s conduct to the Security Council.26 The following month, President Ahmadinejad called for Israel to be ‘wiped off the map’, and Iranian Ambassadors to France, the UK, Germany and the UN (among others) were dismissed from the diplomatic service.27 The IAEA voted to report Iran to the UN Security Council, and in response Iran withdrew from the Additional Protocol – which it had not yet ratified. The Security Council was not then ready to impose sanctions on Iran but a statement was sent by the five permanent members plus Germany giving Iran 30 days to show that its nuclear programme was as peaceful as claimed. The US joined the diplomatic efforts where the EU had already taken the lead – offering direct talks with Iran for the first time since 1979 (when the hostages were detained in the US Embassy) as well as technical assistance and a guaranteed supply of nuclear fuel.28 In July 2006 the Security Council, acting under Chapter VII of the Charter, required Iran to suspend all uranium enrichment and reprocessing, and from that point Iran could no longer rely on its rights under the NPT, on the inconsistency of the conduct of the US, the UK and France, or on their failure to carry out meaningful disarmament as required by the NPT, but was in clear breach of international law by continuing and later extending its activities.29 The Security Council in response to Iran’s continuing

24

Iran’s Nuclear Programme: A Collection of Documents, Cm. 6443; Report of the IAEA Director-General 15.11.2004, GOV 2004/03; E Denza, ‘The EU, Iran and Non-proliferation of Nuclear Weapons’ (2005) 10 European Foreign Affairs Review 289, 301–311. 25 The Times, 10, 12 and 15 August 2005. IAEA Resolution of 11 August in Doc. GOV/2005/64. 26 Joint statement by the Foreign Ministers of the UK, France and Germany and the EU High Representative, 22 September, published in The Wall Street Journal; The Observer, 25 September 2005. 27 The Times, 29 October 2005, 2 November 2005 and 4 January 2006. 28 The Times, 31 March 2006, 1 June 2006, 7 June 2006. 29 UNSC Res 1696 (31 July 2006); DH Joyner, ‘Why Less is More: Law and Policy Considerations on the Iranian Nuclear Issue’ (Harvard Law and Policy Review, 23 October

Non-proliferation of weapons of mass destruction 189 enrichment imposed wider sanctions on Iran’s export of arms and on financial assistance to Iran and froze the assets of more individuals.30 While the US and Israel refused to rule out the use of force to destroy Iran’s nuclear facilities, the EU, while implementing wider sanctions than those required by the UN, continued to express the readiness of its High Representative to work for a diplomatic solution.31 The election of President Obama opened the way to direct though largely discreet negotiations in which the US – which harboured the greatest suspicions of Iran’s activities and which held the most attractive carrots as well as the most persuasive sticks – took the lead. There was, however, continuing and constant support (and pressure) from the EU. This wider support offered cover to the US and to Iran, both of which for historical reasons found it difficult to be seen to engage in bilateral relations. In November 2013 the negotiations between Iran and the Permanent Members of the Security Council (the P5), Germany and the EU resulted in a provisional deal in Geneva. Iran undertook to halt uranium enrichment above 5 per cent purity, to dilute or transform uranium stockpiles enriched above that level and – for the first time in eight years – to open its facilities to intensive IAEA inspection. On the other side there was a commitment not to impose new sanctions and some relaxation of sanctions on precious metals and on oil. If Iran was fully compliant, talks on a permanent arrangement would begin after six months.32 The final deal – a non-binding Joint Comprehensive Plan of Action (JCPOA) – was concluded in July 2015. The non-binding nature of the deal avoided the need for the US Senate to give positive approval (though other complex constitutional procedures did offer Congress an opportunity to block it). The commitments on both sides in the 2013 deal were extended, there was an elaborate timetable for implementation, which would require IAEA confirmation that Iran had taken the key steps required, and oversight by a Joint Commission as well as a ten-year ‘snapback provision’ enabling the Security Council effectively to re-impose sanctions on Iran on the demand of only one of its permanent members. President Obama made clear that the snapback provision was the real guarantee of Iran’s compliance with the JCPOA. He made no claim that the deal would result in any improvement in Iran’s human rights performance or lessen its destabilization activities in Syria or Yemen, but said that it would ‘result in cutting off all four pathways Iran could use to develop enough fissile material for a nuclear weapon’.33 2010) available at http://harvardlpr.com/online-articles/why-less-is-more-law-and-policyconsiderations-on-the-iranian-nuclear-issue/. 30 UNSC Res 1747 (24 March 2007). Later UNSC Resolutions over the next few years reaffirmed and extended sanctions on Iran. 31 Conclusions of the European Council, 17.6.2010, Annex II; The Times, 12 June 2010; T Sauer, Coercive Diplomacy by the EU: The Case of Iran (Netherlands Institute of International Relations, ‘Clingendael’ 2007); M Tocha, ‘The EU and Iran’s Nuclear Programme: Testing the Limits of Coercive Diplomacy’, EU Diplomacy Papers 1/2009. 32 The Times, 26 November 2013. 33 The Times, 15 July 2015; ‘Contemporary Practice of the US relating to International Law’ (2015) 109 American Journal of International Law 649, 874; (2016) 110 American Journal of International Law 347; S Blockmans, ‘The Nuclear Deal with Iran: Le Moment Suprême?’, CEPS Commentary, 16 July 2015.

190 Research handbook on the EU’s common foreign and security policy The JCPOA was met with suspicion and hostility, particularly in Israel, Saudi Arabia and the US, where there were strenuous attempts to block it. In January 2016 the IAEA confirmed Iran’s compliance – so opening the way to the dismantling of nuclear-related sanctions. The EU High Representative, Federica Mogherini, and the Iranian Foreign Minister issued a joint statement saying that ‘with political will, perseverance and through multilateral diplomacy, we can solve the most difficult issues and find practical solutions that are effectively implemented’.34 Thirteen years elapsed between the decision by the EU to engage Iran at the highest level in a sustained diplomatic effort to contain the risks to nuclear non-proliferation arising from its activities and ambitions and the full verification and implementation of the final deal. As with most foreign relations initiatives, success took a very long time and readiness to accept many setbacks. The EU was able to deploy political and economic carrots at the outset and was ready to engage with Iran at a time when the US could not do so, and this helped the US to move away from its historically understandable refusal to negotiate directly with Iran.35 When agreements patiently negotiated were flouted and further offers were contemptuously rejected, the EU continued to insist on the possibility of further diplomacy rather than retaliating and threatening even tougher sanctions or the use of force. It was ready to pass the baton to the US when the ground had been prepared and the US showed itself willing to assume primary responsibility. It showed sensitivity to the preoccupations of Iran, worked in close cooperation with the IAEA and later made effective use of Security Council powers. Taken as a whole it was a good example of action through collective diplomacy, which could not have been achieved by any one of the states acting alone. The EU bears a continuing responsibility for the implementation of the JCPOA since under its terms the High Representative is coordinator of the supervisory Joint Commission. In May 2016, responding to Iranian complaints that lifting of sanctions had been imperfect, the EU and the US issued a joint statement emphasizing that the JCPOA did not extend to sanctions unrelated to Iran’s nuclear development programme.36 The EU, and France and the UK in particular, have sought to restrain President Trump from carrying out his campaign threat to ‘tear up or rewrite the agreement’ and have contradicted his suggestions that it is not being honoured by Iran. The UK Foreign Secretary has told the US Congress that it was vital to preserve the deal.37

34 The Observer, 17 January 2016; European Council Press Release, 16 January 2016. For extensive analysis of the JCPOA, its operation and wider implications, see S Blockmans, A Ahteshami and G Baghat (eds), EU-Iran Relations after the Nuclear Deal (CEPS e-book 2016); DH Joyner, Iran’s Nuclear Program and International Law (OUP 2016) esp ch. 7. 35 A Cottey, ‘The EU’s Non-proliferation Strategy Ten Years On’ (2014) 19 European Foreign Affairs Review 45, 57–59. 36 US Department of State Press Release, Statement by France, Germany, UK, US and the High Representative of the EU for Foreign Affairs and Security on Post-JCPOA Business with Iran (2016) 110 American Journal of International Law 789–795. 37 The Times, 17 January 2017, 1 February and 10 November 2017.

Non-proliferation of weapons of mass destruction 191 3.2 Failures with High-risk States: North Korea and Syria By the terms of the 1992 Joint Declaration of the Denuclearization of the Korean Peninsula, both North and South Korea agreed not to ‘test, manufacture, produce, possess, store or deploy, or use nuclear weapons’. They undertook to use nuclear energy for peaceful purposes and not to possess reprocessing or uranium enrichment facilities.38 North Korea had become a party to the NPT in 1985 – motivated by the need to secure nuclear energy – and concluded a Safeguards Agreement with the IAEA in 1992 on the basis of assurances from the US and from South Korea that no NPT Party would deploy nuclear weapons on the Korean peninsula. In 2003, however, it withdrew from the NPT, citing a ‘serious threat’ from the US in the form of breach of undertakings to provide reactors and oil shipments, though at first it said that its activities would continue to be limited to peaceful purposes. For some time there was uncertainty as to whether North Korea possessed nuclear weapons but evidence emerged of assistance to it being given by the A.Q. Khan network, and in 2006, 2009, 2013, 2016 and 2017 it carried out nuclear tests which have provided evidence of its growing capacities both in regard to bombs and in regard to means of delivery.39 In response, the Security Council has under Chapter VII of the Charter imposed comprehensive sanctions on North Korea.40 The EU has gone beyond the requirements of the Security Council sanctions and has terminated financial assistance to North Korea. The sanctions have, however, had no impact in terms of influencing its defiant behaviour – largely because of the limited dependence of the country on imports and the willingness of its government to inflict severe suffering on its people. Nor have attempts at compromise under the Six Power Talks among North and South Korea, the US, Russia, Japan and China had any detectable success. The EU by virtue of Euratom competence became a member of the Korean Peninsula Energy Development Organization (KEDO) – a body set up to assist North Korea to build nuclear reactors for peaceful purposes but whose activities have largely been discontinued. It was not invited to participate in the Six Power Talks – which were dismissed as no longer necessary by North Korea in 2009.41 EU statements of condemnation can be said to have had no effect whatsoever.42 Years of threats from the US have had no greater effect – and the US is constrained by the fact that any use of force would probably lead to retaliatory annihilation of its ally South Korea. In September 2016 North Korea launched three ballistic missiles with a range of 800 38

33 ILM 569 (1994). C Moxley, ‘The Sword in the Mirror – the Lawfulness of North Korea’s Use and Threat of Nuclear Weapons Based on the United States’ Legitimization of Nuclear Weapons’ (2004) 27 Fordham International Law Journal 1379; Corera (n 20) 166–167; The Times, 15 April 2009, 26 May 2009, 9 February 2016, 28 April 2016. 40 See, in particular, UNSC Res 1874 (2009), which requires North Korea to return to its commitments under the NPT and Safeguards Agreement. On the legal implications of the withdrawal and of the Security Council’s demand for North Korea to re-accede, see DH Joyner and M Roscini (eds), Non-proliferation Law as a Special Regime (CUP 2012) 151–169. 41 C Portela, ‘The EU’s Evolving Responses to Nuclear Proliferation Crises: from Incentives to Sanctions’ (2015) EU Non-Proliferation Paper 46. 42 See, for example, statement by High Representative Mogherini on 6 January 2016. 39

192 Research handbook on the EU’s common foreign and security policy miles in the direction of Japan – with no apparent purpose other than to send a defiant signal to the G20 leaders, and in particular the leaders of South Korea, the US and Japan, then meeting in China.43 A few days later, following a fifth test, it claimed to have developed a nuclear weapon small enough to be mounted on one of its ballistic missiles.44 In May 2017 outside experts endorsed a claim by North Korea that a missile which it had just successfully tested was an intermediate-range missile capable of carrying a heavy nuclear warhead.45 Some EU Member States as well as the EU itself attach importance to retaining diplomatic missions in North Korea for delivery of humanitarian aid. Many states – and China in particular – are alarmed at the prospect of total collapse of North Korea as a state because of the impact of consequential flows of refugees and the difficulty of securing North Korea’s massive arsenal of nuclear weapons. China has demanded that North Korea desist from behaviour which worsens the situation, Russia and the UK have said that Security Council sanctions should be extended and enforced – but Kim Jong Un gives no outward sign of being frightened. Syria is a case where international action – led by a Framework Agreement between the US and Russia – appeared for a time to have had some success in limiting stockpiles of chemical weapons. The Security Council in 2013 required Syria to declare and destroy its stockpiles of these weapons.46 Syria – which had admitted possession of chemical weapons and sought to justify this as a deterrent to possible nuclear attack from Israel – became a Party to the Chemical Weapons Convention in October 2013. The OPCW became responsible for monitoring the identification and elimination of Syria’s chemical weapons. The EU committed 17 million euros by way of assisting the OPCW in its task – providing armoured cars and satellite imagery for the first phase of identification and contributing to a trust fund in support of the later phases. Syrian assets blocked under EU sanctions legislation were used for the project.47 The EU has more recently suggested use of these assets to finance verification and destruction of chemical weapons but the Syrian government has not agreed to this.48 In June 2014 the shipping-out of chemical weapons was declared to be complete, although suspicions remained that Syria’s disclosures fell short of being comprehensive. The Security Council in 2015 set up a Joint Investigative Mechanism (UN and OPCW) to investigate allegations of use of chemical weapons prohibited by the CWC.49 In August 2016 a three-person Leadership Panel reported back to the Security Council on behalf of the Joint Investigative Mechanism that after investigation of a number of cases of suspected use of chlorine gas it had found sufficient evidence of such use in three cases. Two of these it attributed to the Syrian air force and one to ISIL. Responsibility for the attacks was denied by Russia and by Syria – but the finding 43

The Times, 6 September 2016. The Times, 10 September 2016. 45 The Times, 16 May 2017. 46 UNSC Res 2118 (27 September 2013). For detailed analysis of events, see Krutsch, Myjer and Trapp (eds) (n 5) 689–701. 47 Council Decisions 2013/726/CFSP OJ L 329/41; 2014/74/CFSP OJ L 40/63. 48 Statement on behalf of the EU by the Special Envoy for Disarmament and NonProliferation at the 22nd OPCW Conference, 27 November 2017. 49 UNSC Res 2235 (2015). 44

Non-proliferation of weapons of mass destruction 193 confirmed that earlier disclosures by Syria were incomplete.50 Suspicions of withholding chemical weapons were confirmed in April 2017 when the Syrian air force used sarin gas against civilians. The violation of obligations under the CWC and Security Council Resolution was almost immediately condemned by Federica Mogherini, EU High Representative, who said that President Assad bore ‘primary responsibility’.51 When the Security Council failed to agree on a response, the attack was countered three days later by a US airstrike against the airfield from which the offending planes had taken off.52 In February 2018 a leaked report to the Security Council from the OPCW disclosed that 40 shipments of essential components enabling the 2017 sarin attack had been supplied to Syria from North Korea.53 3.3 Non-participating States: India and Pakistan India is one of the most difficult challenges for the EU policies on non-proliferation. Originally a state strongly committed to disarmament and a ban on nuclear testing, it declined to join the NPT because of its approach of discriminating between the states which already had nuclear weapons and the others. In 1974 India carried out its first nuclear explosion. It claimed that this was entirely for peaceful purposes, but this claim was not universally accepted. India carried out a further test and declared itself a nuclear weapon state in 1998, and very shortly afterwards Pakistan – also a non-party to the NPT – followed suit by carrying out a nuclear test. Since acquiring nuclear weapons, India has sought to participate in those arrangements which it sees as economically useful, while making clear that it has no intention of giving up its weapons in order to accede to the NPT as an NNWS. It has consistently refused to accept the inclusion of the EU’s clause (described below) requiring adherence to the main international conventions forming the framework for non-proliferation efforts. It claims special treatment on the grounds that since 1998 it has observed a moratorium on nuclear testing on a voluntary basis, has an explicit policy of no-first-use of nuclear weapons, observes faithfully those conventions to which it is a party and enforces effective export controls. In 2006 its efforts succeeded in that the US concluded an agreement – later ratified after much argument and special concessions in Congress – which authorized the US to provide assistance to civilian nuclear power development in India. Fourteen of its nuclear reactors were to be irreversibly classed as civilian and opened to IAEA inspection under a special Safeguards Agreement. The remaining eight reactors were to be shielded from inspection. India would continue its freeze on nuclear testing. India’s first nuclear explosion in 1974 led to the formation of the Nuclear Suppliers’ Group (NSG) – an informal regime which seeks strict control of nuclear exports to prevent 50

The Times, 26 August 2016. Al Arabiya, 4 April 2017. 52 ‘The US Air Strike after the Use of Chemical Weapons in Syria’, The Times, 6, 8 and 11 April 2017. For analysis of possible legal justifications for the US response, see M Hayashi, (2017) 21(8) ASIL Insights. The President of the EU Commission stated that the repeated use of chemical weapons must be answered and that the airstrikes ‘seek to deter further chemical weapons atrocities’: Press Release on the situation in Syria, 7 April 2017. 53 The Times, 28 February 2018; BBC News, 28 February 2018. 51

194 Research handbook on the EU’s common foreign and security policy non-proliferation, and membership of which is restricted to parties to the NPT. In 2008 India strengthened its special status through the grant of a waiver of NSG restrictions on India – which opened the way to activation of the deal with the US. Austria and Ireland, alone among EU Member States, opposed the waiver but gave in when India confirmed its assurances that it would not test nuclear weapons. EU Member States have remained divided on these developments, which in effect give India the benefits of NWS status without the need to disarm. France and the UK have nuclear cooperation agreements with India and are generally sympathetic to its position and its objectives, but to others it is unacceptable that India remains indefinitely outside the network of formal non-proliferation agreements while sharing benefits intended to be conditional on formally accepting their reciprocal constraints. It appears to be accepted that there is no realistic possibility of amending the NPT to extend the list of NWS because of the precedent this would set and the difficulty of drawing any wider boundary. In consequence, the EU has assumed a position of silence on the issues – which do not feature either in its regular reports on the WMD Strategy or in bilateral arrangements such as the Joint Action Plan of 2008. The record of the EU-India Summit on 30 March 2016 does not even mention WMD. There is justified criticism that this privileged treatment discriminates against states such as South Africa, Argentina and Brazil, which gave up nuclear ambitions in order to secure similar benefits under the NPT.54 Pakistan, like India, began with a strong aversion to acquiring nuclear weapons, but changed its approach in the wake of the secession of East Pakistan, which became independent as Bangladesh in 1971 with the military support of India. The new President Bhutto had already promised: ‘If India builds the bomb, we will eat grass or leaves, even go hungry, but we will get one of our own. We have no other choice.’55 The clandestine development of nuclear bombs received crucial support both from the Pakistani scientist A.Q. Khan, whose work for the international consortium URENCO in the Netherlands enabled him to steal secrets necessary to master the technology of enriching uranium, and also from China.56 Over the years Pakistan has made several concessionary overtures to India with a view to reciprocal disarmament or joint participation in relevant international agreements – such as simultaneous adherence to the NPT – but these have all been rebuffed. The prime ministers of India and Pakistan did, however, agree to continue a moratorium on nuclear testing in 1999, they have set up a hotline to warn one another of accidents which could be misconstrued as attacks, and they warn each other of impending missile tests. Because the stance of Pakistan is so closely dependent on that of India, the EU appears to have concluded that there is little point in pursuing the question of nuclear capability separately with Pakistan, and it does not feature in any account of EU–Pakistan relations. Pakistan is a party to the CWC and to the BWC. 54 The Times, 3 March 2006, 8 September 2008; B Rienzle, ‘Integrating Without Quite Breaking the Rules: The EU and India’s Acceptance Within the Non-proliferation Régime’ (Non-Proliferation Paper No. 43, EU Non-Proliferation Consortium 2015); Joyner (n 3) 37–40. 55 Corera (n 20) 9. 56 ibid, c 1; Joyner (n 3) xviii–xix.

Non-proliferation of weapons of mass destruction 195 3.4 Non-participating States: Israel Israel is generally believed to have possessed nuclear weapons since 1966 but has always followed a policy of ‘deliberate ambiguity’ or ‘opacity’. In 1966 Israel stated that it would ‘not be the first to introduce [nuclear weapons] into our region’.57 It appears to construe the word ‘introduce’ as comprising testing or publicly acknowledging. It regards its stance as an absolute guarantee of Israeli security and does not accept the possibility of other states in the region acquiring nuclear weapons. In pursuit of this policy it has carried out pre-emptive strikes against nuclear facilities in Iraq (Osirak) in 1981 and in Syria in 2007, claiming self-defence as a justification. There are wide discrepancies in estimates of the number of weapons it possesses, but it is undisputed that it has the means of effective delivery. It is not a party to the NPT or to any of the other key international conventions. It maintains high levels of secure export control without participating in the Australia Group.58 Israel supports the concept of a nuclear-free zone in the Middle East, while making clear that active steps to secure it must follow the establishment of a comprehensive regional peace settlement. EU external policy is strongly committed and active in its support for the Middle East Peace Process, and the EU appears to have accepted the immutability of the Israeli position to the extent that there is no attempt to raise the question of WMD in any bilateral forum such as the EU-Israel Action Plan or any mention of it in its six-monthly reports on the EU Strategy on WMD.

4. LONGER-TERM EU ACTION The EU now divides its priorities in the area of non-proliferation of WMD into four main categories – strengthening the international system of agreements, pursuing universal participation in these agreements, reinforcing compliance, and cooperating closely with key partners in the fight. These reflect the objectives identified in the 2003 EU Strategy, and confirmed in the Global Strategy for the EU’s Foreign and Security Policy in 2016.59 4.1 Strengthening the International System The highest priority is given to securing the entry into force of the CTBT. The Treaty was adopted by the UN General Assembly in 1996 and – in contrast to the NPT – the prohibition on all nuclear testing is universal and non-discriminatory. Although 164 57

Statement by Prime Minister Levi Eshkol to Israel’s Knesset, 18 May 1966, confirming assurance given to US in Memorandum of Understanding of 10 March 1965, printed in Foreign Relations of the United States 1964–68, doc. 185. 58 A Cohen, The Worst-Kept Secret (Columbia University Press 1998); N Gordon in Al-Jazeera, 13 December 2013. 59 ‘Shared Vision, Common Action: A Stronger Europe, a Global Strategy for the European Union’s Foreign and Security Policy’, EEAS 28 June 2016. It is suggested by Lars-Erik Lunden that following this Strategy there should now be a further EU Strategy specific to WMD, see (2017) EU Non-Proliferation Consortium Paper 58.

196 Research handbook on the EU’s common foreign and security policy states have ratified, further ratification by eight of the states specifically listed in Annex II is necessary to bring the CTBT into force. Three of these – North Korea, India and Pakistan – have never signed. The US, China, Egypt, Iran and Israel signed but have never ratified. To some extent they take cover under the failure of the US to ratify following early rejection of the Treaty by the US Senate. However, the US implements a moratorium on nuclear testing and President Obama in 2009 committed himself to seeking Senate advice and consent a second time. There is little sign of success from EU diplomatic efforts to bring the Treaty into force. The Treaty, however, has been influential to the extent that only North Korea has actually carried out nuclear tests in recent years, and these have been monitored, verified and reported to the Preparatory Commission of the CTBT. The EU offers financial support to states signatories to enable them to strengthen their capabilities in detection, as well as support for training and education programmes hosted by the Preparatory Commission.60 A long-standing objective is the Fissile Material Cut-off Treaty,61 which would ban production of fissile material for nuclear weapons without limiting production for peaceful purposes, and would complement the NPT and the CTBT. Negotiations were endorsed by the UN Conference on Disarmament and a group of experts was established but progress was at first blocked by US refusal to accept international verification procedures. The US has recently indicated readiness to accept verification but active negotiations are now blocked by Pakistan, which has taken the position that any treaty would specifically target Pakistan (whose stocks are significantly lower than those of India). Although there was some support among EU Member States for freezing development aid to Pakistan, there has never been the necessary agreement for deploying tough measures – and as a non-party to the NPT, Pakistan is not in breach of any international obligations.62 The EU has consistently supported further negotiations both in General Assembly statements and in private lobbying, but without success. The EU has from the outset supported the Proliferation Security Initiative (PSI) launched by President Bush in Krakow in 2003 within the framework of the US National Security Strategy. If all states effectively implemented domestic and international rules on export controls on nuclear material and other items with the potential for use in WMD, the PSI would not be necessary, but given inevitable inadequacies or deceit, the system is designed to enable states with intelligence relating to suspect materials in transit by sea to intercept and seize them. The PSI has no binding international legal framework but operates as a multilateral political commitment within the framework of existing domestic and international legal rules. Participating states subscribe to a Statement of Interdiction Principles – sharing information on ships suspected of carrying WMD in their cargo, developing procedures for swift authorization of interception of such ships with necessary consent, and searching and taking 60

Cottey (n 35) 52–53. Detail of the EU support is set out in Council Decisions 2012/699/CFSP, OJ 2012 L 314/27 and 2014/C 54, OJ 2014 C 54. See statement by EU High Representative Mogherini at Vienna gathering to mark 20th anniversary of CTBT opening for signature, 6 January 2016. 61 Treaty banning the production of fissile material for nuclear weapons or other explosive devices. 62 Cottey (n 35) 56.

Non-proliferation of weapons of mass destruction 197 enforcement action. Seven EU Member States were among the original ‘core members’ of the scheme – which now has 105 participating states notwithstanding some criticism of its legality. The interception of WMD destined for Libya – described above – was an early example of its successful use and there have been numerous others.63 4.2 Pursuing Universal Participation in WMD Conventions The EU now seeks inclusion in all mixed bilateral international agreements of a clause requiring the parties (in effect the non-member state) to comply with their existing obligations under the core treaties on non-proliferation and to implement these treaties effectively in domestic law, to take steps to become parties to those treaties by which they are not already bound, and to set up and maintain an effective system of national export and transit control of WMD-related material. The general commitment to non-proliferation is described as an ‘essential element’ of the agreement – so opening the way in the event of default to suspension by the EU and its Member States of their reciprocal undertakings – and the more precise undertakings may be made an ‘essential element’ on a case-by-case basis.64 Because of the limited EU competence in the area, the clause is not proposed in free trade agreements to be concluded by the EU acting alone. As stated in the Progress Report covering the first half of 2015, ‘[t]he negotiations contribute to awareness raising of the EU’s policy on non-proliferation and disarmament and provide a forum to increase mutual understanding of relevant provisions, identify areas for future cooperation and to encourage partner countries to make concrete headway with regard to the international non-proliferation regime’.65 The clause has been included in an impressive number of agreements – but finds its limits in the context of negotiations with high-risk states. India, Pakistan, Syria, Russia and China – to take obvious examples – have not been prepared to accept EU supervision of their policies regarding non-proliferation.66 The obstacles to insisting on the clause in dealings with high-risk states, and even with states such as South Korea and Canada with reliable records of pursuit of non-proliferation objectives, are political, legal and sometimes humanitarian, but the most important go to the reluctance of sovereign states to accept external monitoring of their decisions to accept or reject treaties or to apply effective internal rules of enforcement.67

63

M Byers, ‘Policing the High Seas: The Proliferation Security Initiative’ (2004) 98 American Journal of International Law 526; D Guilfoyle, ‘The Proliferation Security Initiative: Interdicting Vessels in International Waters to Prevent the Spread of Weapons of Mass Destruction’ (2005) Melbourne University Law Review 23. For analysis of the legal and practical difficulties with the scheme, see Joyner (n 3) ch 8. 64 The model clause, first developed in the context of the EU Strategy of 2003, is in Council Doc. 14997/03, 19 November 2003. 65 2015/C 54, OJ 2015 C 296, para 10. 66 Cottey (n 35) 53–57. 67 For analysis of these difficulties, see L Grip, ‘The EU’s weapons of mass destruction non-proliferation clause: a ten-year assessment’, EU Non-proliferation Consortium Nonproliferation Papers No. 40/2014; Portela (n 41).

198 Research handbook on the EU’s common foreign and security policy The Council Joint Actions adopted in 2006 and 2012 in support of the Biological and Toxin Weapons Convention68 reflect another initiative aimed at securing universal participation in that Convention as well as effective implementation of its requirements in domestic law. These Joint Actions were adopted within the framework of the 2003 EU Strategy against Proliferation of WMD and in advance of successive Review Conferences of the Convention. They envisage provision of legal and technical support for individual states as well as the organization of regional seminars and conferences to promote awareness of the Convention. 4.3 Reinforcing Compliance There are two aspects to enforcing compliance with the NPT – first, moving towards implementation by the NWS of their commitments to taking effective measures towards nuclear disarmament and, secondly, strengthening the practical efficacy of the safeguards system and of export controls. It is in regard to the first aspect that the sharp divisions between the capacities and the policies of the Member States make it impossible to formulate a constructive common European position. The differences on disarmament are not simply between the UK and France as NWS and the other Member States as NNWS. Austria, Ireland, Sweden, Finland, Cyprus and Malta are now strongly committed to the Humanitarian Initiative, which claims new evidence of the catastrophic impact of any use of nuclear weapons in populated areas and seeks greater momentum towards the disarmament objective which was fundamental to the bargain enshrined in the NPT.69 France and the UK regard this Initiative as a distraction from practical steps towards reductions in nuclear arms. Germany, Poland and the Netherlands are committed to the Non-proliferation and Disarmament Initiative – a cross-regional group which also includes Australia, Canada and Japan among others and seeks to find common ground between the NWS and the NNWS.70 Every five years the NPT is reviewed by a Conference of the Parties, an assessment made of progress towards its objectives, and plans made for its future. The links between individual Member States or groups with wider international consortia were in the context of the 2010 Review Conference regarded as helpful in securing a useful outcome for the objectives agreed collectively in the context of formulating the EU common position.71 By 2015, however, disillusionment with the slow progress towards nuclear disarmament by the NWS had come to assume greater prominence among the preoccupations of the parties taking part in the Review Conference. Russia and the US argue that they have in fact made substantial progress towards nuclear disarmament in 68

Joint Action 2006/184/CFSP, OJ 2006 L 65/51; Joint Action 2012/421/CFSP, OJ 2012 L 196/61. 69 J Nielsen and M Hanson, ‘The European Union and the Humanitarian Initiative in the 2015 Non-proliferation Treaty Review Cycle’ (2015) EU Non-Proliferation Paper 41. 70 M Smetana, ‘Stuck on Disarmament: The European Union and the 2015 NPT Review Conference’ (2010) 92 International Affairs 137, 141. 71 M Dee, ‘Standing Together or Doing the Splits? Evaluating European Union Performance in the Nuclear Non-proliferation Treaty Review Negotiations’ (2012) 17 European Foreign Affairs Review 189, esp. 207–211.

Non-proliferation of weapons of mass destruction 199 terms of reducing the number of their missiles and their maintenance on a state of high alert, so reducing the risk of release through accident, sabotage or misunderstanding. Strategic arms limitation talks (SALT) began in 1969 between the Soviet Union and the US and under successive treaties there have been substantial reductions in warheads, missiles and bombers. The current treaty between Russia and the US, known as New START, entered into force in 2011 and will when fully implemented result in further limitations, though the 1,550 warheads and 700 missiles and bombers on each side falls well short of ‘complete nuclear disarmament’.72 The Treaty sets up a more effective verification regime permitting satellite, remote and on-site inspections. The US and Russia argue that the commitment in Article VI of the NPT is to negotiate towards a treaty securing universal nuclear disarmament under effective international control and that this is a long-term goal. The UK and France position is that unilateral measures would not enhance international security if the effect were to leave nuclear weapons as the sole prerogative of states such as Russia and China, which have in the last few years shown greater disregard for the rules of international law, and states such as India, Pakistan and North Korea, which have not taken even the first step of participation in the NPT. They also maintain that the non-proliferation obligations in the NPT are more fundamental than the disarmament obligations.73 Reflecting the fact that the majority of EU Member States are now members of NATO, the statement of the EU position by High Representative Mogherini at the 2015 Review Conference echoed the stance of the UK and France, arguing that progress towards universal disarmament should be made ‘through an overall reduction in the global stockpile of nuclear weapons’, with the primary responsibility placed on the states possessing the largest numbers of nuclear weapons, namely the US and Russia.74 The 2015 Review Conference was generally regarded as unproductive.

72

The Times, 2 April 2009. Their stance is supported by N White in Joyner and Roscini (eds) (n 40) 117–118, who claims that the NPT is no longer a contractual bargain but has become a constitutional treaty under which imbalance remains acceptable: ‘It is the acceptance of that inequality by other states (by becoming members of the UN or parties to the NPT), and by the fact that those instruments contain the rules and institutions necessary to try to ensure the survival of the planet, that give them their legitimacy as constitutional instruments’. See also the judgment of the International Court of Justice (and dissenting judgments) in the case Obligations concerning Negotiations relating to Cessation of the Nuclear Arms Race and to Nuclear Disarmament (Marshall Islands v United Kingdom) 5 October 2016; UN Treaty on the Prohibition of Nuclear Weapons, October 2017. Not one of the nine states known or believed to possess nuclear weapons supported this treaty. 74 Statement made on 28 April 2015, available at eu-un.europa.eu/articles/en/article_16386_ en.htm; M Dee, ‘The EU’s Performance in the 2015 NPT Review Conference: What Went Wrong?’ (2015) 19 European Foreign Affairs Review 591, esp. 597–598. In October 2016 Russia suspended an agreement with the US requiring both parties to eliminate weapons-grade plutonium, claiming that the US method of implementation retained redeployment potential for the plutonium. President Putin indicated that the agreement could continue if the US met unrelated demands such as cancelling sanctions against Russia: The Times, 4 October 2016. 73

200 Research handbook on the EU’s common foreign and security policy 4.3.1 Strengthening safeguards All EU Member States are parties to Safeguards Agreements and Additional Protocols with the IAEA, and their implementation is verified by the European Commission Nuclear Safeguards Directorate in close cooperation with the IAEA. The EU has taken a number of measures with the objective of strengthening practical compliance by non-member states with high standards of safeguarding nuclear material in their territories – providing financial support to the IAEA, and to the G8 Global Partnership against the Spread of Weapons and Materials of Mass Destruction,75 urging conclusion by other states of Additional Protocols to the NPT, assisting them to improve their facilities for technical control, and suspending nuclear cooperation in the absence of assurances by the IAEA that its safeguarding standards are being met. Financing of improvements to safety of reactors in non-member states is currently undertaken under the Instrument for Nuclear Safety Cooperation adopted under Article 203 of the Euratom Treaty.76 The new instrument authorizes expenditure on safety of installations and radiation protection as well as safeguards and underpins cooperation with 20 states. It is used to fund support for Ukraine in its continuing efforts to limit the damage caused by the Chernobyl accident in 1986. Similar action is taken to ensure effective enforcement of the CWC and the Biological and Toxin Weapons Convention.77 4.3.2 Export controls The current EU rules for control of exports, including items in transit and dual-use items, are set out in Council Regulation 428/2009.78 This instrument took account of changes in EU policy on WMD, as well as updates to rules established in several informal international proliferation control groups – the Nuclear Suppliers Group, the Wassenaar Arrangement, the Missile Technology Control Regime and the Australia Group. There is constant review of the procedures applied and the items covered – particularly the dual-use items listed – and dialogue is maintained with non-member states on the efficiency of their controls.79

75

L Grip, ‘Assessing Selected European Union External Assistance and Cooperation Projects on WMD Non-proliferation’ (2011) EU Non-Proliferation Paper 6. The EU is the second-largest donor to the IAEA Nuclear Security Fund. 76 Council Regulation (Euratom) 237/2014, OJ 2014 L 77/109. The contribution of the European Parliament to the Regulation is described in L Grip, ‘The European Parliament and WMD Non-proliferation: Policy-making Processes and Decision-Making Outcomes’ (2013) 18 European Foreign Affairs Review 563, 578–579. 77 See, for example, Council Joint Action 2006/184/CFSP and Council Common Position 2006/242/CFSP, OJ 2006 L 65/51; Council Decision 2011/429/CFSP, OJ 2011 L 188/42; M Alvarez-Verdugo, ‘Mixing Tools Against Proliferation: The EU’s Strategy for Dealing with Weapons of Mass Destruction’ (2006) 11 European Foreign Affairs Review 417, 426–427. 78 OJ 2009 L 134. 79 See Progress Reports on the Implementation of the EU Strategy, in 2014, OJ 2014 C 54/9–10, and OJ 2015 C 41. Particular attention has been given recently to cooperation with China, Jordan and Kazakhstan.

Non-proliferation of weapons of mass destruction 201 Coordination and dialogue among research bodies and 65 European think tanks – as well as new projects – is supported by EU funding through an EU Non-proliferation Consortium operating since 2011.80 4.4 Cooperation With Key Partners Cooperation is most productive with the US – and in the case of Iran, described above, has produced important results. The EEAS Principal Adviser and Special Envoy for Non-proliferation and Disarmament – since 2013 the Polish diplomat and former Head of NATO’s WMD Non-Proliferation Centre Jacek Bylica – represents the EU in meetings of the IAEA, the G8 Non-proliferation Directors Group, at the UN General Assembly First Committee and at meetings of the supervisory bodies of the main international conventions. Non-proliferation is consistently integrated into political dialogue and informal contacts with all relevant non-member states.

5. CONCLUSIONS OF WIDER SIGNIFICANCE FOR CFSP Non-proliferation of WMD – as was stressed at the outset – is a multifaceted activity in which all the strands are of long-term importance and high-profile successes are very few. These are equally characteristics of the foreign policies of sovereign states. It is tempting to focus on some specific event which may be assessed as a ‘success’ or a ‘failure’ – such as the Review Conferences of the NPT – and to draw overall conclusions about EU policy without putting the event in the wider and longer-term context of more tedious activity or in the context of the inherent difficulty of the challenge. Certain conclusions may, however, be drawn from the selective account given above of what the EU has attempted and what it has achieved. The usefulness of EU strategies Strategies demonstrate the wide-ranging nature of the challenge of the WMD nonproliferation threat and the need for a multifaceted approach which uses financial as well as political incentives and sanctions. The regular Council reviews at six-monthly intervals enable dispassionate assessment of where progress has been made in all the relevant areas and open the way to a balanced appraisal of the objectives of the Strategy and to its possible revision.81

80 The enterprise is now based on Council Decision 2010/430/CFSP of 26 July 2010 establishing a European network of independent non-proliferation think tanks in support of the implementation of the EU Strategy against Proliferation of Weapons of Mass Destruction, OJ 2010 L 202/5, which lists current projects in its Annex. 81 On the procedure for preparing the Reviews and their merits in promoting interinstitutional collaboration, see Zwolski (n 15).

202 Research handbook on the EU’s common foreign and security policy Integration of initiatives with wider diplomacy The EU has been effective in using the wide range of tools at its disposal to achieve low-key successes – as with improving safeguards and export controls in states which may not appear obvious proliferators but where weak administrative systems make them a useful transit point for suspect items. Study of the channels used by the A.Q. Khan network illustrates the importance of this technical support work. The systematic inclusion of WMD clauses in EU and Member State agreements may appear to have been of limited value because high-risk states have been unwilling to subscribe to them – but the number of subscribing states has built up a pattern of awareness in foreign ministries of the multilateral instruments as well as a critical mass of general compliance with their terms. Readiness for the long haul with recalcitrant partners The multilateral settlement eventually reached with Iran – with its effective system of monitoring and safeguards against deception – shows the value of the EU’s willingness to persist with diplomatic negotiations even in the face of apparent contemptuous rejection. It is likely that – given the unhappy history of US–Iran relations and the consequent mutual suspicion – the US could not have embarked on serious negotiation with Iran had the European efforts not paved the way. Effectiveness of sanctions The record in the context of non-proliferation efforts shows that sanctions need to be assessed over a substantial period, since the initial political effect of imposing sanctions is usually to cause an unhelpful political reaction of resentment rather than a reappraisal of policy on the part of the targeted state. To succeed, sanctions usually need to be deployed in tandem with a range of incentives. The EU has the advantage over a single state of a wider range of potential sticks and carrots and of uniformity of enforcement which makes diversion more difficult if not impossible. Over the years, sanctions have been refined by the EU – not only to make them ‘smarter’ but also to address problems of injustice to individuals adversely affected. Method of representation There is no doubt that the deployment on behalf of the EU of a highly skilled and experienced arms control diplomat has produced effective results in this field. Both Annalisa Giannella on her appointment in 2003 by the High Representative Javier Solana and now Jacek Bylica (appointed in 2013) had highly impressive qualifications for the post of EU Special Envoy for Non-proliferation and Disarmament, which carried weight in multilateral bodies such as the G7 and the IAEA. The different nature of the separate strands of non-proliferation resulted in fragmentation of the policy between the Commission, which at the outset had the technical expertise as well as the financial resources, and the Council, which assumed the main role in diplomatic negotiations with disarmament aspects, and this led to difficulties between the two

Non-proliferation of weapons of mass destruction 203 institutions as well as some confusion on the part of non-member states.82 Although the establishment of the European External Action Service may smooth some of the problems, it will take some time before habits and loyalties are integrated among the disparate elements making up the relatively new service so as to enable a proper assessment to be made. There can, however, be no doubt that the central cause of the EU’s difficulty in making progress with the disarmament pillar of the NPT is due not to any inadequacy in its arrangements for representation but to fundamental differences of policy among the Member States, causing radically different approaches to new initiatives. Non-proliferation of WMD is a challenge of immense difficulty, complexity and importance – where the EU through the CFSP has assumed an increasingly important role. Given that WMD cannot be uninvented and that their components have many peaceful and beneficial purposes, the immediate objective should not be regarded as universal and complete disarmament but the limitation, control and containment of weapons so that they are never used in anger. It is now 70 years since at Hiroshima and Nagasaki nuclear weapons were used to end a war, and their actual use puts beyond any argument their terrible and long-lasting consequences. It is a significant achievement that in 70 years – covering 45 years of Cold War involving all the states which possessed these weapons – they have never again been used and that proliferation has been limited to the extent that only nine states are now generally acknowledged to have this power of mass destruction. So there is a continuing need to be fearful, vigilant and responsible – but also to have some hope and confidence that the policies reflected in the key international treaties, in the EU’s unremitting efforts to support, to enlarge and to enforce them, and in the national safeguard and monitoring laws have been the right ones. It follows from that assessment that EU policy should continue to be integrated with other aspects of the CFSP, that it should be acknowledged that discrimination between actual and potential possessors of WMD may be justified in the light of wider assessment of danger of their misuse, and that political focus should be directed towards achievable goals and towards the most obvious risks of misuse. Substantial progress has been made towards halting the testing of nuclear weapons – through the Partial Test Ban Treaty, the Comprehensive Test Ban Treaty and the general moratorium now observed by all states with the single exception of North Korea. To persuade India and Pakistan jointly to become parties to the CTBT, and the new government in Washington to ratify it, is an objective where the EU might, by giving the issue greater prominence and political investment, make a contribution – even if it is unlikely that the current Republican administration in Washington will make this a priority. The deal agreed with Iran was a major success against heavy odds, and – while it must continually be monitored – Iran may need to be reassured that it has to some extent opened the way to its re-integration into the international community. North Korea is clearly the most dangerous and intransigent threat and one where the EU cannot lead except perhaps by continuing to argue that the multifaceted approach, together with an 82 ibid. For extensive analysis of EU achievements and recommendations for further action, see I Anthony and L Grip, ‘Strengthening the EU’s Approach to WMD Non-Proliferation’ (2013) EU Non-Proliferation Paper 37.

204 Research handbook on the EU’s common foreign and security policy emphasis on patient diplomacy, have yielded dividends in other cases. The EU itself lacks incentives and threats likely to influence North Korea but this is not the case with China – and the US is actively trying to apply pressure on China to deploy them. China is conscientiously giving effect to UN sanctions imposed on North Korea but is resistant to measures going beyond these, which might lead to the collapse of the regime.83

83 See S Chull Kim and MD Cohen (eds), North Korea and Nuclear Weapons: Entering the New Era of Deterrence (Georgetown University Press 2017).

PART C THE NEXUS BETWEEN CFSP/CSDP AND OTHER EXTERNAL POLICIES

10. The law and practice of EU sanctions Christina Eckes*

1. INTRODUCTION EU sanctions play a very particular role in CFSP. On the one hand, they are measures of general application, i.e. a tool of EU foreign relations that serves general policy objectives; on the other, they are individualized decisions to interfere directly with the fundamental rights of specific persons. Sanctions are deployed with exceptional frequency. They are a forceful tool that impose directly applicable obligations on private parties. More case law exists on sanctions than on any other aspect of CFSP. This is subject to the caveat that the CJEU has so far predominantly reviewed measures that the EU takes under the TFEU to give effect to the CFSP sanctions rather than the CFSP decisions themselves.1 This chapter explains the unique nature of sanctions (Section 2), traces the implications of sanctions for other external action policies and fields of law (Section 3), examines the intrinsic tension between the political objectives of sanctions and their regulatory, administrative or even criminal character (Section 4), identifies new trends (Section 5) and makes recommendations on how sanctions should be improved (Section 6).

2. SANCTIONS ARE EXCEPTIONAL AMONG CFSP POLICIES Public policy can be defined as a set of ideas or principles, or a plan of what to do in particular situations that has been agreed to officially by a group of people, a government or a political party.2 The EU has at least the intention to adopt sanctions as a policy that is meant to follow preconceived principles and serves ex ante identifiable objectives.3 This is therefore the standard against which EU sanctions should be evaluated, even if they may very well be criticized for being fragmented, inconsistently applied and even adopted in a way that does not achieve equal treatment of similar cases. * The author would like to thank Minke de Haan for her research assistance. 1 The exception is the recent Case C-72/15 Rosneft Oil Company OJSC v Her Majesty’s Treasury EU:C:2017:236 and arguably Case C-355/04 Segi and Others v Council EU:C: 2007:116. 2 This definition draws from http://dictionary.cambridge.org/dictionary/english/policy. 3 See Council, ‘Guidelines on implementation and evaluation of restrictive measures (sanctions) in the framework of the EU Common Foreign and Security Policy’ (15 June 2012) http://data.consilium.europa.eu/doc/document/ST-11205-2012-INIT/en/pdf.

206

The law and practice of EU sanctions 207 2.1 Cross-Treaties Bridge EU sanctions do not fit easily amongst other CFSP policies. They consist of a peculiar combination of a CFSP decision, adopted under Title V of the TEU, and a TFEU regulation adopted pursuant to the TFEU. Indeed, Article 215 TFEU, the only legal basis that has so far been used for sanctions since the entry into force of the Lisbon Treaty, is the closest explicit link between TEU and TFEU. It is not a joint legal basis but a legal basis within the TFEU, which requires the prior adoption of a CFSP instrument as a ‘prerequisite for the validity of a regulation’.4 As a result, EU sanctions are adopted on a combination of legal instruments based on a legal basis from each Treaty, which puts the powerful legal instrument of a directly applicable TFEU regulation at the service of CFSP objectives. This is structurally peculiar in that it is an exception to the rationale of Article 40 TEU that the CFSP and policies under the TFEU should remain separate.5 2.2 Direct Rights Relevance of EU Sanctions EU sanctions play a particular role in CFSP in that they directly legally target individuals, i.e. natural and legal persons, and list them as supporting either a targeted political regime or terrorism. They also predetermine in every detail the adoption of directly applicable regulations under the TFEU, including the list of targeted persons. Accordingly, sanctions are a strand of CFSP with exceptional operational means. They can make use of measures adopted under the TFEU to target individuals and bind EU citizens and businesses. Already the adoption of the CFSP decision itself has fundamental rights consequences since it has a negative impact on the reputation of those sanctioned.6 Moreover, the CFSP decision of the Council, which is mirrored in the TFEU regulation, should be seen as the actual origin of the latter’s legal effects. This makes EU sanctions the only CFSP strand which by definition and directly legally affects the rights of individuals, including in a very severe manner. They name individuals and as a consequence immediately tarnish their reputation, even without any further action by banks or border agencies.7 This is different from, for example, CFSP decisions which mandate EU naval forces to target ‘pirates’ as a category of

4

Rosneft (n 1), para 55. C Eckes, ‘The CSFP and Other EU Policies: A Difference in Nature?’ (2015) 20 European Foreign Affairs Review 535. 6 C Eckes, EU Counter-Terrorist Policies and Fundamental Rights: The Case of Individual Sanctions (OUP 2009) chapter 5; C Eckes, ‘Annotatie, “Abdulrahim v Council and Commission [2013] Case C-239/12 P (28 May 2013) – Hof van Justitie van de Europese Unie (Grote Kamer)”’ (2013) 8 EHRC. 7 See e.g. the current list of singled-out individuals who allegedly are associated with the ISIL (Da’esh) and Al-Qaida organizations. Annex I to Council Regulation (EC) 881/2002, most recently amended by Commission Implementing Regulation (EU) 2017/700, 266th amendment OJ 2017 L 103/22. 5

208 Research handbook on the EU’s common foreign and security policy persons that take actions that make them fall into this category, while they are taking these actions.8 In the words of Advocate General Sharpston: ‘It is worth recalling that the consequences of listing are very serious. Funds and other financial assets of economic resources are frozen … for a person, entity or group that is named in the … list, normal economic life is suspended.’9 Other external action policies may interfere with fundamental rights.10 Yet usually this is a matter of either implementation or application, either by national actors or carried out under an EU banner. Rights interferences of the latter category usually still raise relevant questions of attribution before it can be established that the action was indeed an act of the EU. In the case of sanctions, attribution to the EU is unproblematic since the CFSP decision immediately labels those listed as supporters of a sanctioned regime or terrorism, as contributing to nuclear proliferation or as building concentration camps. They interfere with fundamental rights of individuals immediately by virtue of the legal act itself. 2.3 Jurisdiction Article 275 TFEU excludes the CJEU’s jurisdiction ‘with respect to the provisions relating to the [CFSP]’ and ‘with respect to acts adopted on the basis of those provisions’. Article 275(2) TFEU makes an exception to this exclusion. It stipulates exceptional judicial review by the Court of the legality of CFSP decisions providing for restrictive measures against natural or legal persons. Substantively this can be explained by the direct rights relevance of CFSP sanctions measures described above. Arguably this direct rights dimension is the substantive reason for introducing exceptional review of CFSP decisions that impose ‘restrictive measures against natural and legal persons’. Yet the CJEU held in the case of Rosneft that the jurisdiction of the CJEU under Article 275(2) TFEU should be interpreted to cover not only the specific measures against individualized persons, but also measures of general application to the extent that they prohibit everyone from economically supporting or interacting with the individualized persons on the list. The Court did not extend its jurisdiction under Article 275(2) TFEU to cover measures of general application unrelated to the list of targeted individuals, such as embargoes and trade restrictions on categories of products.11 This may be reasonable because of their close connection with individualized measures and because of their potential high impact on rights; yet this distinction also results in a fragmentation of jurisdiction over parts of a CFSP decision.12 8 See e.g. Council Joint Actions 2008/851/CFSP, on a European Union military operation to contribute to the deterrence, prevention and repression of acts of piracy and armed robbery off the Somali coast (Atalanta) OJ 2008 L 301/33 and the subsequent amendments. 9 Case C-599/14 P, Council v LTTE EU:C:2016:723, Opinion of AG Sharpston, para 102. 10 E.g. EUNAVFOR Med operation Sophia has the competence to board, search, seize and divert vessels suspected of being used for human smuggling or trafficking on the high seas. For recent activities see https://eeas.europa.eu/csdp-missions-operations/eunavfor-med_en. 11 Rosneft (n 1), paras 94–107, with the conclusion in para 107. See Chapter 4 in this volume. 12 ibid.

The law and practice of EU sanctions 209 Sanctions have been the subject of an exceptionally large number of rulings, both of the General Court and the CJEU. At the time of writing (June 2017), 181 cases on restrictive measures have been decided by the ECJ,13 94 of which have been successful.14 Sixty-four additional cases are pending before the GC.15 Ten appeals were brought to the ECJ in 2013 and 2014 only. Two appeal cases were decided in April 2016.16 This overview does not count cases that only relate to or refer to restrictive measures. However, until the case of Rosneft in 2017 these rulings all reviewed the TFEU regulations rather than the CFSP decision. The Court has agreed in settled case law to review the TFEU regulations, irrespective of whether they merely reiterate a CFSP decision.17 Because the TFEU regulations are in relevant parts prescribed by the underlying CFSP decisions, rulings on most aspects of the procedural and substantive legality of the regulation – even if they do not directly address the legality of the CFSP measure – will also carry great persuasive weight with regard to the legality of the CFSP decision. 2.4 The Actual Decision Takes Place Outside of the EU Legal Order Three types of EU sanctions can be distinguished: first, EU sanctions giving effect to UN Security Council Resolutions that impose sanctions, for instance the sanctions at stake in the Kadi case, implementing Security Council Resolution 1267; secondly, supplementary measures that are adopted by the EU and that are related to but go beyond UN sanctions, such as the autonomous EU sanctions against Iran and the Democratic People’s Republic of Korea focusing on the worsening of the human rights situation, which run parallel to and supplement the UN sanctions against the regimes of these countries; and thirdly, autonomous EU sanctions that are unrelated to a UN sanctions regime, for example autonomous counter-terrorist sanctions based on Common Position 2001/931/CFSP and the new ISIL (Da’esh)/Al Qaeda regime (Section 5.1 below).18

13 Information drawn from the EURLEX search engine in June 2017, at http://eur-lex. europa.eu. 14 ibid. Five in 2017, 16 in 2016, 18 in 2015, 24 in 2014, 21 in 2013, eight in 2012, two in 2010. Most restrictive measures so far have been annulled for infringing procedural rights, essentially because the Council did not produce the relevant evidence either to the targeted person or the EU Courts. This was found to be incompatible with the right to a fair hearing and the right of access to justice. It also breached the closely related obligation to state reasons. See Sections 4 and 5.4 below. 15 Information drawn from the CJEU’s search engine on 1 June 2017, at http://curia. europa.eu. 16 Case C-266/15 P Central Bank of Iran v Council EU:C:2016:208 and Case C-200/13 P Council v Bank Saderat EU:C:2016:284. 17 ibid, paras 105–106 on the close link between the CFSP decision and the TFEU regulation. 18 T Biersteker and C Portela, ‘EU sanctions in context: three types’ (European Union Institute for Security Studies, 17 July 2015).

210 Research handbook on the EU’s common foreign and security policy Where EU sanctions are adopted to give effect to UN Security Council Resolutions, the EU in principle faithfully follows all changes made to UN sanctions.19 If the UN Security Council Sanctions Committee adds a name, the Commission adds the same name. If the UN Sanctions Committee deletes a name, the Commission deletes that name.20 The UN Sanctions Committee lists persons and entities based on a statement of case submitted by a UN member state.21 It remains the choice of that state what information is made public.22 The information is routinely considered too sensitive to share even with the UN Sanctions Committee. Consequently, the Sanctions Committee often does not actually have information to share with the EU.23 In short, not only does the Commission not ask any questions, there are in fact no structures in which the EU could ask any questions about the substantive justification of UN listings. This external control over an EU policy is also a peculiarity of sanctions. 2.5 Level of National Disobedience The CFSP decision is directly binding on EU Member States and requires them to take the appropriate measures to give effect to its objectives. When CFSP Council decisions set out arms embargoes and travel bans, these measures are directly implemented at national level. By contrast, asset freezes and export bans, i.e. economic measures, are an EU competence. They are implemented by EU regulations. Nonetheless, 16 of the 27 Member States continue to adopt parallel financial sanctions within the framework of their national legislation,24 which are often identical to the EU lists.25 Some national 19

See, e.g., Council Implementing Regulation (EU) 2017/199 implementing Council Regulation (EC) 1183/2005 OJ 2017 L 32/1; Council Implementing Decision 2017/203/CFSP implementing Council Decision 2010/788/CFSP OJ 2017 L 32/22, which changed the identifying information for 21 people and one entity listed on the sanctions list against the Democratic Republic of Congo; Commission Implementing Regulation (EU) 2017/44 amending Council Regulation (EC) 1210/2003 OJ 2017 L 6/36 on Iraq sanctions. 20 UN targeted sanctions regimes are imposed under Charter VII, Article 41 of the UN Charter, binding on all states. 21 Consolidated Security Council Sanctions Committee List. See https://www.un.org/sc/ suborg/en/sanctions/un-sc-consolidated-list. 22 UNSC Res 1735 (2006), para 6. 23 See also A Cuyvers, ‘Give Me One Good Reason: The Unified Standard of Review for Sanctions’ (2014) 5 CML Rev 1759. 24 See also UK Supreme Court, Her Majesty’s Treasury (Respondent) v Mohammed Jabar Ahmed and others (FC) (Appellants); Her Majesty’s Treasury (Respondent) v Mohammed al-Ghabra (FC) (Appellant); R (on the application of Hani El Sayed Sabaei Youssef) (Respondent) v Her Majesty’s Treasury (Appellant) [2010] UKSC 2, para 22. For the UK see Counter-Terrorism Act 2008 and the Terrorist Asset-Freezing Act 2010. For the Netherlands see Sanctieregeling terrorisme 2007-II. 25 For example, the EU and UK lists of individuals and entities threatening the sovereignty and territorial integrity of Ukraine consist of 150 individuals and 37 entities. See http://eurlex.europa.eu/legal-content/EN/TXT/?uri=uriserv:OJ.L_.2017.067.01.0034.01.ENG&toc=OJ:L: 2017:067:TOC and https://www.gov.uk/government/uploads/system/uploads/attachment_data/ file/512570/ukraine_sovereignty.pdf, respectively. The EU and Dutch lists of terrorist suspects giving effect to UNSC Res 1373 are equally identical. See http://www.eeas.europa.eu/cfsp/

The law and practice of EU sanctions 211 laws are, for example, expressly intended to implement UN Security Council Resolution 1373 (2001) but do not refer to the relevant EU laws implementing that Resolution.26 This is a breach of EU law, which prohibits parallel national measures that ‘implement’ directly applicable EU regulations.27 Yet so far the Commission has not introduced any proceedings against those Member States that keep their own national lists.

3. IMPLICATIONS OF SANCTIONS FOR OTHER POLICIES 3.1 Full Set of Judicial Remedies: Strengthening Fundamental Rights and the Rule of Law under CFSP Sanctions have made a significant contribution to the constitutionalization of CFSP, understood as the development and application of common EU norms and principles.28 As identified by Piet Eeckhout, ‘there is a tendency to apply broadly the same constitutional rules, principles, and disciplines to European foreign policy as are applied to the EU’s internal policies’.29 This increasingly includes CFSP. Because of the great amount of sanctions litigation this policy area has given the CJEU an opening to apply general constitutional rules of EU law to a hybrid combination of TFEU regulations that hinge on and effectively copy CFSP decisions. Moreover, the recent case of Rosneft30 is a good example of the Court applying constitutional rules and principles purely to CFSP. It is the first case in which the Court ruled on its scope of jurisdiction over a CFSP sanctions decision31 and it comes as one of four recent rulings of the CJEU in which the Court addressed the scope of its jurisdiction over CFSP more generally.32 In Rosneft, the Court explained that the different entry points of jurisdiction (Article 40 TEU and Article 275(2) TFEU) have different implications under the different types of procedures. The CJEU found that in the absence of any express

sanctions/consol-list_en.htm and https://www.government.nl/documents/reports/2016/01/15/ national-terrorism-list respectively. 26 See e.g. Article 2 of the Dutch Regulation on sanctions for the suppression of terrorism 2007-II (‘Sanctieregeling terrorisme 2007-II’), which refers neither to Council Regulation (EC)2580/2001, OJ 2001 L 344/70 nor to Council Common Position 2001/931/CFSP, OJ 2001 L 344/93. 27 Case 39/72 Commission v Italy EU:C:1973:13; Case 50/76 Amsterdam Bulb BV v Produktschap voor Siergewassen EU:C:1977:13. 28 C Eckes, ‘Common Foreign and Security Policy: The Consequences of the Court’s Extended Jurisdiction’ (2016) 22 European Law Journal 492. 29 P Eeckhout, ‘The Constitutionalization of European Foreign Policy’: Proceedings of the Annual Meeting (American Society of International Law) (2013) 107 International Law in a Multipolar World 171. 30 Rosneft (n 1), paras 66ff. 31 Segi (n 1) was decided under the former third pillar. 32 Case C-455/14 P H v Council et al EU:C:2016:569; Case C-439/13 P Elitaliana SpA v Eulex Kosovo EU:C:2015:753 and Opinion 2/13 EU:C:2014:2454, EU Accession to the ECHR.

212 Research handbook on the EU’s common foreign and security policy limitation, its jurisdiction to review compliance with Article 40 TEU extended to the preliminary ruling procedure.33 This interpretation is applicable across all CFSP policies. Article 275(2) TFEU, by contrast, specifically refers to Article 263(4) TFEU and could hence be interpreted as limiting this entry point to direct actions for annulment. Yet the CJEU, following the Opinion of Advocate General Wathelet,34 found that it also had jurisdiction to rule on the legality of CFSP sanctions decisions in preliminary ruling procedures.35 This is in line with the case of Segi and with the CJEU’s settled case law on the full set of judicial remedies.36 Yet it specifically confirms both a fundamental right and rule-of-law reading of judicial review, which also extends across the whole of CFSP.37 Attempts have been made to broaden structurally the entry point for jurisdiction under Article 275(2) TFEU. The term ‘restrictive measures’ used in this provision is within the EU context a technical term referring to sanctions, such as asset freezing and travel bans. However, fuelled by the dissatisfaction that the Court’s limited jurisdiction under CFSP may leave individuals without the necessary judicial protection, the question arose whether Article 275(2) TFEU could be interpreted much more broadly in order to establish a general rationale that CFSP measures, which adversely affect the rights of individuals in a direct manner, must be subject to judicial review. Textually this came together in the question of how narrowly the term ‘restrictive measures’ in Article 275(2) TFEU should be interpreted. Indeed, the Commission suggested a wide reading covering all direct adverse effects of CFSP policies, which would also have conveniently (albeit perhaps only partially) addressed the problem that the Court’s limited jurisdiction creates in the context of EU accession to the ECHR.38 By contrast,

33

Rosneft (n 1), paras 62–63. See further Chapter 4 in this volume. Rosneft (n 1), Opinion of AG Wathelet, paras 61ff. 35 ibid, paras 71–81. 36 ibid, paras 67–68: it is part of the ‘complete system of legal remedies or procedures that persons bringing proceedings must, when an action is brought before a national court or tribunal, have the right to challenge the legality of provisions contained in European Union acts’. 37 ibid, paras 69–75. 38 Opinion 2/13; H (n 32), Opinion of AG Wahl, para 34: ‘The Commission considers that Articles 24(1) TEU and 275 TFEU should be read as excluding the jurisdiction of the CJEU only with regard to CFSP acts which are an expression of sovereign foreign policy (‘actes de Gouvernement’), and not acts merely implementing that policy. In the alternative, the Commission takes the view that Articles 24(1) TEU and 275 TFEU exclude the CJEU’s review of alleged breaches of CFSP provisions alone, but not of alleged breaches of other EU provisions. Thus, the CJEU would be empowered to review the lawfulness of acts adopted in the framework of the CFSP when the alleged invalidity stems from a possible infringement of non-CFSP provisions. Nonetheless, the present appeal is, according to the Commission, inadmissible for the following reasons: first, the contested decisions cannot be considered mere acts of implementation, since they are of an operational nature; second, the grounds for annulment submitted by the appellant at first instance either required the General Court to interpret Decision 2009/906 (for which that court lacked jurisdiction) or had to be directed against the Italian authorities (and thus submitted in the context of an action lodged before the Italian courts).’ Criticized by AG Wahl in paras 60–66. 34

The law and practice of EU sanctions 213 Advocate General Wahl in the case of H suggested a narrow reading, which the CJEU followed. In his words: I do not believe that the concept of ‘restrictive measures’, although nowhere expressly defined in the Treaties, may, as the applicant suggests, be considered to cover all EU acts which adversely affect the interests of individuals. A textual, systematic and historical interpretation of Article 275 TFEU, in fact, reveals that concept to be of more limited scope.39

Arguably the Court’s jurisdiction should extend to actions for damages for harm caused by CFSP sanctions decisions.40 While non-contractual liability in Article 340 TFEU is applicable under Union law in its entirety this does not as such address the issue of jurisdiction. In this context it is important to realize that the dividing line between the TEU and the TFEU is not easily drawn in the context of sanctions. In a recent sanctions case, for example, the General Court awarded compensation for non-material damage for an unlawful listing of an entity as supporting nuclear proliferation in Iran.41 The Court did so specifically for the reputational damage resulting from the public sanctions measures. However, while the Court stated at the outset that the Council regulation adopted under the TFEU is a direct consequence of the adoption of a CFSP decision under the TEU it did not consider this fact in the context of non-material damages. While this case confirms that an unlawful listing in a TFEU regulation may lead to compensation it does not immediately allow the conclusion that the Court would accept jurisdiction for a standalone CFSP listing. Logically, the CFSP listing caused the reputational damage. It is the first public listing of the applicant’s name and is only copied into the regulation. However, AG Wathelet in Rosneft specifically argued: [a]ctions for damages which relate to a CFSP act are covered by the ‘carve-out’ provision in the last sentence of the second subparagraph of Article 24(1) TEU and the first paragraph of Article 275 TFEU, but not by the ‘claw-back’ provision in the last sentence of the second subparagraph of Article 24(1) TEU and the second paragraph of Article 275 TFEU.42

In practical terms the denial of damages for harm caused by CFSP decisions could leave individuals without compensation even if their listing was found to be illegal by the Court within its exceptional jurisdiction under Article 275(2) TFEU. This would be the case in a situation where the applicants challenged the CFSP decision, rather than also the regulation, or if they were only listed in the CFSP decision but not in the regulation.43 39

H (n 32), Opinion of AG Wahl, para 74. C Eckes, ‘EU Counter-Terrorist Sanctions against Individuals: Problems and Perils’ (2012) European Foreign Affairs Review 113. 41 Case T-384/11 Safa Nicu Sepahan Co. v Council EU:T:2014:986, paras 78–92. See also Case T-293/12 Syria International Islamic Bank PJSC v Council EU:T:2014:439, para 70, where the Council accepted that the Court is competent to establish whether it has jurisdiction to award damages for CFSP measures. 42 Rosneft (n 1), Opinion of AG Wathelet. 43 This was the case for Segi (n 1). 40

214 Research handbook on the EU’s common foreign and security policy 3.2 Institutional Consequences: Closed Evidence Sanctions have in the past regularly been annulled because the Council was unable to substantiate the reasons for listing. This speaks to the need for a closed-evidence procedure. Article 105 of the General Court’s Rules of Procedure, which entered into force on 1 July 2016, sets out that information is ‘confidential’ if publication of that information ‘would harm the security of the Union or that of one or more of its Member States or the conduct of their international relations’.44 This will for the first time give the Council the ability to support their listing decisions with closed evidence. Indeed, the adoption of Article 105 was triggered in particular by the procedural difficulties of dealing with information in sanctions cases.45 However, Member States would have to trust the closed-evidence procedure sufficiently to share national intelligence supporting the decision of their competent national authorities. This should not be taken for granted. The General Court’s new rules of procedure do not specifically refer to sanctions or to CFSP. Information related to the CFSP is therefore treated like any other confidential information. The difficulties of being unable to share confidential information in sanctions cases triggered the creation of a general closed-evidence procedure, which now also allows for more judicial secrecy in other areas. This could ultimately lead to a spill-over of secret judicial proceedings to other, non-CFSP areas. This would be most likely for asylum or criminal law, for example in cases with a national security dimension. 3.3 Implications Under National Law In some Member States an explicit national rule attaches additional consequences to an EU sanctions listing. National law may, for example, exclude accepting refugees who are members or supporters of any of the groups listed under the EU sanctions regime. In a preliminary reference from a German court, the Court of Justice was asked to rule on this consequence of EU sanctions under national asylum law.46 A German asylum authority had decided that membership of an organization included on the terrorist lists justified excluding the person from refugee status. The CJEU held that exclusion from refugee status must be decided on a case-by-case basis, but also stated that inclusion on an EU sanctions list is a ‘factor to be considered’ in evaluating whether someone has committed a ‘serious non-political crime’ or an ‘act against the principles of the UN’.47 EU autonomous counter-terrorist sanctions are often but not always based on national criminal proceedings that are intended to give effect to Framework Decision 44

General Court, Rules of procedure, OJ 2015 L 105/1. The introduction to Chapter 7 of the Rules of Procedure specifically refers to restrictive measures, even if Article 105 is phrased in general terms and could be applied to other security matters. 46 Joined Cases C-57 & 101/09 Bundesrepublik Deutschland v B and D EU:C:2010:661. 47 ibid, para 90. 45

The law and practice of EU sanctions 215 2002/475/JHA on Combating Terrorism.48 Member States retain great discretion over the details of implementation since framework decisions are ‘binding upon the Member States as to the result to be achieved but shall leave to the national authorities the choice of form and methods’.49 The definition of what constitutes a terrorist objective or a link with terrorism is consequently not necessarily identical in the different legal orders, national or European. The Dutch legislature, for instance, added specific provisions on terrorism to the existing criminal law, which allows an increase of the maximum penalty for acts with a terrorist objective by up to 50 per cent.50 This demonstrates the circular interaction and interlocking of the EU and the national legal contexts: EU law requires Member States to take action against terrorist activities, Member States do so, and EU sanctions may then attach additional consequences to the national counter-terrorist actions. At times there is another loop to this circularity where the EU law measure is used to justify a criminal conviction or the prohibition of an organization at the national level.51 In the Netherlands, for instance, all legal persons, listed by the EU pursuant to Common Position 2001/931/CFSP, are prohibited ipso jure.52 In Sweden and in Italy, the courts consider being on the UN list of terrorist suspects a relevant fact when, in the course of criminal proceedings, they determine whether the accused is connected with terrorism or not.53 The circularity is complete when a national decision is adopted ‘pending the adoption’ of an EU measure, then used to justify the adoption of the latter and immediately repealed when the EU measure comes into force.54 Furthermore, the breach of EU sanctions legislation leads to so-called secondary sanctions under national law. Pursuant to Article 83(2) TFEU, minimum rules for secondary sanctions could in principle be established by the Union. However, since this has not yet happened, Member States remain in charge of adopting secondary sanctions. Secondary sanctions do not necessarily have to be criminal but they have to be effective. Again, the specific national provisions differ considerably. In Austria, secondary sanctions are regulated by the national sanctions law.55 Any natural or legal person who makes funds available to a listed individual faces a fine of up to €50,000, and, if more than €100,000 is made available, imprisonment of up to one year or 360 Tagessätze.56 In Germany, deliberate breaches are punishable by imprisonment for a 48

For example, Case T-47/03 Sison v Council and Commission (‘Sison I’) EU:T:2007:207; Case T-341/07 Sison v Council (‘Sison II’) EU:T:2011:687. In these cases, the national decision concerned the rejection of refugee status in the Netherlands rather than criminal proceedings. 49 Article 34(2)(b) TEU (pre-Lisbon). 50 EJ Husabø and I Bruce, Fighting Terrorism through Multilevel Criminal Legislation (Martinus Nijhoff Publishers 2009) 184. 51 See Case C-550/09 Generalbundesanwalt beim Bundesgerichtshof v E and F EU:C:2010: 382. 52 ‘Van rechtswege verboden’, Art 2:20 Section 3 of the Civil Code (Burgerlijk Wetboek). 53 See Sixth report of the Analytical Support and Sanctions Monitoring Team, UN S/2007/132 (8 March 2007), box 1, 14, http://www.un.org/en/ga/search/view_doc.asp?symbol= S/2007/132. 54 See Case T-348/07 Al Aqsa v Council EU:T:2010:373, para 177. 55 See Section 12(1) Sanktionengesetz (SanktG). 56 Fine calculated on the daily rate of income.

216 Research handbook on the EU’s common foreign and security policy period of between six months and five years.57 The ECJ ruled that, where the EU listing has been found to be flawed because the listed individuals had not been given the necessary opportunity to exercise their procedural rights, the listing is illegal and cannot justify secondary sanctions under national law.58 The CJEU held that the EU listing cannot, in any circumstances, be relied upon … as a basis for a criminal conviction in respect of facts relating to that period [in which those listed did not enjoy the necessary procedural rights], without infringing the principle of the non-retroactivity of provisions which may form the basis for a criminal conviction.59

The latter demonstrates the dependence of national criminal secondary sanctions on the EU listings. Yet secondary sanctions can also result in a challenge of the underlying EU listing. In 2017 the CJEU ruled in a preliminary reference concerning secondary sanctions60 that the listing of an organization could be challenged in a preliminary ruling request in the context of national proceedings against secondary sanctions imposed on natural persons for having supported that organization, including after the period for bringing an action for annulment against that listing in the EU Courts had expired. 3.4 EU Counter-Terrorist Sanctions as Part of EU Counter-Terrorist Policies Autonomous EU counter-terrorist sanctions are not only a particular type of sanction but also a particular type of EU counter-terrorist policy. This raises the question of how they interact with other counter-terrorist policies. In a recent preliminary ruling, a Dutch court suggested that the definition of a ‘terrorist act’ within the field of EU sanctions61 should be interpreted in line with the general Framework Decision on Combating Terrorism.62 The CJEU rejected this interpretation.63 Rather than seeing EU counter-terrorist policies as one coherent whole, it focused on the objective and nature of each of the legal instruments. It argued that the objective of the Framework Decision was to approximate the definition of terrorist offences imposing penalties for past conduct and that it fell within the sphere of the Area of Freedom, Security and Justice (AFSJ). The autonomous EU sanctions regime, by contrast, essentially constituted a CFSP policy and imposed preventive rather than punitive measures. The Court also recalled that the relationship between national and EU law is different in both contexts. While the Framework Decision aimed to approximate national criminal law, sanctions were adopted according to a system operating on two levels, in the sense that the Council may include on that list only persons and entities in respect of which a 57

Section 34(4)Nr.2, (5), (6)Nr.4, (7) Aussenwirtschafsgesetz (AWG). Case C-550/09 Generalbundesanwalt beim Bundesgerichtshof v E und F EU:C:2010:382. 59 ibid, para 59. 60 Case C-158/14 A, B, C & D v Minister van Buitenlandse Zaken (A. and others) EU:C: 2017:202. 61 Common Position 2001/931 and Regulation 2580/2001 (n 26). 62 Council Framework Decision 2002/475/JHA, OJ 2002 L 164. 63 A and others (n 60). 58

The law and practice of EU sanctions 217 decision taken by a competent national authority exists.64 In particular, the Court held that the different nature and purpose of the two legal instruments justified the fact that, while the Framework Decision specifies that it does not govern actions by armed forces during periods of armed conflict, the autonomous EU sanctions regime was not subject to that limitation. The CJEU’s fragmented purposive interpretation limits the implications of one policy field for others. The CJEU’s classification of EU sanctions as preventive rather than punitive as an argument for not applying rights protection that is offered under more general counter-terrorist instruments is problematic in and of itself. First, if sanctions remain in place for very long periods of time they restrict rights in a manner comparable to criminal measures, so the classification as preventive should be irrelevant from a rights perspective.65 Secondly, this classification allows a class of counter-terrorist measures that is not subject to the same rights protection as other counter-terrorist measures. This seems not only unnecessary in this particular case, but also undesirable from the perspective of legal certainty, coherence and rights protection. It confirms once more that sanctions are constructed to operate as far as possible outside of the system of rights guaranteed under EU law. The CJEU decided to treat the different legal instruments as reflecting a different internal logic and subjected them to different limits. This results in a situation in which those sanctioned, who are, for example, participants in an armed conflict, may benefit from certain immunities under international humanitarian law with regard to the Framework Decision but are at the same time targeted by EU restrictive measures and secondary sanctions under national criminal law. It also limits the spill-over from TFEU policies to CFSP and vice versa. 3.5 Extraterritorial Effects Sanctions have certain extraterritorial effects. EU citizens are bound by them, irrespective of where they are. So are companies and organizations incorporated under the law of a Member State, including branches of companies in third countries.66 The EU also invites certain third countries to align with its imposed sanctions measures.67 Alignment with EU sanctions is seen as an act of political commitment.68 In the latest renewal of sanctions against Russia over the conflict in Ukraine, only Montenegro, Albania, Norway and Ukraine aligned themselves with the EU. In 2015 this group still included

64

ibid, para 84. C Eckes, ‘EU Restrictive Measures Against Natural and Legal Persons: From Counterterrorist to Third Country Sanctions’ (2014) 51 CML Rev 869. 66 Fact lists EU sanctions, http://www.consilium.europa.eu/uedocs/cms_data/docs/pressdata/ EN/foraff/135804.pdf. 67 The alignment is published in a separate declaration from the High Representatives in a Press release, see e.g. Press release 239/17 of 5 May 2017 regarding the restrictive measures in respect of actions undermining or threatening the territorial integrity, sovereignty and independence of Ukraine. 68 E Hellquist, ‘Either With Us or Against Us? Third-country Alignment with EU Sanctions Against Russia/Ukraine’ (2016) 29 Cambridge Review of International Affairs 997, 1001. 65

218 Research handbook on the EU’s common foreign and security policy Iceland, Liechtenstein and Georgia.69 It is not clear why the latter states did not align themselves with the EU for the specific sanctions regime. At times third countries choose to follow suit with EU imposed sanctions, without publicly aligning themselves.70 Furthermore, non-alignment can be due to time pressure. A request for alignment comes without previous political dialogue and the third country may simply not have enough time to decide if it should align itself.71 This does not rule out the possibility that it will nonetheless implement the sanctions later on.72

4. TENSION BETWEEN LEGAL RULES AND POLITICAL OBJECTIVES The intrinsic tension between legal rules and procedures and political objectives could not be more apparent. The EU institutions claim that ‘[s]anctions are one of the EU’s established tools to promote the CFSP objectives of peace, democracy and the respect for the rule of law, human rights and international law’.73 They consider sanctions as ‘always [forming] part of a wider, comprehensive policy approach involving political dialogue and complementary efforts’.74 The repeated annulments of EU sanctions measures, predominantly for infringing procedural rights of those listed,75 could be read as a situation in which individual rights limit the ability of the EU to take political decisions. Previously, comprehensive state sanctions, such as those imposed on Iraq in reaction to its 1990 invasion of Kuwait, also had a severe humanitarian impact. They suffered from a great lack of precision in targeting those who contributed the most to the situation that triggered the imposition of the sanctions in the first place. The move from measures of general application, such as comprehensive state embargoes, to measures targeting specific persons that restrict their rights at a level of severity comparable to criminal measures, was intended to limit humanitarian consequences and fundamental rights violations. Yet this move also creates more far-reaching procedural guarantees for those sanctioned in a Union of law. This may result, somewhat counterintuitively, in a situation where it is more difficult for the Council to defend the measure with more limited fundamental rights consequences (targeted sanctions) in court than the measure with more far-reaching fundamental rights consequences (trade embargoes). 69 Press release 622/15 of 28 June 2015 regarding restrictive measures in response to the illegal annexation of Crimea and Sevastopol. 70 Sanctions may have more effect for one country than for another. Iceland suffered from the Russian boycott of Icelandic fish. After 2016 it stopped publicly aligning itself with the EU sanctions. See B Thorhallsson and P Gunnarsson, ‘Iceland’s Relations with its Regional Powers: Alignment with the EU-US sanctions on Russia’ (Working Paper 874, NUPI 2017). 71 ibid, 24, confirmed by interviews with Icelandic officials. 72 For Iceland see https://www.mfa.is/foreign-policy/sanctions/. For Liechtenstein see http:// www.llv.li/#/114812?scrollto=true. 73 See https://eeas.europa.eu/headquarters/headquarters-homepage_en/423/Sanctions%20 policy. 74 ibid. 75 See n 14 above and Section 5.4 below.

The law and practice of EU sanctions 219 4.1 Relisting In practice, the EU institutions have been able to maintain the sanctions despite repeated annulments by the EU Courts. The Council regularly relisted natural and legal persons after the EU Courts had annulled their listing. This may be justified and reasonable if the procedural flaws of the listing decision can be remedied in the relisting decision. However, it is at least prima facie less justified if the reason for annulment was that the reasons for listing could not be substantiated and that the reasons for listing are only reformulated in the relisting decision. The Council is not obliged to demonstrate that the relisting is based on new or newly discovered relevant facts. Nor does the Council have to specifically explain why the person is relisted or why the information was missing in the original listing. The specific example of Mr Kadi has attracted the most attention in this regard. Mr Kadi had not been delisted following the decisions of the CJEU of 3 September 200876 as the Court had maintained the effects of the listing for up to three months. Subsequently, the Commission provided the narrative summary of reasons provided by the UN to list Mr Kadi on the UN Sanctions List and allowed Mr Kadi to comment on the narrative summary.77 However, it maintained the position that Mr Kadi should be included on the EU sanctions list due to his association with the Al Qaeda network.78 The Commission did not and most likely could not produce new or better reasons for listing Mr Kadi since his case concerned EU sanctions giving effect to a decision of the UN Security Council Sanctions Committee. Finally, in October 2012, the EU delisted Mr Kadi, following the delisting by the Security Council only days earlier.79 The regular relistings have been the means deployed by the Council to prevent the Court’s interventions from interfering with the political objectives of these sanctions. This has been facilitated by the fact that the Court regularly maintains the effects of the annulled decisions for a transition period, presumably with the aim of allowing the Council to relist. Consequently, while the regular annulments may have negatively impacted on the credibility of the EU’s sanctions policy, they have not reduced the legal or factual effects of asset freezing or travel bans.

76

Joined Cases C-402 & 415/05 P Yassin Abdullah Kadi and Al Barakaat International Foundation v Council of the European Union and Commission of the European Communities EU:C:2008:461. 77 Case T-85/09 Yassin Abdullah Kadi v European Commission EU:T:2010:418, paras 53–57. 78 Commission Regulation (EC) 1190/2008, amending for the 101st time Council Regulation (EC) 881/2002 imposing certain specific restrictive measures directed against certain persons and entities associated with Usama bin Laden, the Al-Qaida network and the Taliban, OJ 2008 L 322/25. 79 Commission Implementing Regulation (EU) 933/2012 amending for the 180th time Council Regulation (EC) 881/2002 imposing certain specific restrictive measures directed against certain persons and entities associated with the Al Qaeda network, OJ 2012 L 278/11.

220 Research handbook on the EU’s common foreign and security policy 4.2 General Policy Objectives and Specific Fundamental Rights Restrictions The political objectives, including the symbolic or signalling functions of sanctions regimes, cannot be easily explained without raising difficult fundamental questions of justice. Does the EU legal order allow justifying measures that strongly interfere with the fundamental rights of targeted individuals with general policy objectives? While sanctions decisions essentially concern a familiar weighing of the common good and the rights of individuals, it is difficult to justify them in a Kantian ethical framework using individuals as means to serve a more general purpose. However, a combination of different purposes for a rights-sensitive measure is widely accepted, including for criminal punishment. Part of the motivation to punish can be the protection of the population, including from other perpetrators who are discouraged from breaching the law because they fear the example of those punished. Yet any punishment finds its limits in the severity of the criminal act, the damage and the subjective responsibility of the perpetrator. This is where the difficulties in the context of sanctions begin. The limits of subjective responsibility are in general terms guaranteed by the proportionality principle. Yet the proportionality principle is not able to ensure these limits in the context of sanctions because of the way it is interpreted in this context.80 The tension between the general purpose and the individual rights restriction may also be one of the reasons why the EU does not identify the objectives more clearly and make an argument as to how any specific objective is served by targeting a particular person. 4.3 Change in Behaviour The 2012 EU Sanctions Guidelines state that ‘[i]n general terms, restrictive measures are imposed by the EU to bring about a change in policy or activity by the target country, part of country, government, entities or individuals, in line with the objectives set out in the CFSP Council Decision’.81 Sanctions are aimed at changing the ‘policy’ of the third-country regime or the ‘activity’ of those targeted, i.e. to achieve a change in behaviour. The EU has suspended sanctions for political developments in the country. This has, for example, been the case for Belarus and Zimbabwe.82 However, there is no publicly available information that the EU has ever delisted an individual from any of the targeted sanctions lists because that person or entity has demonstrated a change in behaviour. The aim of achieving a change in behaviour is highly problematic in the context of sanctions regimes, in particular with regard to those persons who are more remote from the regime.83 The generality of the policy objectives of the legal instruments, e.g. aiming for ‘democratization’ of a country, makes it in fact impossible to link the 80

See Section 5.4 below. Restrictive measures guidelines (n 3), II.A.4. 82 See, for example, Council Decision 2013/160/CFSP amending Decision 2011/101/CFSP concerning restrictive measures against Zimbabwe, OJ 2013 L 90/95. 83 The problems connected with the assessment of a change in behaviour were brought to the fore by the CJEU in Tay Za with regard to sanctions against the most remote group of targets, family members of those suspected of economically supporting and benefiting from the regime in focus: Case C-376/10 P Tay Za v Council EU:C:2012:138, para 67. 81

The law and practice of EU sanctions 221 objectives to the behaviour of individuals in the vast majority of cases. Is it possible to demonstrate that someone is no longer supporting terrorism or a third-country regime, in particular if the original support was economic and that person’s funds are frozen? Or does their listing ultimately depend on a policy change of the third-country regime, as in the case of Myanmar, or a regime change, as in the case of Iraq or Libya?

5. NEW TRENDS? 5.1 Ever More Sanctions, EU Powers and Objectives Since their inception within the EU context in 2002, instruments and objectives used in the EU’s targeted sanctions policy have multiplied. Currently, the EU has restrictive measures in force against 30 countries, and additionally separate sanctions regimes against terrorist groups, most significantly against persons and entities associated with Al Qaeda and ISIL (Da’esh), but also against terrorist offences in general.84 Of these 30 regimes, six were implemented before 2000, eight between 2000 and 2010, and 16 after 2010. A recent example of an autonomous EU sanctions regime is that adopted against ISIL (Da’esh)/Al Qaeda in September 2016. This new regime is targeting foreign fighters, persons travelling outside the EU to join the cause of terrorism.85 It will allow the EU for the first time to designate people for being associated with terrorism (ISIL (Da’esh)/Al Qaeda) who have not been previously identified by the UN or by the competent authorities of the Member States. Until the adoption of this new regime, all listing decisions under the autonomous EU counter-terrorism sanctions regime had been taken in a composite administrative procedure in which national authorities took the relevant decision, which then triggered the EU listing. The EU, by contrast, had only been competent to draw up listings without input of national authorities in the context of regime sanctions. As these counter-terrorist sanctions are not related to any political regime governing any geographically identifiable territory, they constitute the most indeterminate sanction tool the EU deploys. This new regime should therefore be seen as an extension of the powers of the Council. The objectives that targeted sanctions regimes pursue have equally multiplied. Besides internal conflicts, non-proliferation, counter-terrorism, democratization and protection of civilians, including their human rights, have become objectives of sanctions imposed by both the UN and the EU. The EU sanctions against Russia are an example of a sanctions regime that is characterized by a great diversity of objectives and measures. They were introduced in March 2014.86 In the first year, not only the list 84

European Commission, ‘European Union, Restrictive measures (sanctions) in force’. Last updated 26 April 2017, https://eeas.europa.eu/sites/eeas/files/restrictive_measures-2017-04-26clean.pdf. 85 Council Regulation (EU) 2016/1686 OJ 2016 L 255/1 and Council Decision 2016/1693/ CFSP repealing Common Position 2002/402/CFSP OJ 2016 L 255/25; see in particular the last alternative of Article 3(1). 86 Council Decision 2014/145/CFSP OJ 2014 L 78/16.

222 Research handbook on the EU’s common foreign and security policy of persons and entities, but also listing criteria were amended every month.87 The following range of prohibited actions and protected objectives gives a good impression of increasing diversification of objectives and measures: protection of the territorial integrity, sovereignty and independence of Ukraine; the misappropriation of public goods; a ban on imports of goods originating in Crimea or Sevastopol unless they have Ukrainian certificates; a prohibition on investing in Crimea;88 a ban on providing tourism services in Crimea or Sevastopol;89 goods and technology for the transport, telecommunications and energy sectors or the exploration of oil, gas and mineral resources may not be exported to Crimean companies or for use in Crimea; technical assistance, brokering, construction or engineering services related to infrastructure in the same sectors must not be provided.90 5.2 Institutional Tug-of-War Since the entry into force of the Lisbon Treaty, the EEAS has formally been given an important institutional role in the adoption of sanctions. With regard to autonomous sanctions in particular, the 2012 Guidelines stipulate that the EEAS ‘should have a key role in the preparation and review of sanctions regimes as well as in the communication and outreach activities accompanying the sanctions, in close cooperation with Member States, relevant EU delegations and the Commission’. The review is conducted by the relevant Council working parties and committees, where appropriate on the basis of EU Heads of Mission reports.91 Heads of Missions may, for instance, help to ensure unambiguous identification of the targeted persons92 and are ‘invited to provide, where appropriate, their advice on proposals for restrictive measures or additional designations’.93 The strong involvement of the EEAS underlines the hybrid CFSP/TFEU nature of sanctions since the EEAS is equally of a hybrid nature, dealing with both CFSP and

87

See for an overview of all the amendments: Restrictive measures in force (n 84) 115. Europeans and EU-based companies can no longer buy real estate or entities in Crimea, finance Crimean companies or supply related services. In addition, they may not invest in infrastructure projects in six sectors. 89 European cruise ships may not call at ports on the Crimean peninsula, except in case of emergency. This applies to all ships owned or controlled by a European or flying the flag of an EU Member State. 90 See Council Regulation (EU) 692/2014, OJ 2014 L 183/9, last amended by Council Regulation (EU) 1351/2014 concerning restrictive measures in response to the illegal annexation of Crimea and Sevastopol; see for the other sanctions regimes Council Regulation (EU) 269/2014 OJ 2014 L 78, last amendment of the criteria for listing in Council Regulation 959/2014 and Council Regulation (EU) 208/2014, OJ 2014 L 66/1 concerning restrictive measures against persons. 91 Council, Guidelines on implementation and evaluation of restrictive measures (sanctions) in the framework of the EU CFSP, 11205/12 of 15 June 2012 (EU Sanctions Guidelines). 92 ibid, II.D.22. 93 ibid, Annex I Recommendations for working methods for EU autonomous sanctions, para 3. 88

The law and practice of EU sanctions 223 TFEU policies.94 The EEAS involvement, with its focus on external policies, including CFSP, does not make the use of a legal basis under the AFSJ (Article 75 TFEU) more likely. In November 2016 the Foreign Relations Counsellors Working Party (RELEX) agreed that the assessment of the impact and functioning of autonomous EU sanctions against ISIL (Da’esh)/Al Qaeda and the examination of appropriate possible improvements should be entrusted to the former CP 931 Working Party with an enlarged scope and a new name: COMET WP.95 COMET WP has the mandate to ‘examine and evaluate information with a view to listing and de-listing of persons, groups, undertakings and entities, as well as assess whether the information available meets the [relevant] criteria’, ‘make recommendations for listings and de-listings’, ‘prepare the regular review’, and ‘assess the impact and functioning’. COMET WP largely works in secret.96 The six-monthly rotating Council Presidency chairs it. The documents are held by the Council Secretariat. As mentioned above, the ISIL (Da’esh)/Al Qaeda regime provides for the first time the possibility for the EU itself to take autonomous listing decisions that are not based on UN lists or on the decision of competent national authorities. This may justify a new institutional arrangement. Yet these new institutional developments are also a clawing back of competences from the EEAS.97 5.3 Ever More Vague Listing Criteria While the sanctions criteria of counter-terrorist sanctions have always been very broad and arguably rather vague, for example requiring an association with an alleged terrorist organization, the sanctioning criteria used in regime sanctions have equally become broader and more vague over time. An example is the sanctions regime against Iran, which has been fundamentally adapted to make the sanctions more likely to stand up in court. They started with a focus on individuals who contribute to the proliferation of nuclear materials before shifting more generally to those supporting the government of Iran. Two appeal cases before the CJEU specifically examined the criterion of ‘support to the Government of Iran’ set out in Article 20(1)(c) of Council Decision 2010/413/CFSP, as amended by Council Decision 2012/635 CFSP.98 The ECJ explained that the 94

S Blockmans and C Hillion (eds), EEAS 2.0: A Legal Commentary on Council Decision 2010/427/EU Establishing the Organisation and Functioning of the European External Action Service (CEPS 2013), available at https://www.ceps.eu/publications/eeas-20-legal-commentarycouncil-decision-2010427eu-establishing-organisation-and. 95 Council Foreign Relations Counsellors Working Party, ‘Fight against the financing of terrorism – Establishment of a Council Working Party on restrictive measures to combat terrorism (COMET WP)’ (23 November 2016), 14612/1/16 REV 1, http://www.statewatch.org/ news/2016/dec/eu-council-comet-terrorist-lists-wp-14612-16-rev1.pdf. 96 See ibid, practical arrangements on page 4: even organizational information of the working party falls short of being classified but is labelled ‘restraint’, which in actual fact makes it inaccessible. 97 Compare also: EEAS 2.0: A Legal Commentary (n 94). 98 Central Bank of Iran (n 16); Case C-440/14 P National Iranian Oil Company v Council EU:C:2016:128.

224 Research handbook on the EU’s common foreign and security policy purpose behind the addition of that criterion was to target the relevant person or entity’s own activities which, even if they have no actual direct or indirect connection with nuclear proliferation, are nonetheless capable of encouraging it by providing the Government of Iran with resources or facilities of a material, financial or logistical nature which allow it to pursue proliferation activities.99

It further upheld the General Court’s conclusion that financial services ‘such as the holding of accounts, the performance and conclusion of financial transactions or the purchase and sale of bonds, constitute material, logistical and financial support to that State and, as a result, support to the Government of that State’, qualified as financial support even if the Central Bank of Iran did not place ‘its own financial resources at the disposal of the Government of Iran’.100 This considerably extends the circle of targeted persons. Another recent example is the sanctions regime against Syria. Article 3(1) and Article 4(1) of Decision 2011/273 provided for the adoption of restrictive measures against persons responsible for the violent repression of the civilian population in Syria and persons associated with them, as listed in the annex to that decision.101 The following section will, inter alia, discuss how broadly this particular listing criterion is interpreted. 5.4 Trends in Litigation The Court’s decisions on the legality of sanctions measures occur in the following setting. The Council designates an individual, sets out a number of reasons, and the individual challenges the listing in an action for annulment. The applicant usually disputes, amongst other things, the factual accuracy of the listing reason put forward by the Council. The right to judicial review under Article 47 EU Charter of Fundamental Rights requires that the ECJ is in a position to ensure that the listing decision, which affects that person individually, is taken on a sufficiently solid factual basis. That entails, in this instance, a verification of the factual allegations in the summary of reasons underpinning the contested acts, in order to review whether those reasons, or, at the very least, one of those reasons, deemed sufficient in itself to support those acts, is substantiated.102

99

Central Bank of Iran (n 16), para 44. ibid, paras 45–46. 101 Council Decision 2011/273/CFSP concerning restrictive measures against Syria, OJ 2011 L 121/11. 102 Case C-193/15 P Tarif Akhras v Council ECLI:EU:C:2016:219, para 56. With reference to the judgments in Joined Cases C-584, 593 & 595/10 Commission and Others v Kadi EU:C:2013:518, para 119; Case C-630/13 Anbouba v Council EU:C:2015:247, para 46; and Case C-605/13 Anbouba v Council EU:C:2015:248, para 45. 100

The law and practice of EU sanctions 225 The burden of proof lies with the Council to demonstrate that the listing criteria are met.103 Most annulments then happen because the Council is unable or unwilling to produce evidence, or even information and clues, supporting the reasons for designation. Recent cases, for instance regarding the sanctions regimes against Russia and Zimbabwe, continue to turn on similar issues as previous case law in the context of other sanctions regimes: choice of the legal basis, manifest error of assessment, obligation to state reasons, rights of the defence, fundamental rights, and proportionality. Yet one new trend is the bringing together of large number of individuals and entities in collective challenges.104 The most relevant trend, however, is that the litigation statistics have improved in two regards: fewer cases are brought and of these the Council wins a higher proportion. The reasons for this are not so much that the Council is producing more or better evidence. The reason rather lies in the case law of the EU Courts itself, in which the Court has accepted that certain legal constructions, such as presumptions based on inferences, can lead to a reversal of the burden of proof. Other circumstances, such as the construction of the proportionality test, stand in the way of strict scrutiny of the specific listing, as opposed to the general policy decision that sanctions are imposed. Finally, the Court considers it sufficient if the Council substantiates one out of the many reasons brought forward to justify a listing. As to the first point, the CJEU appears to have changed its position on the acceptability of presumptions that the targeted individual meets the listing criteria, which are not codified under secondary law and which lead to a reversal of the burden of proof based on inferences. In 2012 the Grand Chamber of the ECJ overturned the General Court’s position in the case of Tay Za v Council that the family members of those in charge of businesses could be targeted by sanctions based on a presumption (for which no provision was made in the relevant secondary law) that they benefited from the economic policies of the targeted regime. The ECJ held: [b]y finding that it may be presumed that the family members of leading business figures benefit from the functions exercised by those businessmen, so that such family members also benefit from the economic policies of the government, and that there is therefore a sufficient link … between the appellant and the [targeted regime] the General Court erred in law.105

Whether or not presumptions based on inferences should be permissible as a matter of principle is a different question. The case law demonstrates a development in the Court’s position on their acceptability. In the more recent case of Afrasiabi, the ECJ emphasized the need for a contextual assessment106 and the difficulties that the Council 103

See for two cases that concerned inter alia the burden of proof: Case C-72/11 Afrasiabi and Others ECLI:EU:C:2011:874 and Case C-630/13 P Anbouba v Council EU:C:2015:247. 104 See, e.g., a case brought by 109 individuals and 12 companies against their designation on the sanctions list against Zimbabwe: Case T-190/12 Tomana and others v Council and Commission EU:T:2015:222; appeal: Case C-330/15 P Tomana and Others v Council and Commission EU:C:2016:601. 105 Case C-376/10 P Tay Za v Council EU:C:2012:138, paras 69–71. 106 ibid, para 60.

226 Research handbook on the EU’s common foreign and security policy encounters in producing evidence ‘because of the state of war that prevails in Syria’.107 It held that ‘the Council discharges the burden of proof that lies on it if it presents to the [EU Courts] a set of indicia sufficiently specific, precise and consistent to establish that there is a sufficient link between the person subject to a measure freezing his funds and the Syrian regime’.108 This led the Court to the conclusion that the applicant’s status as a businessman and his leading positions in the networks of Syrian businessmen such as the Chambers of Commerce, and his role as a representative of Syrian businessmen constituted such ‘a set of indicia sufficiently specific, precise and consistent to establish that [the applicant] was providing economic support to the Syrian regime or benefiting from it’.109 Accordingly, while in Tay Za a presumption that family members of those associated with a regime benefit from that closeness to the regime was rejected, in Afrasiabi the presumption that a well-connected businessman in Syria is supporting the regime or benefiting from it was accepted. Moreover, the ECJ explicitly confirmed that a set of indicia is sufficient to discharge the burden of proof that lies on the Council when taking a listing decision. Effectively this amounts to a threshold that is lower than ‘probable cause’. As to the second issue, the ECJ confirmed that the starting point for the test of whether sanctions are proportionate could be the abstract objective of maintaining international peace and security. It upheld the General Court’s proportionality assessment of the Iranian regime sanctions. The General Court had first acknowledged that the measures caused considerable harm to the applicant, both financially and to its reputation.110 When assessing the proportionality, the General Court had then simply held that ‘the difficulties caused to the applicant as a result of the contested acts are not disproportionate to the importance of the aim of maintaining international peace and security that is pursued by those acts’.111 The remainder of the proportionality assessment was dedicated to explaining how the harm to the applicant had been limited in the specific case. The General Court had not considered to what extent – in terms of quality or quantity – the applicant’s support of the regime with quite general financial services had contributed to the overpowering general aim of maintaining international peace and security. This is an indirect way of limiting judicial review: by taking weighty abstract objectives as the yardstick the proportionality test is voided of meaning.112 As to the third point, in the case of Afrasiabi the Council had first put forward reasons for listing including that the applicant, Mr Akhras, besides being a prominent businessman benefiting from and supporting the Syrian regime and a member of the Board of the Federation of Syrian Chambers of Commerce, was the founder of the Akhras group, had close business relations with President Assad’s family and provided industrial and residential premises for improvised detention camps and logistical 107 108 109 110 111 112

ibid. ibid. ibid. ibid, para 118. ibid, para 119. See Eckes (n 65).

The law and practice of EU sanctions 227 support for the regime.113 In the proceedings the Council could only substantiate the most general reason, namely that he was ‘a prominent businessman who is part of the economic ruling class in Syria’.114 It should, however, be added that the other originally stated reasons for listing, for which the Council did not or could not put forward any evidence, could arguably be seen to have caused the most reputational damage. By confirming that substantiating one of the reasons for listing is sufficient to make the listing lawful, the CJEU effectively opens the door to a practice of giving a combination of more specific and more general reasons that damage the listed person’s reputation to different degrees, without ever having to put forward any evidence for most and potentially the gravest of these reasons.

6. CONCLUSION AND RECOMMENDATIONS EU sanctions targeted at specific individuals, be they sanctions against terrorist suspects or regime supporters, have replaced comprehensive state sanctions. At the same time, the combination of ever more vague listing criteria and the fact that the ECJ accepts sets of indicia as sufficient to justify a listing also reduces the targeted nature of sanctions and allows them to target categories of people. What seems fairly certain is that for the moment sanctions are here to stay. However, the EU sanctions policy continues to raise fundamental rights concerns. Some of these concerns could be addressed by taking Declaration 25 on Articles 75 and 215 TFEU, as it was attached to the Lisbon Treaty, more seriously. This Declaration reads as follows: The Conference recalls that the respect for fundamental rights and freedoms implies, in particular, that proper attention is given to the protection and observance of the due process rights of the individuals or entities concerned. For this purpose and in order to guarantee a thorough judicial review of decisions subjecting an individual or entity to restrictive measures, such decisions must be based on clear and distinct criteria. These criteria should be tailored to the specifics of each restrictive measure.

This does not seem to leave room for a combination of a presumption of association of a regime in the broadest sense possible because someone is an economically active person under this regime, or even a family member of such an economically active person. The requirements of tailoring the sanctions criteria to the specific regime should be translated into a requirement that they are directly linked to the objectives of that regime. Since targeted sanctions were introduced in 2002, the adoption procedures, including the statement of reasons and notification, have improved. Yet it is difficult for the EU to create a procedure that stands up in a Union of law, in which individuals are subjected to far-reaching rights restrictions, possibly without any personal wrongdoing, in order to serve political objectives. This was possible for comprehensive state sanctions 113

Case C-72/11 Afrasiabi and Others EU:C:2011:874, para 58. See S Blockmans, ‘Curbing the Circumvention of Sanctions Against Iran Over Its Nuclear Programme: Afrasiabi’ (2013) 50 CML Rev 623. 114 ibid, 59.

228 Research handbook on the EU’s common foreign and security policy because measures against entire states belong to the realm of international law and international relations and are by definition political, even if they entail far-reaching fundamental rights implications for large parts of a population. This tension must be addressed with greater care. The specific sanctions regime should state as precisely as possible the link between the CFSP objectives as set out in Article 21 TEU: the protection of peace and democracy and respect for the rule of law, human rights and international law, and the measures imposed on specific individuals or groups of people. The EU Guidelines should attempt to explain in more detail what could constitute a change in policy or behaviour in this context that should lead to delisting. One requirement should be that only those who are actually in a position to contribute to achieving the objectives – in terms of the quality and quantity of the alleged support – could justifiably be made targets of sanctions. Furthermore, any change can only be evaluated against the backdrop of previous behaviour. This seems impossible if the person was listed on the basis of an inferred presumption of involvement. The tension between the policy goals and the specific restriction of the rights of individualized persons may also be the reason why the EU Courts have constructed a level of judicial review which leads to fewer annulments and ultimately to less litigation being brought against EU sanctions. Indeed, the combination of the presumptions based on inferences that place the burden of proof effectively on those to whom the presumption applies, the broader and more vague listing criteria, and the acceptance of a test of proportionality that weighs any individual rights infringement against the overpowering objective of international peace and security, have made sanctions litigation less successful. It has simply lowered the threshold of justification. The consequence that sanctions have for the legal culture, as well as for trust and confidence in the legal system, must be considered. These costs go beyond the infringement of human rights in any particular case. Sanctions should not set dangerous precedents of quasi-criminal charges located in the grey zone of criminality. The EU should for reasons of credibility commit to meeting the procedural standards of criminal law, even if the general position of the EU institutions and the Court remains that sanctions are preventive and not punitive. Yet the case law of the Court, which effectively lowers the burden of proof, gives the Council room to go in the opposite direction. Additionally, detailed and repeated impact assessments which consider legal and political costs should be made for each sanctions regime. Public and private costs of sanctions should also be separately considered in these assessments. The latter are often disregarded and even sometimes the private bodies and individuals, such as financial institutions,115 are unaware of their full extent.116 In the globally interconnected world, financial institutions, for instance, not only have to conduct customer name checks 115

For an impression of the complexity of combating the financing of terrorism, see World Bank, Combating Money Laundering and the Financing of Terrorism – A Comprehensive Training Guide: Workbook 1. Effects on Economic Development and International Standards (World Bank 2009); House of Lords European Union Committee, ‘Money Laundering and the Financing of Terrorism’ (HL 2008-09 132-I). 116 W Wensink, M van de Velde and L Boer, Estimated Costs of EU Counterterrorism Measures (European Parliament Directorate-General for internal policies 2011).

The law and practice of EU sanctions 229 against numerous parallel and complementary sanctions lists, they also have to monitor to whom customers transfer funds and from whom they receive funds. Finally, measures like the new autonomous EU sanctions against ISIL (Da’esh)/Al Qaeda, in which the EU actually designates individuals truly autonomously, should be based on Article 75 TFEU rather than Article 215 TFEU. This would effectively remove this type of sanction from the realm of CFSP and bring them within the AFSJ. Under both Article 75 and Article 215(2) TFEU, sanctions are adopted by the Council acting by qualified majority following a Proposal of the Commission (Article 75 TFEU) or a Joint Proposal from the Commission and the High Representative (Article 215 TFEU). However, since Articles 215 and 75 TFEU set out very different procedures for the basic policy decision (CFSP decision or AFSJ framework, respectively), the choice of the legal basis is crucial not only for the division of competences between the EU and its Member States, but also for the influence of the different EU institutions. The involvement of the European Parliament would increase democratic scrutiny. This is particularly relevant where the EU institutions take the substantive listing decision, rather than rubber-stamping the decisions of competent national authorities. As has been the case under the Al Qaeda regime up to now, the adoption of a legal framework pursuant to the rules and standards of the ordinary legislative procedure is necessary to avoid the impression of arbitrary exercise of executive power.

11. The nexus between the CCP and the CFSP: achieving foreign policy goals through trade restrictions and market access Andrea Ott and Guillaume Van der Loo

1. INTRODUCTION: THE KANTIAN TRIANGLE In contrast to the development–security nexus1 and the undisputed links between trade and development,2 the relationship between CFSP and trade is less visible and has attracted less scholarly attention.3 However, it can be argued that trade and foreign affairs have formed an intrinsic bond since the inception of the EU as a value-based regional trade organization. Free and fair trade is a precondition for peace and stability between nations4 and is thus part of the ‘genetic code’ of the EU. The establishment of a customs union and an internal market has served to achieve and maintain peace and prosperity between its members.5 Coined the Kantian Triangle, political science theory argues that in international relations, democracies, economic interdependence and cooperation in international organizations mutually reinforce peace and stability.6 Exporting the EU’s values of peace and prosperity is confirmed explicitly by the Lisbon Treaty. The Union’s external action, and specifically the Common Commercial Policy (CCP), is guided by the principles of promoting peace, its values and the well-being of its people (Arts 3(5) and 21(1) TEU). And these values shall be pursued in a ‘coherent perspective for the EU’s external action as a whole’ by deploying CFSP 1 On this issue, see Chapter 12 in this volume; Hans Merket, The EU and the SecurityDevelopment Nexus (Brill 2016); Mark Furness and Stefan Gänzle, ‘The Security–Development Nexus in European Union Foreign Relations after Lisbon: Policy Coherence at Last? (2016) 35(4) Development Policy Review 475. 2 Maurizio Carbone and Jan Orbie, The Trade-Development Nexus in the European Union – Differentiation, Coherence and Norms (Routledge 2014). Lorand Bartelt, ‘Trade and Development Policy of the European Union’ in M. Cremona (ed.), New Developments in the EU’s External Relations (OUP 2008). 3 Eeckhout points out this awkwardness, see Piet Eeckhout, EU External Relations Law (OUP 2011) 35. On this issue, see also Fabienne Bossuyt, Lotte Drieghe and Jan Orbie, ‘Living Apart Together: EU Comprehensive Security from a Trade Perspective’ (2013) 18 European Foreign Affairs Review 63. 4 This ‘trade promotes peace’ theory is based on the writings of Immanuel Kant and Adam Smith. 5 See in this regard the speech by Jean Monnet to the French National Liberation Committee, 5 August 1943 or the Harvard Speech by Secretary of State George Marshall on 5 June 1947 on the Marshall Plan triggering European unification post-1945. 6 John R. Oneal and Bruce Russett, Triangulating Peace: Democracy, Interdependence, and International Organizations (W.W. Norton 2001).

230

The nexus between the CCP and the CFSP 231 instruments alongside trade instruments.7 Such lofty and ambitious statements articulated in primary rules and policy documents need to be backed up by regulatory practice. The CCP–CFSP nexus operates under the wider context of, on the one hand, trade forming a vehicle for achieving peace and stability in third countries and aiming for coherence between external action values and, on the other hand, the EU normative trade system acting in and complying with the wider framework of the World Trade Organization (WTO) rules. The fundamental principles of national treatment and non-discrimination enshrined in the WTO Agreements restrict its members in their pursuit of a ‘politically tainted’ trade policy.8 For example, the 2003 Kimberley process, in which the EU participates,9 introduced a certification, import and export control system to stem the trade in ‘blood diamonds’ and hence contributed to conflict prevention in Africa. This practice is, however, not compatible with the WTO principles and required a waiver by the WTO members.10 The WTO framework, as discussed in the last section of this chapter, also restrains the EU’s use of unilateral trade measures for political purposes. This chapter will analyse the normative framework to create a coherent EU foreign policy11 and streamline CFSP aims with trade aims. This normative framework is characterized by the dichotomy between achieving CFSP goals through traderestrictive measures and trade-facilitation measures. Whereas trade-restrictive measures are to be understood as all EU policy measures that directly or indirectly limit or confine trade (in general or towards a specific country), trade-facilitation measures are considered to be all policy measures that directly or indirectly promote trade or offer preferential market access. The former encompasses, for instance, not only trade embargoes or the suspension of preferential market access (e.g. FTAs) but also the measures that limit the modus operandi of the CCP to bring this policy in line with EU foreign policy objectives (e.g. sustainable development) or international obligations (ratification of the Arms Trade Treaty of 2012). The latter group of measures covers, for example, EU trade agreements concluded for predominantly political purposes (e.g. FTAs included in association agreements) or unilateral preferential market access offered through the GSP+ regime or autonomous trade measures (ATMs). This chapter 7

High Representative of the Union for Foreign Affairs and Security Policy, Implementing the EU Global Strategy, Year 1 – report, 2017. 8 Art. XXI(c) GATT allows WTO members to take action in line with the obligations under the UN Charter and can be only be exceptionally enabled through waivers, see Isabel Feichtner, ‘The Waiver Power of the WTO: Opening the WTO for Political Debate on the Reconciliation of Competing Interests’ (2009) European Journal of International Law 615. See also TBT Agreement, which applies on technical standards and regulations which find exceptions in ‘national security requirements; the prevention of deceptive practices; protection of human health or safety, animal or plant life or health, or the environment; fundamental climatic or other geographical factors; fundamental technological or infrastructural problems’. 9 Gloria Fernández Arribas, ‘The European Union and the Kimberley Process’ (CLEER Working Paper No. 3, TMC Asser 2014) . 10 Council Regulation 2368/2002 of 20 December 2002, OJ 2002 L 358/28. First waiver granted in 2003. Decision of 11 December 2012 extends the waiver for the third time from 1 January 2013 until 31 December 2018, WTO, WT/L/876, 14 December 2012. 11 Stefan Oeter, ‘Art. 21 TEU para. 1’ in Herman-Josef Blanke and Stelio Mangiameli (eds), The Treaty of the European Union (TEU), a Commentary (Springer 2013).

232 Research handbook on the EU’s common foreign and security policy will first explore the legal (Section 2) and policy (Section 3) dimensions of the CCP–CFSP nexus. Trade-restrictive measures are then discussed (Section 4), focusing on the essential elements clauses in international agreements (4.1) measures that implement international law obligations (4.2) and the EU’s value-based trade agenda (4.3). Finally, the trade facilitating measures are analysed (Section 5). In particular, the EU’s GSP+ scheme (5.1) and ATMs (5.2) as foreign policy instruments are discussed and compared.

2. THE LEGAL DIMENSION OF THE TRADE–FOREIGN POLICY NEXUS Since the establishment of the European Economic Community (EEC), the CCP has been considered a crucial and central policy for the Union’s external relations, equipping the EEC, as it then was, with an effective tool to pursue policy goals that go beyond mere trade objectives. This, however, was an incremental process over the years, for institutional and policy reasons. The CCP evolved following the completion of the customs union (end of the 1960s) into an area of EU exclusive competence. From the end of the transitional period, the EEC engaged in trade relations with third countries based on trade liberalization and uniform principles.12 Hence, with the completion of the customs union in 1968, important decisions of the Court since the 1970s confirmed and consolidated the exclusivity of the trade competence and policy.13 The Court’s interpretation of the CCP remained important but trade policy further evolved through institutional practice and Treaty reforms in its scope and institutional set-up. Significantly, the latest Treaty revision – the Treaty of Lisbon – not only broadened the scope of the CCP,14 it also brought fundamental changes to the parliamentarization and politicization of EU trade policy through institutional change and the constitutional mandate of policy coherence.15 The CCP moved from the bureaucratic safe haven governed by European Commission trade specialists and national ministries in the 133 Committee (now Trade Policy Committee) to being opened up to the political scrutiny rights of the European Parliament (EP) through the ordinary legislative and consent

12 See the original Art. 110 EEC Treaty and now Art. 206 TFEU and Art. 113 EEC, now Art. 207 TFEU. 13 Opinion 1/75 Local Costs 1975; Case 41/76 Donckerwolcke ECLI:EU:1976:182. 14 For an overview, see Marc Bungenberg and Christoph Herrmann (eds), Common Commercial Policy After Lisbon (Springer 2013). 15 Peter Hilpold, ‘The “Politicization” of the EU’s Common Commercial Policy – Approaching the “Post-Lockean” Era’ in Marise Cremona, Peter Hilpold, Nikolaos Lavranos, Stefan Staiger Schneider and Andreas Ziegler (eds), Reflections on the Constitutionalisation of International Economic Law Liber Amicorum for Ernst-Ulrich Petersmann (Brill 2013); Christoph Vedder, ‘Linkage of the Common Commercial Policy to the General Objectives for the Union’s External Action’ in Bungenberg and Herrmann (eds) (n 14) 142.

The nexus between the CCP and the CFSP 233 procedure.16 And the EP emphasized its agenda to consider human rights principles and other aims under Article 21 TEU in EU (trade) agreements.17 The Treaty of Lisbon also established a horizontal set of principles and objectives guiding the Union’s external policymaking, including the CCP.18 In particular, Articles 205 and 207(1) TFEU explicitly submit the CCP to the general external policy principles and objectives of the Union’s external actions.19 The link created between the CCP and the general provisions of Article 21 TFEU has been considered as leading to a ‘politicization’ of trade policy.20 It is, therefore, mandatory that the CCP should now aim not only at the gradual liberalization of trade, but also at non-economic policy objectives such as the protection of human rights and fundamental freedoms, the promotion of sustainable and environmental development and the strengthening of international security.21 However, in the pre-Lisbon era, the CCP was also political. Even since the early days of the CCP, trade had been a tool to implement political aims through sanctions or embargoes and by international agreements such as association agreements and FTAs linking trade(-related) aspects with non-trade aims. The Court had already recognized in several cases that the Community could pursue policy objectives other than solely commercial ones through CCP instruments, including development cooperation,22 security policy,23 and social and environmental policy objectives.24 Also the streamlining exercise between trade and security had already been sanctioned by pre-Lisbon case law.25 However, Articles 21 TEU and 205 and 207(1) TFEU have now constitutionalized the incremental process of the politicization of the CCP. Since the Treaty of Lisbon, the pursuance of these horizontal principles and objectives in all areas of EU external action is of a binding nature and consequently

16

Before Lisbon and since 1973 the EP was informed on the conclusion of trade agreements. 17 European Parliament Resolution of 17 February 2014 on the implementation of the Treaty of Lisbon with respect to the European Parliament (2013/2130(INI)), paras 43, 44. 18 See Chapter 1 in this volume, and Markus Krajewski, ‘The Reform of the Common Commercial Policy’ in Andrea Biondi, Piet Eeckhout and Stefanie Ripley (eds), EU Law after Lisbon (OUP 2012) 297. 19 Opinion 2/15 Singapore FTA ECLI:EU:C:2017:376, paras 143–145. 20 Vedder (n 15) 142. 21 On the other hand, Arts 3(5) and 21(2) also make explicit references to trade objectives as general objectives of EU trade policy. 22 Opinion 1/78 International Agreement on Natural Rubber 1979 ECR 2871. 23 For example, in the cases Werner and Leifer, the Court held that the foreign policy objectives with regard to dual goods do not necessarily bring this matter outside the framework of the CCP (Case C-70/94 Werner ECLI:EU:C:1995:328, para 11; Case C-83/94 Leifer ECLI:EU:C:1995:329, para 10). 24 See Case 45/86 Commission v Council EU:C:1987:163; Case C-62/88 Greece v Council EU:C:1990:153; and Case C-281/01 Commission v Council EU:C:2002:761. 25 For example, the cases concerning dual-use goods: Werner and Leifer, the Court held that the foreign policy objectives with regard to dual goods do not necessarily bring this matter outside the framework of the CCP (Case C-70/94 Werner (n 23), para 11; Case C-83/94 Leifer (n 23), para 10).

234 Research handbook on the EU’s common foreign and security policy poses challenges resulting from the ambiguity of values and contrasting objectives. The former can be exemplified by the promotion of fair trade as part of a fundamental principle in Article 3(5) TEU next to free trade.26 Article 3(5) TEU is shorthand for the extensive external relations values and principles listed in Article 21(1) TEU. However the reference to free and fair trade in Article 3(5) TEU is neither repeated in Article 21 TEU nor referred to in Article 207 TFEU. It addresses the question of whether it reappears in another value found in Article 21 TEU and 207 TFEU or is to be understood as being shorthand for all non-economic values which the CCP has to take into account.27 If fair trade is considered a separate and stand-alone term next to other non-economic values, it is, however, riddled with ambiguity. And the decisive impulse for fair trade does not derive from a top-down approach through the legislator or policymaker but from a bottom-up initiative.28 Its non-legal and ambiguous character notwithstanding, the introduction of fair trade as a constitutional value is in line with the increasing emphasis on fair trade since the 1990s in policy documents.29 In addition, fair trade featured in the Cotonou Agreement in 2000,30 was low profile in other agreements31 until it made a comeback with specific trade and sustainable development chapters in all recent FTAs,32 and became a reason for withdrawal of GSP+ preferences.33 And finally, the slow death of the WTO Doha Development Round since 2003 also contributed to the EU’s refocus on a fairer approach to trade and the distribution of wealth between developed and developing countries.34 With regard to contrasting objectives, there may be situations where the trade-related aims of the CCP clash with the principles and objectives enshrined in Article 21 TEU, especially considering that the trade and non-trade-related goals of the CCP are of 26

Unlike fair trade, free trade is referred to in Arts 21(2)(e) TEU and 206 TFEU. Markus Krajewski, ‘The Reform of the Common Commercial Policy’ in A Biondi, P Eeckhout and S Ripley (eds), EU Law after Lisbon (OUP 2012). 28 So have fair trade, a labelling organization coordinating national labelling initiatives, and the World Free Trade Organization influenced private businesses to achieve greater equity in international trade and support development that is socially, economically and environmentally sustainable; see also Deborah Martens and Jan Orbie, ‘The European Union and Fair Trade: Hands-off?’ in Sangeeta Khorana and Maria Garcia (eds), Handbook of European Union and International Trade (Edward Elgar Publishing, forthcoming 2018). 29 See for example the Communication from the Commission to the Council on ‘fair trade’ COM (1999) 619 final, Brussels, 29 November 1999. 30 Art. 23(i) Cotonou Agreement, OJ 2012 L 303/1 and Art. 324(1)(c) FTA with Andean States, OJ 2012 L 354/3. 31 Art. 19 of the 2000 FTA with Mexico: the purpose of customs cooperation shall be to ensure fair trade, however, there was no mention in other FTAs, association agreements or the Economic Partnership with ECOWAS and EUMOA from 2014. 32 For instance, in the FTAs with Singapore and Vietnam (referring to economic development, social development and environmental protection). For earlier examples, see the FTA with the Andean States: Art. 324(2)(c): ‘promoting fair and equitable trade, facilitating access to the benefits of this Agreement for all production sectors, the weakest in particular’. 33 Art. 19(1)(d) Regulation No 978/2012 mentions ‘serious and systematic unfair trading practices including those affecting the supply of raw materials, which have an adverse effect on the Union industry and which have not been addressed by the beneficiary country’, OJ 2012 L 303/1. 34 Art. 23(i) Cotonou Agreement. 27

The nexus between the CCP and the CFSP 235 equal value.35 For example, can the EU restrict trade (thus breaching the economic objectives of the CCP) in order to realize its political objectives (e.g. security or the promotion of human rights)? It appears that such restrictive trade measures are still allowed if they indeed pursue one, or several, of the horizontal principles and objectives. The Court has recognized that the EU institutions must be allowed a broad discretion in areas which involve political, economic and social choices on its part [and] that the legality of a measure adopted in those fields can be affected only if the measure is manifestly inappropriate having regard to the objective which the competent institution is seeking to pursue.36

In particular, the Court held that the EU legislature enjoys wide discretion in the field of external economic relations, including when negotiating and concluding FTAs.37 Indeed, the political goals of Article 21 TEU are formulated in sufficiently general terms to give the EU institutions political discretion to assess and deal with a potential conflict between the economic objectives of the CCP and broader foreign policy considerations. However, the policy implications of the obligation to meet the horizontal political objectives enshrined in EU primary law remain unclear. For example, how – and to what extent – do the EU institutions have to take into account the human rights situation in a third country before engaging in FTA negotiations? In the Polisario case, the General Court partially annulled the EU Council decision concluding the EU-Morocco liberalization agreement for agricultural products because the Council had failed to examine a priori, carefully and impartially, the relevant facts to ensure that this agreement did not indirectly infringe the fundamental rights of the population of the Western Sahara.38 This judgment was set aside by the Court of Justice on the ground of no legal standing. However, AG Wathelet also concluded that the Council had an ‘obligation’ to check a priori that the agreement did not contribute to an infringement of human rights.39 In any case, it appears that a decision to maintain or develop a trade agreement with a third country that openly violates human rights and fundamental principles should in principle require from the EU institutions an a priori detailed explanation and strong motivation as to how and why the pursuance of the EU’s economic goals should prevail over its fundamental principles and values. 35 The Treaty of Lisbon has turned gradual trade liberalization into a binding objective of the EU’s trade policy by replacing the soft obligation in Art. 131 TEC (i.e. ‘aims to contribute’) with a binding commitment (i.e. the Union ‘shall’ contribute to trade liberalization) in Art. 206 TFEU. 36 Case C-440/14 National Iranian Oil Company v Council EU:C:2016:128, para 77. See also, to that effect, Case C-344/04 IATA and ELFAA EU:C:2006:10, para 80; Case C-266/05 Sison v Council EU:C:2007:75, para 33; Case C-127/07 Arcelor Atlantique and Lorraine and Others EU:C:2008:728, para 57. 37 See Case C-122/95 Germany v Council EU:C:1998:94, paras 77 and 79; and Case T-572/93 Odigitria v Council and Commission EU:T:1995:131, para 38. 38 Case T-512/12 Front Polisario v Council ECLI:EU:T:2015:953. 39 Opinion of AG Wathelet, Case C-104/16 Front Polisario v Council ECLI:EU:C:2016:677, para 262.

236 Research handbook on the EU’s common foreign and security policy

3. THE POLICY DIMENSION OF THE CCP–CFSP NEXUS As discussed above, Article 21 TEU mandates the EU to standardize and streamline other external action aims with EU trade policy. For this reason, the EU is increasingly emphasizing the interaction between trade and the CFSP in different policy documents. For example, the 2003 European Security Strategy merely recognized that ‘trade policies can be a powerful tool for promoting reform’ and that ‘targeted trade measures’ are an important instrument to realize the EU’s foreign policy and security objectives.40 However, in the 2016 Global Strategy for the European Union’s Foreign and Security Policy, trade plays a more important role since it is framed as both an objective and an instrument of the EU’s external action.41 One of the key objectives of the Global Strategy is, next to the promotion of peace and security, democracy and a rules-based global order, the advancement of prosperity inside and outside the EU. The Strategy sees an international economic system and open markets as a crucial objective to safeguard a prosperous Union. But trade is also considered as an instrument, for example as part of an integrated approach to conflicts and crises (e.g. restrictive measures and smart sanctions and fighting a criminal war economy by modernizing the EU’s policy on export control for dual-use goods).42 Moreover, ambitious bilateral trade agreements that promote international regulatory standards as well as labour, environmental, health and safety norms with strategic partners (Transatlantic Trade and Investment Partnership (TTIP), EU–Canada Comprehensive Economic and Trade Agreement (CETA), the Deep and Comprehensive Free Trade Areas (s) with the European Neighbourhood Policy (ENP) countries and FTAs with Asian countries) are considered as crucial instruments to achieve a strong multilateral rules-based order and economic system. A more integrated policy approach has also been visible since the European Commission decided in 2002 to conduct a Trade Sustainability Impact Assessment (Trade SIA) before the start of trade negotiations. This finds its roots in the 1999 sustainable development clause and has evolved into an overall economic, social, human rights and environmental analysis.43 This SIA accompanies the decision by the College of Commissioners to request a negotiating authorization from the Council of the EU, together with the draft negotiating directives to be issued by the Council of the EU. As part of the EU’s better regulation agenda, the Commission applies impact assessment and ex post evaluation for its EU trade policy instrument. The Council and the Commission have also committed themselves to include human rights considerations in their impact assessments of EU FTAs.44 In this context, the European 40

A Secure Europe in a Better World, European Security Strategy, 12 December 2003. ‘Shared Vision, Common Action: A Stronger Europe. A Global Strategy for the European Union’s Foreign and Security Policy’ (European External Action Service 2016). 42 These elements fall under the heading ‘Political Economy of Peace’. 43 See on this . 44 See Council conclusions of 25 June 2012 on Human Rights and Democracy, the EU Strategic Framework on Human Rights and Democracy and an EU Action Plan on Human Rights and Democracy, 11855/12, Annex III, I.1, p. 11. See also, to that effect, Joint Communication from the Commission to the European Parliament and the Council of 28 April 2015 JOIN (2015) 16 final. 41

The nexus between the CCP and the CFSP 237 Ombudsman found in February 2016 that the Commission’s refusal to carry out a prior human rights impact assessment of the EU–Vietnam FTA constituted a ‘maladministration’.45 Although the Ombudsman agreed with the Commission that there appears to be no express and specific legally binding requirement to carry out such an impact assessment, she took the view that it would be in conformity with the spirit of the legal provisions mentioned in the Treaties.46 The Commission aims to improve the ex post monitoring of EU trade agreements through implementation reports. The FTA with the Andean states, for instance, includes a chapter on trade and sustainable development, including multilateral labour standards (Art. 267), and provides for institutional structures of monitoring with domestic mechanisms. The implementation reports are, however, based on the mandate under Regulation 19/2013 implementing the bilateral safeguard clause and stabilization mechanism for bananas in the Agreement.47 Through this link the Commission acquires the mandate to assess the implementation of the Agreement and the application of the safeguard measures and to review ‘challenges in Colombia and Peru as regard human, social, labour and environmental rights’.48 As such, the Commission widens its mandate to live up to the parliamentarization of this policy and the pressures from the EP that human rights clauses need to be linked to effective implementation and compliance with Article 21 TEU values in trade agreements.49

4. TRADE-RESTRICTIVE MEASURES TO ACHIEVE CFSP AIMS Through the WTO multilateral system, the EU is limited in its capacity to impose trade restrictions or sanctions to achieve CFSP aims. In this section, we will highlight the exceptional use of such trade-restrictive measures because an agreement can be suspended for fundamental breaches of human rights or comparable reasons or if the EU has a mandate to tackle illegal trade through unilateral trade bans. This illegality is determined on the basis of international law protecting higher-ranking values such as environmental or conflict prevention.

45 On this issue, see also Marise Cremona, ‘A Quiet Revolution: The Common Commercial Policy Six Years after the Treaty of Lisbon’ (SIEPS Paper No. 2 2017). 46 European Ombudsman, Decision in case 1409/2014/MHZ on the European Commission’s failure to carry out a prior human rights impact assessment of the EU-Vietnam free trade agreement. 47 Second Report 2016 . 48 Indent 11 Regulation 19/2013, OJ 2013 L 17/2. 49 See European Parliament resolution of 13 June 2012 on the EU trade agreement with Colombia and Peru. 2012/2628/RSP; EP Resolution of 26 November 2015 on the accession of Ecuador to the Trade agreement concluded between the EU and its MS and Columbia and Peru; see generally, Andrea Ott, ‘The European Parliament’s Role in EU Treaty-making’ (2016) Maastricht Journal of European and Comparative Law 1022.

238 Research handbook on the EU’s common foreign and security policy 4.1 Suspension of Trade and Cooperation Agreements Through Human Rights Clauses Several examples can be highlighted in which the EU restricts trade for CFSP purposes. The most well known are trade sanctions or restrictive measures under the wider framework of international law, which are not discussed in this chapter.50 In addition to trade sanctions based on UN law, other obligations under international law and international treaty law with a peace and security focus need to be implemented by EU trade measures. For example, the suspension or termination of an international trade (and cooperation) agreement on the basis of an essential element clause falls into this category. Since its introduction by the Lomé IV Association Agreement in 1989, the scope of the essential element clause has broadened, covering provisions on, inter alia, human rights, the prohibition of weapons of mass destruction and migration and good governance.51 The original motivation to include such essential standard clauses from the 1990s was to enable the suspension of such agreements, thereby avoiding the difficulties of suspending trade agreements in line with international law and remedying human rights violations.52 However, the number of (trade) agreements which have formally been suspended is confined to two examples.53 If a reaction from the Union is required to address a specific situation in a third country (e.g. human rights violations), the EU might prefer to act through sanctions or diplomatic procedures instead of opting for the ‘nuclear’ option, i.e. suspending the legal framework (and overall trade relations) with that country.54 Human rights clauses, therefore, amount to a deterrent

50

See Chapter 10 in this volume. These clauses have evolved since their inception but vary through different generations of agreements. Especially the Baltic clause in the Europe Agreements with these countries raised the issue of consistency because all eight Europe Agreements demonstrate variations of these clauses: Peter Van Elsuwege, From Soviet Republics to EU Member States: A Legal and Political Assessment (Brill 2008) 106–107. 52 Uganda and Equatorial Guinea, which could not be remedied by the existing contractual relations under Lomé I and the Trade Agreement with the former Yugoslavia, was suspended in 1991 but challenged unsuccessfully in the court judgment in Racke. See further Barbara Brandtner and Allan Rosas, ‘Human Rights and the External Relations of the European Community: An Analysis of Doctrine and Practice’ (1998) European Journal of International Law 479. 53 In 1991 the bilateral trade agreement with the former Yugoslavia was suspended. However, this suspension could not be based on a human rights clause, instead the agreement was suspended based on UN Security Council Resolution 713 (Council Regulation (EEC) No 3300/91 of 11 November 1991 suspending the trade concessions provided for by the Cooperation Agreement between the European Economic Community and the Socialist Federal Republic of Yugoslavia, OJ 1991 L 315/1). In the case of the Cooperation Agreement with the Syrian Arab Republic a partial suspension took place, see Council Decisions 2011/523/EU, OJ 2012 L 16/1 and 2012/123/CFSP, OJ 2012 L 54/18. 54 For example, as a response to Russia’s ‘unprovoked violation of Ukrainian sovereignty and territorial integrity’ and annexation of Crimea, the EU did not suspend its PCA with Russia, but adopted a series of CFSP sanctions and suspended negotiations on the ‘New Agreement’ (the envisaged successor of the PCA). On this issue, see Guillaume Van der Loo, The EU-Ukraine 51

The nexus between the CCP and the CFSP 239 against abuses. However, more sophisticated human rights clauses, such as the Articles 96 and 97 Cotonou Agreement mechanism with the African, Caribbean and Pacific (ACP) countries, allow the EU to take ‘appropriate measures’.55 Such measures may include temporary cuts in budgetary aid,56 as in the case of Madagascar in 2010, for example, or the suspension of the 10th European Development Fund (EDF) Country Strategy Paper, as well as all cooperation projects implemented through the government.57 Recently, the Union has aimed to apply a standardized toolbox of human rights and essential clauses but this can create obstacles in treaty negotiations with third countries, and discrepancies in the treatment of countries remain.58 Negotiations with Thailand on a Partnership Cooperation Agreement (PCA) were halted due to Thai concerns about compliance with the essential clauses, in particular regarding the role of the Thai monarch. In 1996 Australia refused to accept the inclusion of a human rights clause in a trade and cooperation agreement.59 Despite standardization, a discrepancy between developed and developing countries in the application of these human rights clauses and their conditionality is evident. Negotiations with Canada hit rock bottom in 2013 on CETA and the parallel negotiated Strategic Partnership Agreement (SPA) once the EU put its standard clauses on the negotiating table. Canada found it difficult to accept standard clauses and conditionality drafted in line with the solution found with the PCA with Singapore, and the human rights clauses are therefore toned down.60 In contrast to the PCA Vietnam (Art. 57 PCA), no conditional link between the FTA and the SPA in case of a breach of the SPA is created but only the suspension of provisions of the Agreement from both sides (Art. 28 SPA Canada) is included. Despite fulfilling its commitment to Article 21 TEU law objectives and the EP’s insistence on consistent Association Agreement and Deep and Comprehensive Free Trade Area: A New Legal Instrument for EU Integration Without Membership? (Brill/Nijhoff 2016). 55 This procedure already existed as Art. 366a in Lomé IV. The current Cotonou Agreement expires in 2020 and the Economic Partnership Agreements rebasing the trade relations between EU and ACP countries, such as the one with the Economic Community of West African States (ECOWAS) states create a link to the human rights principles of the Cotonou agreement. Art. 96 includes consultation and appropriate measures in regard to human rights, democratic principles and the rule of law and Art. 97 covers this in regard to corruption. Art. 96 has been applied 15 times since 2000 in cases involving Fiji (2000, 2007), Zimbabwe (2002), the Central African Republic (2003), Guinea-Bissau (2004, 2011), Togo (2004) and Madagascar (2010). 56 Other appropriate measures were taken against Fiji 2007, Madagascar 2010, Zimbabwe 2002–2014, Liberia 2003, and Guinea-Bissau July 2011–July 2014. 57 Council Decision 2010/371 of 6 June 2010 concerning the conclusion of consultations with the Republic of Madagascar under Article 96 of the ACP-EU Partnership Agreement, OJ 2010 L 169/13. See also Council Decision of 18 July 2011 concerning the conclusion of consultations with the Republic of Guinea-Bissau under Article 96 of the Cotonou Agreement, OJ 2011 L 203/2. 58 This template for negotiations is divided into the three components: (1) Human rights clauses and prohibition of weapons of mass destruction clause, (2) Migration and good governance clauses and (3) Technical clauses on the non-fulfilment of obligations. 59 See Brandtner and Rosas (n 52) 474. 60 Art. 1 SPA: ‘shared values’, Art. 5 ‘share a common commitment’, Art. 6 ‘shared priority’, defining case of special urgency and serious and substantial violation such as a coup d’état, Art. 28 and ‘in the unlikely and unexpected event’.

240 Research handbook on the EU’s common foreign and security policy standardization, the EU needs to adapt to the interests of stronger and strategic trading partners. These variations need to be pragmatically adapted without becoming double standards applied to developed or developing nations. In addition to the examples discussed above, two other groups of trade-restrictive measures can be identified. The first group encompasses those trade-restrictive measures that the EU is required to adopt to implement international commitments. The second group consists of measures that aim to realize some of the EU’s normative foreign policy goals, codified in, for example, Articles 3(5) and 21 TEU. Both of them are discussed here. 4.2 Tackling Illegal Trade by Restrictive Measures or Trade Bans 4.2.1 Importing international law obligations into EU trade policy In all its actions, the EU needs to recognize its international law commitments and the UN Charter according to Articles 3(5) and 21(2)(b) and (c) TEU, including EU trade policy and law. Article 215 TFEU stipulates that cross-policy measures are taken to implement EU sanctions against the backdrop of CFSP measures. These sanctions were based on the CCP legal base in the past,61 such as in 1982 when the EEC applied trade sanctions against Argentina in reaction to its invasion of the Falkland Islands.62 Nowadays, trade embargoes against states and individuals are one of the instruments in the toolbox of EU sanctions.63 Other international law obligations arise from UN Security Council Resolution 1540, the Chemical Weapons Convention and the Biological Weapons Convention to restrict and ban trade. The 2015 ‘Trade for All’ Communication announced ‘an ambitious modernisation of the EU’s policy of export controls of dual-use goods, including the prevention of the misuse of digital surveillance and intrusion systems that results in human rights violations’.64 Unlike weapons of mass destruction and landmines,65 trade in conventional weapons had not been subject to any legally binding global instrument until the Arms Trade Treaty (ATT) of 2012. Based on a UN Resolution, the ATT obliges its signatory parties to monitor arms exports and they may not authorize the transfer under the circumstances that UN arms embargoes need to be upheld if the transfer would violate international treaty obligations or if a state has the knowledge at the time of authorization that these conventional arms are used in case of genocide, crimes against humanity and grave breaches of the Geneva Convention of 1949 (establishing standards for humanitarian treatment in times of war). The EU contributed 61 Though see the legal difficulties encountered when the EEC had for the first time to implement UN sanctions against Rhodesia in 1965; see in more detail Panos Koutrakos, Trade, Foreign Policy and Defence in EU Constitutional Law: EU Constitutional Law (Hart Publishing 2001) 59. 62 Regulation 596/82, OJ 1982 L 72/15 and for further examples see Sara Poli and Maria Tzanou, ‘The Kadi Rulings: A Survey of the Literature’ (2009) Yearbook of European Law 533. 63 For instance, in the case of Eritrea, embargo on arms and related material, ban on provision of certain services, Council Decision 2010/127/CFSP, OJ 2010 L 51. 64 European Commission, ‘Trade for all. Towards a more responsible trade and investment policy’ COM (2014) 497, 14 November 2015. 65 See Chapter 9 in this volume.

The nexus between the CCP and the CFSP 241 actively to the negotiation process and argued for a best practice example in its export regime on dual-use goods and its common regime on defence-related products. The EU could not sign this ATT, however, because it is not a state and is only involved through an observer status in the UN. The EU was also not granted a RIO (regional integration organization) exception to this UN Treaty by the UN members.66 On 27 May 2013 the EU adopted Council Decision 2013/269/CFSP authorizing Member States to sign, in the interest of the Union and according to Article 2(1) TFEU, the ATT in the subject field which falls within the EU’s exclusive trade competence.67 These examples demonstrate that despite trade being an exclusive EU competence, once trade policy enters the realm of CFSP, Member States might get involved for institutional and substantive reasons and curtail effective policymaking. 4.2.2 Trade bans, sustainable development and human rights protection The above analysis makes the point that the EU is under an obligation in all its external relations policies to ‘uphold and promote its values and interests’, which include ‘sustainable development of the Earth, free and fair trade, eradication of poverty and the protection of human rights’ (Art. 3(5) TEU). In different legal and policy documents, the EU emphasizes the interdependence of sustainable development, fair and ethical trade and human rights68 and the importance of responsible management of supply chains through responsible business conduct and corporate social responsibility practices.69 The responsible management of supply chains addresses new forms of global trading resulting from the fragmentation of production processes.70 In addition, illegal trade in environmentally sensitive goods such as wildlife, timber, fish and hazardous waste is covered by national and international regulatory regimes. To counter illegal trade in these environmentally sensitive goods, the EU adopted regulatory measures on conflict minerals (the 2017 Conflict Minerals Regulation),71 illegal logging (EU Timber Regulation, in force since 2013)72 and the IUU Regulation to

66 This exemption enables the EU’s participation in the conclusion of a UN multilateral agreement and such a RIO exception is provided in the UN Convention on the Law of the Sea. 67 OJ 2013 L 155/9. 68 European Commission, ‘Trade for all’ (n 64) 24; European Commission ‘Non-paper on Trade and Sustainable Development (TSD) chapters in EU Free Trade Agreements (FTAs)’, 11 July 2017. 69 So, for example, the EU-Mercosur FTA draft, 10 March 2017 . 70 See on this a 2013 UNCTAD study on global supply chains: Trade and economic policies in developing countries, Trade and economic policies in developing countries, Policy issues in International Trade and Commodities Study Series No. 55. 71 Regulation (EU) 2017/821 of the European Parliament and of the Council of 17 May 2017 laying down supply chain due diligence obligations for Union importers of tin, tantalum and tungsten, their ores, and gold originating from conflict-affected and high-risk areas, OJ 2017 L 130/1. 72 Regulation (EU) No 995/2010 of the European Parliament and of the Council of 20 October 2010 laying down the obligations of operators who place timber and timber products on the market, OJ 2010 L 395/23.

242 Research handbook on the EU’s common foreign and security policy combat illegal fishing.73 Another example is the EU Seals Regulation banning the import and sale of seal products with the exception of Inuit hunted seal products.74 These EU regulatory acts aim to export the EU’s values and principles, impact third countries and their traders, and restrict or even exclude trade. The EU pursues these aims in the wider international framework and its own values but without always operating under a clear mandate of binding international rules but an uncoordinated web of autonomous state rules and standard-setting tools forming mainly soft law rules.75 There might be a certain consensus that these measures define higher values and are triggered to prevent Member States from acting unilaterally. However, these measures are at risk of breaching WTO law and have to tread a fine line between illegal and legal trade.76 The EU Timber Regulation and Conflict Minerals Regulation team up with similar regulatory measures taken by other industrial nations, the US, Australia, Japan or Canada.77 Both EU acts implement a ban on illegal products, aim to achieve greater transparency and execute mandatory due diligence systems for traders in timber and conflict minerals in line with international soft law rules. The Timber Regulation bans illegal timber in the EU and requires companies that place wood or wood products on the EU market for the first time to assess the risk that those products may have come from an illegal source and companies have to reduce any identified risks.78 The Conflict Minerals Regulation will be applicable from 2021 onwards and will implement a compulsory system of due diligence.79 Due diligence in 73

Council Regulation (EC) No. 1005/2008 of 29 September 2008 establishing a Community system to prevent, deter and eliminate illegal, unreported and unregulated fishing, OJ 2008 L 286/1. 74 Regulation (EC) No 1007/2009 of the European Parliament and of the Council of 16 September 2009 on trade in seal products, OJ 2009 L 286/36. 75 So is the international (soft law) framework for the EU Timber Regulation relevant: 1972 Stockholm declaration, 1992 Rio Declaration and 2001 Bali Declaration. The Conflicts Minerals Regulation finds its inspiration in the OECD Guidelines for multilateral enterprises and UN guiding principles on business and human rights. On these standard-setting devices see Meng Du and Fei Deng, ‘International Standards as Global Public Goods in World Trading System’ (2016) Legal Issues of Economic Integration 113. 76 Dylan Geraets and Bregt Natens, ‘The WTO Consistency of the European Union Timber Regulation’ (Working Paper No. 120, Leuven Centre for Global Governance Studies 2013) ; Enrico Partiti and Steffen van der Velde, ‘Curbing Supply-Chain Human Rights Violations Through Trade and Due Diligence. Possible WTO Concerns Raised by the EU Conflict Minerals Regulation’ (2017) 51 Journal of World Trade 6. 77 For illegal logging: Australian Illegal Logging Prohibition Act and amendment of the US Lacey Act. For conflict minerals: the US Dodd-Frank Sec.1502, but suspended in February 2017 by US President Trump. 78 The literature critically assessed that it excludes traders buying and selling within the EU market and no printed material, such as books, is covered, see Laurens Ankersmit, Green Trade and Fair Trade in and with the EU (CUP 2017) 54, 55. 79 The European Parliament amended the Commission proposal from a voluntary to a mandatory due diligence system, see further: Steffen Van der Velde, ‘The End of Conflict Minerals on the EU Market?’ (Policy Brief No. 3, Asser Institute March 2017); Wybe Douma and Steffen Van der Velde, ‘Protection of Fundamental Rights in Third Countries through the EU External Trade Policy: The Cases of Conflict Minerals and Timber’ in C. Paulussen, T. Takács,

The nexus between the CCP and the CFSP 243 regard to conflict minerals implies that the companies concerned apply supply chain controls in all conflict-affected or high-risk areas, in order to identify the risk of funding harmful activities. Crucial questions arise about whether such unilateral acts violate the prohibition of extraterritoriality and WTO rules. The former prohibits the international community from moving its laws and jurisdiction beyond its borders, with minor exceptions, such as, for instance, the principle of universal jurisdiction in criminal law.80 The latter concerned the question of whether such regulatory measures form recognized international standards under the TBT (Technical Barriers to Trade) Agreement and can be justified under the Chapeau of Article XX (justification of breaches of GATT Agreement). Both issues have been addressed by the CJEU81 and in the WTO dispute settlement system.82 Whilst WTO compatibility is difficult to establish, no breach of the principle of territoriality can be detected because a sufficient link between the regulatory acts and the EU territory exists.83 The Appellate Body in the EC – Seals dispute argued that, in principle, the EU Seals Regulation could rely on the public morals exception which the EU was not allowed to apply in a discriminatory manner.84

5. TRADE FACILITATION AND PREFERENTIAL MARKET ACCESS TO ACHIEVE CFSP GOALS In addition to trade-restrictive measures, the EU aims to achieve its foreign policy objectives by offering, sometimes under certain conditions, additional preferential market access. As already noted, the EU’s FTA policy has always been ‘political’ although following the 2006 Global Europe Strategy, economic considerations seem to prevail when selecting potential FTA partners. Nevertheless, the EU’s DCFTAs concluded with Ukraine, Moldova and Georgia illustrate that (geo-)political considerations still play an important role in the EU’s trade policy.85 Moreover, the sustainable development chapters included in the EU’s new generation of FTAs also aim to V. Lazic´, and B. Van Rompuy (eds), Fundamental Rights in International and European Law (TMC Asser Press 2016). 80 Cedric Ryngaert and Marieke Koekkoek, ‘Extraterritorial Regulation of National Resources: A Functional Approach’ in J. Wouters, A. Marx, D. Geraets and B. Natens (eds), Global Governance through Trade, EU Policies and Approaches (Edward Elgar Publishing 2015) 245–271. 81 Case C-366/10 ATAA ECLI:EU:C:2011:864. 82 European Communities – Measures Prohibiting the Importation and Marketing of Seal Products, WT/DS400 & 401/R (adopted as modified, 18 June 2014). 83 See especially EC – Seals case and ATAA case, in detail: Ryngaert and Koekkoek (n 80); Meng Du, ‘Permitting Moral Imperialism? The Public Morals Exception to Free Trade at the Bar of the World Trade Organisation’ (2016) Journal of World Trade 686. See on the Conflict Minerals Regulation: Enrico Partiti and Steffen van der Velde, ‘Curbing Supply-chain Human Rights Violations through Trade and Due Diligence. Possible WTO Concerns Raised by the EU Conflict Minerals Regulation’ (Asser Institute Research Paper series, June 2017). 84 In addition, the Appellate Body departed from the Panel’s categorization of the EU’s seal regime as a technical regulation in the sense of Art. 1 TBT. 85 See Chapter 15 in this volume.

244 Research handbook on the EU’s common foreign and security policy contribute to several of the EU’s external policy objectives, related to international labour standards and environmental protection. There are also several other ad hoc trade measures that the EU adopts in light of broader foreign policy objectives. For example, trade facilitation is also implemented through (quasi-CFSP instruments such as border missions. The 2005 EUBAM mission to Moldova and Ukraine, which is strictly speaking not a CFSP instrument, has a strong trade-facilitation dimension, focusing on issues such as trade facilitation, customs fraud and DCFTA implementation.86 Or, as part of a broader support package for Jordan in the context of the Syrian refugee crisis, the EU aimed to facilitate access to the EU market by temporarily relaxing its rules of origin for the country. In order to stimulate the participation of Syrian refugees in Jordan’s formal labour market, this initiative allows producers in Jordan to use an alternative set of rules of origin (i.e. those applied under the Everything But Arms (EBA) schema) for exports to the EU, provided that production takes place in one of the 18 geographic zones in Jordan and includes the creation of jobs for Syrian refugees.87 However, the most obvious example of this practice is the EU’s GSP+ arrangement, which provides additional market access to developing countries complying with a set of international conventions on human and labour rights, environmental protection and good governance. However, the GSP+ can hardly be seen as a flexible foreign policy instrument as the EU can only give additional market access through the GSP+ arrangement if the beneficiary country meets certain economic and political criteria. Moreover, the WTO framework imposes legal restraints on the EU’s discretion to use the GSP+ for political purposes. This may explain why the EU increasingly relies instead on ATMs to offer market access as a foreign policy tool. Yet the adoption of such ATMs may also face procedural challenges within the EU and the WTO. The use of both instruments for broader CFSP purposes is discussed below. 5.1 The GSP+ Preferences: Triangle Between Trade, Development and CFSP Aims Since its adoption by the EC in 1971, the EU’s Generalised System of Preferences (GSP) has always been a ‘pure’ CCP instrument, based solely on Article 207 TFEU (ex. Art. 133 TEC).88 However, positive conditionality was only introduced in the GSP 86 The legal basis of this EUBAM mission is a Memorandum of Understanding signed by the Commission and the governments of Ukraine and Moldova. The EUBAM was established in the pre-Lisbon era as a first-pillar instrument, funded by European Commission instruments. 87 Decision 1/2016 of the EU-Jordan Association Committee amending the provisions of Protocol 3 to the Euro-Mediterranean Agreement establishing an Association ‘between the European Communities and their Member States, of the one part, and the Hashemite Kingdom of Jordan, of the other part, concerning the definition of the concept of originating products’ and the list of working or processing required to be carried out on non-originating materials in order for certain categories of products, manufactured in dedicated development zones and industrial areas, and connected with generating employment for Syrian refugees and Jordanians, to obtain originating status, OJ 2016 L 233/6. 88 Although the first GSP Regulations adopted in 1971 did not explicitly refer to ex Art. 113 EEC on the CCP, they were clearly adopted as CCP measures (Regulations 1308/1971 to

The nexus between the CCP and the CFSP 245 system in the 1994 GSP revision89 and implemented in 1998. Three specific ‘arrangements’ were established granting more favourable treatment to countries introducing effective policies and/or ratifying the relevant international conventions with regard to (i) labour rights (e.g. ILO Conventions), (ii) the environment (e.g. the International Tropical Timber Organization (ITTO)) and (iii) combating drug production and trafficking.90 However, the use of the GSP+ system to pursue political objectives is constrained by both the WTO rules and the political and economic criteria that beneficiary countries have to meet. The legal constraints imposed by the WTO framework became clear when India contested the EU’s special GSP preferences to Pakistan in 2001. In the post-9/11 context, which turned Pakistan into a frontline state in the war against terrorism and drug trafficking, the European Commission added Pakistan to the list of beneficiaries of the ‘drug arrangement’. India challenged this decision in the WTO by arguing that the EU’s special arrangements needed to be granted in a generalized and nondiscriminatory way in order not to treat developing countries differently. In December 2003, the Panel indeed concluded that the ‘GSP drug arrangement’ was inconsistent with Article I:1 GATT (the most-favoured nation (MFN) principle) and was not justified under the Enabling clause.91 On appeal, the Appellate Body also concluded in 2004 that GSP preferences must be based on clear criteria, that identical treatment should be available to all similarly situated GSP beneficiaries and that there must be a clear link between the preferential treatment provided and the development objective.92 Thus, pursuant to WTO rules, the GSP system cannot simply be used to reward countries for their contribution to the EU foreign policy goals, such as the fight against terrorism.93 In response to the WTO ruling, in 2006 the EU replaced the three special incentive-based arrangements by one integrated ‘special incentive arrangement for

1314/1971, OJ 1971 L 142). The subsequent GSP Regulations explicitly had the CCP Treaty provision as a legal basis (see for example Council Regulation (EC) No 3281/94 of 19 December 1994 applying a four-year scheme of generalised tariff preferences (1995–1998) in respect of certain industrial products originating in developing countries, OJ 1994 L 348/1. 89 Council Regulation (EC) No 3281/94, ibid. 90 Council Regulation (EC) No 1154/98 of 25 May 1998 applying the special incentive arrangements concerning labour rights and environmental protection provided for in Articles 7 and 8 of Regulations (EC) No 3281/94 and (EC) No 1256/96 applying multiannual schemes of generalised tariff preferences in respect of certain industrial and agricultural products originating in developing countries. The special arrangement to combat drug production and trafficking was introduced by Council Regulation (EC) No 2501/2001 of 10 December 2001 applying a scheme of generalised tariff preferences for the period from 1 January 2002 to 31 December 2004, OJ 2001 L 346. 91 WTO, European Communities – Conditions for the Granting of Tariff Preferences to Developing Countries, Report of the Panel, WT/DS246/R, 1 December 2003. 92 WTO, European Communities – Conditions for the Granting of Tariff Preferences to Developing Countries, AB-2004-1, Report of the Appellate Body, WT/DS246/AB/R, 7 April 2004. 93 Lorand Bartels, ‘The WTO Legality of the EU’s GSP+ Arrangement’ (2007) 10 Journal of International Economic Law 4.

246 Research handbook on the EU’s common foreign and security policy sustainable development and good governance’, now known as the GSP+ system.94 In light of the Treaty of Lisbon and the Global Europe Strategy, the EU’s GSP system was further revised in 2012.95 These different GSP reforms broadened the list of international conventions relevant to the GSP+, going beyond the core labour conventions and also covering sustainable development and human rights, developed strict economic criteria, and refocused priorities on ‘the neediest’. In order to be eligible for the GSP+ regime, applying countries must ratify and implement 27 conventions96 (without formulating reservations prohibited by these conventions) and must be ‘vulnerable’ in terms of their economies’ diversification and import volumes.97 These strict political and economic criteria limit the number of countries eligible for the GSP+ regime, and therefore also the political discretion of the Commission to randomly add a country to the GSP+ list for pure foreign policy considerations. The Commission still enjoys a broad discretion to withdraw existing GSP preferences for political reasons, however. Negative conditionality was introduced in the 1994 GSP revision and the 2005 Regulation also introduced the possibility of downgrading GSP+ privileges to the general scheme. According to current GSP Regulation, all three GSP arrangements (i.e. the general GSP, GSP+ and EBA) can be withdrawn by the Commission (through an implementing act) in the event of ‘serious and systematic violation of principles’ laid down in the 15 GSP+ conventions related to core human and labour rights.98 In addition, the latest GSP Regulation adds several additional grounds for temporary withdrawal of preferences, including export of goods made by prison labour and failure to comply with international conventions on anti-terrorism and money laundering.99 Although it is easier for the Commission to withdraw GSP+ preferences for political reasons than granting them, the general GSP scheme was only withdrawn in two cases (Myanmar (1997) and Belarus (2006)). Under the previous GSP Regulation Sri Lanka’s GSP+ preferences were also downgraded to the general regime (2010).100 It has been observed that in both cases of complete withdrawal, the GSP withdrawal had followed CFSP sanctions.101 However, 94 Council Regulation (EC) No 980/2005 of 27 June 2005 applying a scheme of generalised tariff preferences, OJ 2005 L 169/1. 95 Regulation (EU) No 978/2012 of 25 October 2012 applying a scheme of generalised tariff preferences and repealing Council Regulation (EC) No 732/2008, OJ 2012 L 303. 96 Fifteen related to ‘core human and labour rights UN/ILO conventions’ and 12 conventions related to the environment and to governance principles. 97 The vulnerability criterion implies that a country can only benefit from GSP+ if it is not competitive on the EU market (import-share ratio) and it does not have a diversified export base (diversification ratio). For a critical review of the 2012 GSP reform, see G. Siles-Brügge, ‘EU Trade and Development Policy Beyond the ACP: Subordinating Developmental to Commercial Imperatives in the Reform of the GSP’ (2014) 20(1) Contemporary Politics 49. 98 Art. 19(1)(a) Regulation (EU) No 978/2012 (n 95). 99 ibid, Art. 19. 100 Implementing Regulation (EU) No 143/2010 of 15 February 2010 temporarily withdrawing the special incentive arrangement for sustainable development and good governance provided for under Regulation (EC) No 732/2008 with respect to the Democratic Socialist Republic of Sri Lanka, OJ 2010 L 45/1. 101 Clara Portela and Jan Orbie, ‘Sanctions under the EU Generalised System of Preferences and Foreign Policy: Coherence by Accident?’ (2014) 20(1) Contemporary Politics 63.

The nexus between the CCP and the CFSP 247 there is no link between CFSP sanctions and GSP withdrawal as both instruments address different kind of violations (whereas CFSP sanctions support the broad CFSP objectives, the grounds for GSP withdrawal are narrower and only relate to breaches of core human and labour rights). Therefore, the EU does not develop mechanisms or policies to ensure coherence between CFSP sanctions and GSP withdrawal.102 Although the withdrawal of GSP+ preferences is rare, the GSP+ monitoring procedures give the Commission a strong and broad mandate to monitor domestic developments related to the GSP+ Conventions. In order to address previous inconsistencies with regard to the application of the GSP+ system, the 2012 GSP+ revision includes stricter monitoring procedures to evaluate whether there is a ‘serious failure’ to effectively implement these conventions.103 GSP+ applicants must also accept without reservation the reporting requirements imposed by those conventions and agree to participate in and cooperate with the EU GSP+ monitoring led by the European Commission.104 The GSP Regulation even states that the biennial GSP+ monitoring report may cover, in addition to compliance with the obligations under the listed conventions, ‘any information the Commission considers appropriate’.105 This monitoring procedure enables the EU to use the threat of GSP+ withdrawal to address a broad range of issues in the GSP+ beneficiary countries. For example, in May 2017 the Commission added Sri Lanka again to the GSP+ scheme as it considered that the country met all the relevant criteria.106 However, the Commission noted that it will closely monitor some issues of concern identified in the application process, including Sri Lanka’s efforts to ensure that its counter-terrorism is fully in line with international human rights conventions and the definitive ending of the use of torture by security forces.107 Or, more recently, as a reaction to the extrajudicial killings taking place in the Philippines linked to President Duterte’s so-called ‘war on drugs’ and the legislative proposals to reinstate the death penalty for drug-related crimes and to lower the age of criminal responsibility from 15 to nine years old, the EP and the Civil Society Dialogue on GSP+ urged the Commission to consider the removal of GSP+ preferences for the Philippines in the absence of substantive improvements.108 102 Clara Portela and Kolja Raube, ‘The EU Polity and Foreign Policy Coherence’ (2001) 8(1) Journal of Contemporary European Research 1. 103 For example, Colombia and Guatemala benefited from the GSP+ regime despite being seriously criticized by the ILO (Jan Orbie and Lisa Tortell, ‘The New GSP+ Beneficiaries: Ticking the Box or Truly Consistent with ILO Findings?’ (2009) 14 European Foreign Affairs Review 663). 104 Art. 9(1)(f) Regulation (EU) No 978/ 2012 (n 95). 105 ibid, Art. 14(2). 106 Commission Delegated Regulation (EU) 2017/836 of 11 January 2017 amending Annex III to Regulation (EU) No 978/2012 of the European Parliament and of the Council applying a scheme of generalised tariff preferences OJ 2017 L 125. For the Commission’s positive report on the assessment of the application for GSP by Sri Lanka, see SWD (2016) 474. 107 European Commission, ‘EU grants Sri Lanka improved access to its market as incentive for reform’ (Press release, 17 May 2017). 108 European Parliament resolution of 16 March 2017 on the Philippines – the case of Senator Leila M De Lima (2017/2597(RSP)); Civil Society Dialogue meeting on the GSP+, 12 July 2017, minutes of the meeting at . After a GSP+ monitoring mission to the Philippines (26 January–2 February 2017),

248 Research handbook on the EU’s common foreign and security policy Finally, it should be noted that there are also cases where geostrategic foreign policy considerations precluded the EU in several cases from withdrawing GSP preferences. For example, it has been argued that the EU did not withdraw trade GSP preferences for Pakistan in 1998, after complaints about forced child labour in the carpet industry in Pakistan, because the EU, in particular several Member States, did not want to jeopardize the ongoing negotiations for a cooperation agreement with Pakistan.109 5.2 Autonomous Trade Measures In addition to the GSP+ scheme, the EU is increasingly relying on ATMs to offer third countries unilateral market access. Over the past decade, the EU has unilaterally offered additional preferential market access to several countries through ATMs, mainly as support measures in emergency or disaster situations. All these ATMs were adopted as CCP instruments, solely based on Article 207 TFEU. However, these ATMs also have a strong ‘political dimension’ as they serve broader foreign policy goals (e.g. support measures in emergency or disaster situations), or were influenced by the EU’s strategic interests. Moreover, in most cases the EU has made the application of these ATMs conditional on the fulfilment of several human rights considerations. For example, trade was – and still is – a key component of the Stabilisation and Association Process. Therefore, in 2000 the Community adopted unilateral trade preferences through ATMs towards the Western Balkan countries which aimed to ‘support economic integration with the EU and hence foster political stability and economic progress in the entire region’. These were linked to the ‘respect for fundamental principles of democracy and human rights and the Western Balkan countries’ and ‘readiness to engage in effective economic reforms and in regional cooperation’.110 After several revisions, in 2015 the EU extended these ATMs to 2020 and added a specific temporary suspension mechanism for ‘systematic violations of human rights, including core labour rights, of fundamental principles of democracy and of the rule of law by its beneficiaries’.111 specific areas ‘in which progress is strongly encouraged’ were indicated, in particular regarding criminal justice and the prevention and treatment of drug use. With regard to the Philippines, the Commission’s most recent GSP report noted progress in some areas such as labour rights and social–economic rights, but stressed that ‘serious concerns remain on the way the campaign against illegal drugs is conducted, in particular concerning the right to life, legal safeguards, effective investigation and prosecution, on the cooperation with UN mechanisms as well as on the possible re-introduction of the death penalty and the possible lowering of the age of criminal responsibility’ (European Commission, ‘Report on the Generalised Scheme of Preferences covering the period 2016–2017’, 19 January 2018, COM (2018) 36 final). 109 Thomas Greven, Social Standards in Bilateral and Regional Trade and Investment Agreements (Occasional Paper 16, Friedrich Ebert Stiftung 2005). 110 Council Regulation (EC) No 2007/2000 of 18 September 2000 introducing exceptional trade measures for countries and territories participating in or linked to the European Union’s Stabilisation and Association process, OJ 2000 L 295. For the Commission’s proposal, see COM (2000) 351. 111 Regulation (EU) 2015/2423 of the European Parliament and of the Council of 16 December 2015 amending Council Regulation (EC) No 1215/2009 introducing exceptional trade measures for countries and territories participating in or linked to the European Union’s

The nexus between the CCP and the CFSP 249 It is also clear that (geo-)political considerations play a role in the adoption of ATMs. These became clearly visible in the EU’s trade policy vis-à-vis the Eastern Partnership (EaP) countries. It is no secret that the Kremlin opposes the EU’s association agreements (AAs) with these countries as it fears that this would pull them out of Russia’s traditional sphere of influence. In this context, Russia adopted several trade-related restrictions against Moldova and Ukraine before and after the signature of these agreements, in order to force these countries away from closer EU association. However, in both cases the EU reacted by adopting ATMs for Moldova and Ukraine, aiming to compensate for their loss on the Russian market and as a diplomatic move to support these countries’ ‘European choice’. For example, in the framework of the ENP, in 2008 the Community granted Moldova unilateral preferential treatment through ATMs.112 Initially, these ATMs mainly had a development goal as the Commission considered that providing Moldova, ‘the poorest country on the European continent’, with improved access to the EU market would support its economic development.113 However, the EU broadened the scope of these ATMs as a reaction to Russia’s import ban on several Moldovan products. For example, in September 2013, not coincidentally two months before the scheduled initialling of the EU-Moldova AA, Russia banned wine imports from Moldova – officially for food safety concerns. In order to show ‘political will to respond to unjustified and arbitrary pressures exerted by Russia on its Eastern partners’114 and to support the government of Moldova in its efforts to initial and sign the Association Agreement, in December 2013 the EU amended the ATMs towards Moldova to fully liberalize, in line with the expected DCFTA, wine imports from this country.115 To support the new pro-EU government in Ukraine after the Maidan Revolution, and as a support measure to the economic crisis in the country in the aftermath of these turbulent events, the EU also adopted ATMs for Ukraine.

stabilisation and association process and suspending its application with regard to Bosnia and Herzegovina, OJ 2015 L 341/18. 112 Council Regulation (EC) No 55/2008 of 21 January 2008 introducing autonomous trade preferences for the Republic of Moldova and amending Regulation (EC) No 980/2005 and Commission Decision 2005/924/EC, OJ 2008 L 20/1. 113 Proposal for a Council Regulation introducing autonomous trade preferences for Moldova and amending Regulation (EC) No 980/2005 and Commission Decision 2005/924/EC COM (2007) 705. 114 European Parliament, ‘MEPs back freeing wine trade with Moldova to offset Russian trade sanctions’ (Press release, 10 December 2013). 115 Regulation (EU) No 1384/2013 of the European Parliament and the Council of 17 December 2013 amending Council Regulation (EC) No 55/2008 introducing autonomous trade preferences for the Republic of Moldova, OJ 2013 L 354/85. Again as a reaction to Russian import bans, the EU broadened these ATMs in December 2015 (see Regulation (EU) No 1383/2014). It has to be noted that as early as 2010 Russia imposed restrictions on imports of wine from Moldova. Also then, the EU responded by increasing the existing tariff rate quotas for Moldovan wine (on this issue, see Frank Hoffmeister, ‘The European Union’s Commercial Policy a Year After Lisbon – Sea Change or Business as Usual?’ in Panos Koutrakos (ed.), The European Union’s Relations One Year after Lisbon (CLEER Working Paper No. 3, 2011) 89.

250 Research handbook on the EU’s common foreign and security policy Following a proposal from the European Commission,116 the EP and Council adopted, after a ‘fast-track’ approval process,117 Regulation 374/2014 ‘on the reduction or elimination of customs duties on goods originating in Ukraine’.118 This Regulation unilaterally reduced or eliminated customs duties on goods originating in Ukraine in full compliance with the EU’s DCFTA, which at that time was not yet provisionally applied.119 This ATM also has to be seen in the context of Russia’s trade-related retaliation measures against the EU-Ukraine AA and DCFTA. For example, the rapporteur in the EP on this Regulation stated that ‘as Putin closes Russian markets for Ukrainian exports, we are opening them’.120 Thus, if the EU wants to unilaterally offer a third country additional preferential market access for political reasons, it is easier to do so through ATMs than through adding the country to the GSP+ scheme. Compared to the GSP+ scheme, there are no political or economic criteria to be met for (potential) ATM beneficiaries. Nevertheless, the adoption of such ATMs can be complicated by political opposition within the EU or at the level of the WTO. With regard to the former, the adoption of an ATM requires a legislative act based on Article 207 in accordance with the ordinary legislative procedure. This implies, since the Treaty of Lisbon, that both the Council and the EP are fully involved in the adoption of ATMs. This is a more demanding procedure compared to adding a country to the GSP+ scheme, which requires only the adoption of a delegated act by the Commission.121 Because most of these ATMs concern sensitive products, including agricultural products, Member States, and in particular MEPs from 116 European Commission, ‘Proposal for a Regulation of the European Parliament and the Council on the reduction or elimination of customs duties on goods originating in Ukraine’ COM (2014) 166 final. 117 The Commission encouraged the EP and Council to ‘fast-track’ the approval process so that the tariff reduction could be in place in June 2014: European Commission, ‘European Commission proposes temporary tariff cuts for Ukrainian exports to the EU’ (Press release, 11 March 2014). With the aim of implementing the unilateral DCFTA tariff reduction as soon as possible, no amendments were made by the EP to the Commission’s proposal. 118 Regulation (EU) No 374/2014 of the European Parliament and of the Council of 16 April 2014 on the reduction or elimination of customs duties on goods originating in Ukraine, OJ 2014 L 118/1. 119 This Regulation incorporates in its Annex the Union’s DCFTA Schedule of Concessions (included in Annex I-A of the EU-Ukraine AA). 120 European Parliament, ‘MEPs cut customs duties on imports from Ukraine’ (Press release, 3 April 2014). But after the signature of the EU-Ukraine AA, Russia also threatened to completely suspend its preferential trade with Ukraine, laid down in the 2011 Commonwealth of Independent States FTA. During a trilateral ministerial meeting between the EU, Ukraine and Russia in September 2014, it was decided, after strong pressure from Russia, to delay the provisional application of the DCFTA until 31 December 2015. Meanwhile, the EU would continue to apply its ATMs towards Ukraine. For a more detailed overview of these developments, see Guillaume Van der Loo, The EU-Ukraine Association Agreement and Deep and Comprehensive Free Trade Area: A New Legal Instrument for EU Integration Without Membership? (Brill/Nijhoff 2016). 121 Art. 10(4) Regulation No 732/2008. However, the Commission needs to notify the EP and the Council when adopting such a delegated act. The EP and the Council may object to the delegated act within a period of two months of the notification, after which the Commission is required to repeal the act (see Arts 36 and 37 of the Regulation).

The nexus between the CCP and the CFSP 251 southern Member States, insist on specific safeguard measures. For example, after the terrorist attacks in Sousse on 26 June 2015, the EU decided as a support measure to boost the Tunisian economy to import 70,000 tonnes of Tunisian olive oil (Tunisia’s main agricultural export). The EU offered a temporary, unilateral duty free tariff rate quota of 35,000 tonnes annually (70,000 in total) for Tunisia’s exports of olive oil, in the form of ATMs.122 This measure was approved by the EP on 25 February 2016, but to meet concerns of the EU olive producers, MEPs inserted additional safeguards.123 Similarly, in view of the difficult economic situation and ongoing reform process in Ukraine, the European Commission proposed to improve access for Ukraine to the EU market for some industrial and agricultural products. The Commission proposed temporary ATMs for Ukraine in 2016, broadening the scope of the DCFTA that provisionally entered into force on 1 January 2016.124 These ATMs would be granted in the form of zero-tariff quotas for certain agricultural products in addition to the preferential tariff rate quotas set out in the DCFTA, and the partial or full removal of import duties on several industrial products. Also in this case, the EP amended the Commission proposal by reducing the volume of the additional quota for certain agricultural products.125 The EU’s ATMs can also be challenged at the WTO level. Given that the unilateral introduction of trade preferences would be in breach of the MFN principle (Article I(1) GATT) and the principle of non-discriminatory administration of quantitative restrictions (Article XIII GATT), the EU needs to request a waiver pursuant to Article IX:3 of the WTO when applying such ATMs.126 After the devastating floods in Pakistan in July 2010, the September 2010 European Council called not only for the provision of traditional humanitarian aid, but also for the adoption of additional measures including granting ‘exclusively to Pakistan increased market access to the EU through the immediate and time limited reduction of key imports from Pakistan’.127 After the 122 Regulation (EU) 2016/580 of the European Parliament and of the Council of 13 April 2016 on the introduction of emergency autonomous trade measures for the Republic of Tunisia, OJ 2016 L 102/1. 123 These include a mid-term assessment, which gives the Commission the competence to adopt an implementing act in order to introduce corrective measures if it turns out that they harm EU olive oil producers and a ‘tracking clause’ obligation which has to ensure that all olive oil under the quota is obtained entirely in, and transported directly from, Tunisia. On this issue, see Guillaume Van der Loo, ‘Mapping out the Scope and Contents of the DCFTA with Tunisia and Morocco’ (2016) Euromesco Paper 28. 124 European Commission, Proposal for a Regulation on the introduction of temporary autonomous trade measures for Ukraine supplementing the trade concessions available under the Association Agreement COM (2016) 0631 final. 125 European Parliament legislative resolution of 4 July 2017 on the proposal for a regulation of the European Parliament and of the Council on the introduction of temporary autonomous trade measures for Ukraine supplementing the trade concessions available under the Association Agreement (P8_TA(2017)0285). The ATMs were approved by the Council on 17 July 2017: Council, ‘Ukraine: Council adopts temporary trade preferences’ (Press release, 17 July 2017). 126 This procedure allows the WTO Ministerial Conference ‘in exceptional circumstances … to waive an obligation imposed on a member by this Agreement or any of the Multilateral Trade Agreements, provided that any such decision shall be taken by three-fourths of the Members’. 127 European Council Conclusions, 16 September 2010, Declaration on Pakistan.

252 Research handbook on the EU’s common foreign and security policy Council and EP reduced the scope of the Commission’s proposal (for example fearing the impact of the proposal on the textiles sector), ATMs were adopted in 2012.128 However, when the EU requested a waiver, pursuant to Article IX WTO, Pakistan’s commercial and geopolitical rivals, including India, blocked this process.129 The waiver was eventually approved in October 2012, after the EU revised its request for a waiver and reduced the trade concessions (for example by increasing the use of tariff rate quotas instead of full liberalization).130 The EU’s ATMs for Pakistan demonstrated that obtaining the required WTO waiver support is not always guaranteed.131 This may explain why the EU did not ask for a waiver for several recent ATMs. For example, the EU did not request a WTO waiver for its ATMs for Ukraine and Tunisia. Concerning the 2014 ATMs for Ukraine, the Commission argued that this was ‘an advance application of a GATT Article XXIV compatible FTA for a very limited time’ and that therefore no waiver was required.132 The EU did not ask for a waiver for its ATM for Tunisia either, most likely because it considers that this too is covered by Article XXIV GATT through the EU-Tunisia FTA, which is included in their Association Agreement.133

6. CONCLUSION Notwithstanding the underlying politicization of EU trade since the beginning of the EU integration process, the recent introduction of a constitutional obligation of mainstreaming and standardization of a joint set of EU external relations principles and values strengthens the nexus between the aims of the CCP and the CFSP. The Lisbon Treaty promotes foreign policy objectives as trade tools or trade as a tool to achieve 128

Regulation (EU) No 1029/2012 of the European Parliament and of the Council of 25 October 2012 introducing emergency autonomous trade preferences for Pakistan, OJ 2012 L 316. On this issue, see Sieglinde Gstöhl, ‘No Strings Attached? The EU’s Emergency Trade Preferences for Pakistan’ in Inge Govaere and Sara Poli (eds), EU Management of Global Emergencies: Legal Framework for Combating Threats and Crises (Brill Nijhoff 2014). 129 Peter Van Elsuwege and Jan Orbie, ‘The EU’s Humanitarian Aid Policy After Lisbon’ in Govaere and Poli (eds), ibid. 130 WTO, ‘European Union Preferences for Pakistan’, Waiver Decision of 14 February 2012, WT/L/851. It has to be noted that the EU did not request a WTO waiver for its ATMs for Ukraine and Tunisia. 131 Although Art. IX:3 WTO requires the approval of three-fourths of the Members, decisions on such matters are generally taken by consensus. On these issues, see Isabel Feichtner, The Law and Politics of WTO Waivers: Stability and Flexibility in Public International Law (CUP 2012). 132 Response of the Commission’s representative to questions during the meeting of the European Parliament ITNA Committee, 19 March 2014. On this point see Van der Loo (n 120) 236. 133 However, the EU asked for an extension of its WTO waiver for the ATM for the Western Balkan countries (Council Decision (EU) 2016/1039 of 16 June 2016 establishing the position to be taken on behalf of the European Union within the General Council of the World Trade Organization on the European Union request for an extension of the WTO waiver relating to the autonomous preferential regime for the Western Balkans).

The nexus between the CCP and the CFSP 253 foreign policy aims. This constitutional obligation is also reinforced by the extended role of the EP, which champions the politicization of EU trade policy throughout the process from the negotiation of legislation or international agreements to the implementation of these instruments. However, the EU’s commitment to international trade and WTO rules limits the EU’s discretion to consider other values when this could either lead to a discriminatory treatment of and between third countries or could not be justified under the limited scope under WTO law to protect other societal values.134 The dichotomy between trade-restrictive and trade-facilitating measures demonstrated that the international law framework imposes more restrictions on the former than the latter. Trade-restrictive measures can only be implemented if based on a sound international law mandate and on values and international commitments such as the protection of peace and security (e.g. dual-use goods or implementation of UN sanctions). Societal values such as sustainable development and protection of human rights are more difficult to weave into trade-restrictive measures. The more the EU works with a broader brush and includes CFSP aims in non-economic objectives to be achieved in trade relations, the more difficult it will be to act in conformity with international (trade) rules and to have a direct impact with its action on third countries. Trade facilitation might fare better but it was demonstrated that also the unilateral granting of trade preferences through the GSP+ scheme for foreign policy reasons is constrained by WTO law and the EU’s own eligibility criteria. This may explain why the EU relies increasingly on ATMs if it wants to offer a third country additional market access for political reasons (e.g. emergency response measures or geopolitical considerations). Nevertheless, such ATMs have to be approved as well by the other WTO members, and need to overcome internal opposition.

134 See further Peter Van den Bossche and Werner Zdouc, The Law and Policy of the World Trade Organization (CUP 2017), p. 592.

12. EU development cooperation and the CFSP: mutual encroachment? Morten Broberg

1. OBJECTIVES AND STRUCTURE OF THIS CHAPTER In February 2017 the Munich Security Conference brought together several hundreds of senior decision-makers from around the world. When speaking at this conference, the German Chancellor, Dr Angela Merkel, stated that ‘we need to do more to join up military capabilities. … We need leadership within the European Union to allow us to design a joined up approach that can include development policy and good governance, not only military capabilities.’1 In this way, Chancellor Merkel made it clear that foreign relations, security and development cooperation go hand in hand. This may not be particularly surprising since development cooperation, at the level of nation states, is generally considered part of foreign policy. However, the Treaty basis of the European Union draws an important distinction between, on the one hand, the Common Foreign and Security Policy (CFSP), and, on the other hand, development cooperation policy. This chapter sets out to present important aspects of the Union’s development cooperation policy and to examine the nexus between this policy and the CFSP.2 My focus will be on the provision of development aid,3 approaching the CFSP from the outside, rather than from within. I first present the contemporary EU development cooperation policy (Section 2).4 There is an intimate relationship between development cooperation and foreign and security policy but in an EU context these policies are placed within very different legal and institutional schemes. It is important, therefore, to clarify their outer limits, and, in particular, whether, and if so how, they may ‘encroach’ upon each other. I therefore examine the boundaries between these two policies (Section 3). Next, I turn to consider the institutional issues inherent in the CFSP–development cooperation nexus (Section 4). Finally, I sum up the main findings (Section 5). 1

German Chancellor, Angela Merkel, in her speech on 18 February 2017 at the 53rd Munich Security Conference, available at accessed 25 June 2017. 2 For an in-depth examination of the security–development nexus in European law, see Hans Merket, The EU and the Security-Development Nexus: Bridging the Legal Divide (Brill Nijhoff 2016). 3 The important question regarding the CFSP and EU external trade is dealt with in Chapter 11 in this volume, while the question of the CFSP and humanitarian aid is dealt with in Chapter 13. 4 For a presentation of the EU development cooperation policy pre Lisbon, see Morten Broberg, ‘What Is the Direction for the EU’s Development Cooperation after Lisbon?’ (2011) 16(4) European Foreign Affairs Review 539.

254

EU development cooperation and the CFSP 255

2. EU DEVELOPMENT COOPERATION POLICY AFTER THE LISBON TREATY 2.1 Overview With the entry into force of the Treaty of Lisbon on 1 December 2009 the previous three-pillar system was replaced with a unitary one, based on two treaties: the Treaty on European Union (TEU) and the Treaty on the Functioning of the European Union (TFEU). A central objective of the new Treaty system was to improve the Union’s position on the international stage. The changes brought about by the Lisbon Treaty affected the European Union’s development cooperation policy both directly and indirectly. Below, I will first consider the consequences of the Lisbon Treaty’s reorganization of the objectives of the European Union’s development cooperation policy (Section 2.2) whereupon I turn to consider the likely consequences of the Lisbon Treaty’s provisions that require the Union to ensure policy coherence in its external relations and to promote its own values in the wider world (Section 2.3). 2.2 Reorganization of the Objectives of EU Development Cooperation Policy The Lisbon Treaty reshuffled and elaborated on the objectives of the European Union in several respects, including with regard to development cooperation policy. When considering the EU development cooperation policy in a CFSP context, this reshuffling entails two important changes: first, with the second paragraph of Article 208(1), the Union’s development cooperation objectives have been streamlined so that there now is a much stronger focus on the primary objective, that is the fight against poverty in developing countries;5 second, the remaining, and hitherto broadly defined objectives of the Union’s development cooperation policy (such as the promotion of democracy and the rule of law) have been reorganized and now form part of the general ‘framework of the principles and objectives of the Union’s external action’ set out in Article 208(1) TFEU. As such, they must be respected in all the Union’s external activities pursuant to Article 205 TFEU and Article 21(3) TEU. In other words, these objectives apply to both the development cooperation policy and to the CFSP.6 Article 21(1) and (2) TEU list these principles and objectives thus: 1.

The Union’s action on the international scene shall be guided by the principles which have inspired its own creation, development and enlargement, and which it seeks to advance in the wider world: democracy, the rule of law, the universality and indivisibility of human rights and fundamental freedoms, respect for human dignity, the principles of equality and solidarity, and respect for the principles of the United Nations Charter and international law.

5 See similarly Opinion of Advocate General Bot in Case C-658/11 Parliament v Council (Mauritius) EU:C:2014:41, para 126. 6 Article 21(3) TEU explicitly refers to both Title V TEU (where we find the CFSP) and to Part Five TFEU (where we find the EU’s development cooperation policy). See also Opinion of Advocate General Bot, ibid.

256 Research handbook on the EU’s common foreign and security policy The Union shall seek to develop relations and build partnerships with third countries, and international, regional or global organisations which share the principles referred to in the first subparagraph. It shall promote multilateral solutions to common problems, in particular in the framework of the United Nations. 2.

The Union shall define and pursue common policies and actions, and shall work for a high degree of cooperation in all fields of international relations, in order to: (a) safeguard its values, fundamental interests, security, independence and integrity; (b) consolidate and support democracy, the rule of law, human rights and the principles of international law; (c) preserve peace, prevent conflicts and strengthen international security, in accordance with the purposes and principles of the United Nations Charter, with the principles of the Helsinki Final Act and with the aims of the Charter of Paris, including those relating to external borders; (d) foster the sustainable economic, social and environmental development of developing countries, with the primary aim of eradicating poverty; (e) encourage the integration of all countries into the world economy, including through the progressive abolition of restrictions on international trade; … .7

First, elevating some objectives, which pre Lisbon could be found in the chapter on the Union’s development cooperation policy (e.g. to support democracy and the rule of law), to horizontal objectives, which must guide all the European Union’s actions in the field of external relations, suggests that these objectives have been strengthened from a constitutional point of view.8 Indeed, references to these objectives and principles have made their way into hard-law CFSP measures9 as well as into a number of TFEU soft-law measures in various fields of the Union’s external relations.10 Moreover, today the Court of Justice uses these objectives and principles as legitimate teleological guidelines in its interpretation of legal acts in all external relations policy areas, including e.g. the Common Commercial Policy.11 We are thus witnessing a generalization of objectives and principles which previously belonged to specific policy areas, meaning that they are becoming more firmly embedded in the Union legal order. Second, the streamlining of development cooperation objectives so that the reduction/eradication of poverty has been made a ‘primary objective’ of European Union development cooperation policy suggests that, within the field of development cooperation, this objective should be accorded particular weight in cases of conflict 7

Emphasis added. With particular regard to the EU’s CFSP it may be noted that prior to the Lisbon Treaty, Article 11 EU laid down very similar objectives for the CFSP as those that apply post Lisbon to the European Union’s policies and actions in all fields of international relations. 9 See, for example, Council Decision (CFSP) 2016/947 of 14 June 2016 amending Joint Action 2008/124/CFSP on the European Union Rule of Law Mission in Kosovo [2016] OJ L157/26. 10 See, for example, ‘New European Consensus on Development: Our world, our dignity, our future’, Joint statement by the Council and the representatives of the governments of the Member States meeting within the Council, the European Parliament and the Commission [2017] OJ C210/1, available at accessed 24 July 2017. 11 See, for example, Opinion 2/15 (Free Trade Agreement between the European Union and the Republic of Singapore (EUSFTA)) EU:C:2017:376, paras 139–147. 8

EU development cooperation and the CFSP 257 with other objectives.12 The requirement that poverty reduction/eradication should be a ‘primary objective’ therefore suggests that, among the traditional economic and social objectives which have always formed a part of the Union’s development cooperation policy, there should be a stronger and more specific focus on poverty reduction/ eradication in the Union’s policy formulation.13 If we accept this line of reasoning, this would seem to mean that the European Union’s specific development initiatives must always (also) pursue a goal of poverty reduction. For example, a development cooperation initiative aimed at improving environmental protection in a developing country should also clearly have a poverty-reducing objective to fully comply with Article 208 TFEU. Thus, individual development cooperation measures, which do not have as their primary objective to reduce/eradicate poverty or which do not contribute to this objective in the overall context of the Union’s development cooperation policy, no longer fall within the Union’s development cooperation competences. Apart from this preference for poverty reduction/eradication, the streamlining did not entail new hard legal obligations as regards the Union’s exercise of competence vis-à-vis the situation pre Lisbon. As regards the Union’s competence to enter into international agreements, Article 209 TFEU provides that ‘[t]he Union may conclude with third countries and competent international organisations any agreement helping to achieve the objectives referred to in Article 21 of the Treaty on European Union and in Article 208 of this Treaty’. Thus, Article 209 TFEU explicitly foresees international agreements that have as their objectives some of the horizontal objectives mentioned in Article 21 TEU. In this light, the reorganization of the Union’s development cooperation objectives did not entail noticeable new restrictions on the Union’s treaty-making competence in this area.14 2.3 Policy Coherence for Development and Value Promotion as Restraints on the Union’s Exercise of Its Development Cooperation Competence? When exercising its development cooperation policy competences, the Union must comply with EU law, including the general principles. This is not new. But in the post-Lisbon field of development cooperation, two principles appear to require particular attention. The first is specifically concerned with development cooperation, whereas the second applies to all fields of external action. First, the Maastricht Treaty introduced a requirement for so-called policy coherence for development (PCD).15 The Lisbon Treaty has maintained this requirement in Article 208(1) TFEU, which provides that ‘[t]he Union shall take account of the objectives of 12 See in this respect also Henrike Klavert, ‘EU External Action Post-Lisbon: What Place Is There for Development Policy?’ (2011) 4 The Bulletin of Fridays of the Commission 18, 18–23

Note that eradication of poverty appears both in Article 21(2)(d) TEU and in Article 208 TFEU. In contrast to the other objectives, eradication of poverty has thus been duplicated. 13 See in support of this para 4 of the ‘New European Consensus on Development’ (n 10). 14 The Court’s judgment in the ‘Philippines Partnership and Cooperation Framework Agreement’ case, Case C-377/12 EU:C:2014:1903, seems to confirm this. The judgment is analysed in Section 3.4 below. 15 cf. Article 178 EC.

258 Research handbook on the EU’s common foreign and security policy development cooperation in the policies that it implements which are likely to affect developing countries’. Article 208(1) TFEU obliges the Union to pursue coherence between its objectives in the field of development policy and its other policies. However, no substantive legal effects flow from this duty of coherence between development cooperation policy and other policies being laid down in primary law. Post Lisbon, the Treaties also include other provisions which more generally require the Union to ensure policy coherence in its external relations, notably in Articles 3(5) and 21(3) TEU as well as Article 7 TFEU. These provisions are broadly concerned with the coherence of the Union’s policies, in particular in the field of external relations.16 The other issue worth mentioning is the enhanced requirement of value promotion. What impact does the reshuffling of the EU development cooperation policy objectives have on the Union’s competence to promote its own values, such as democracy and human rights, as part of that policy? Pre-Lisbon the European Union also actively pursued the promotion of democracy and respect for human rights as part of its development cooperation policy and did so with the approval of the Court of Justice.17 The Lisbon Treaty, however, has taken this promotion one step further. According to Article 3(5) TEU, in its relations with the wider world, the European Union shall uphold and promote its values and interests and contribute to the protection of its citizens. It shall contribute to peace, security, the sustainable development of the Earth, solidarity and mutual respect among peoples, free and fair trade, eradication of poverty and the protection of human rights, in particular the rights of the child, as well as to the strict observance and the development of international law, including respect for the principles of the United Nations Charter.18

This obligation to actively advance European values in the wider world is also reflected in Article 21(1) TEU, which provides that ‘[t]he Union’s action on the international scene shall be guided by the principles which have inspired its own creation, development and enlargement, and which it seeks to advance in the wider world …’.19 While it might be going too far to suggest that Article 3(5) TEU entails that the inclusion of a human rights clause in international agreements (or, at least, in international development cooperation agreements) constitutes a condition of legality of

16

On the general obligation of the European Union to attain coherence in its external relations post Lisbon, see Christophe Hillion, Cohérence et action extérieure de l’Union européenne (EUI LAW, Working Paper No. 14, 2012), available at accessed 29 July 2017. 17 Morten Broberg, ‘From Colonial Power to Human Rights Promoter: On the Legal Regulation of the European Union’s Relations with the Developing Countries’ (2013) 26 Cambridge Review of International Affairs 675, 682–683. 18 Emphasis added. 19 Emphasis added. See also Article 21(2)(a)–(c) TEU as well as Article 205 TFEU. The latter provides that ‘[t]he Union’s action on the international scene, pursuant to this Part, shall be guided by the principles, pursue the objectives and be conducted in accordance with the general provisions laid down in [Articles 21 and 22] of the Treaty on European Union’.

EU development cooperation and the CFSP 259 the latter,20 the Union may have an obligation to ensure a minimum degree of human rights compliance, as well as effective remedies in the event of breach, in its cooperation with developing countries. So far, we only find a limited number of explicit references to the obligation in Article 3(5) TEU to promote European values; first of all in European Parliament resolutions.21 This arguably indicates that, until now, the provision has only been attributed rather limited legal weight. By contrast, it is easy to find examples where the European Union commits to generally promoting its values in the wider world. Thus, for instance, in the ‘New European Consensus on Development’ it is clearly set out that the European principles and values shall guide both the European Union and the Member States in their development action.22 Similarly, the European Council explicitly laid down in the so-called ‘Stockholm Programme’ that ‘[t]he Union should continue to promote European and international standards and the ratification of

20 An argument to this effect was put forward by Advocate General La Pergola in para 29 of his Opinion in Case C-268/94 Portugal v Council (India Agreement) EU:C:1996:207. In the subsequent ruling in the same case the Court of Justice refrained from taking a position in this respect, however (cf. Case C-268/94 Portugal v Council (India Agreement) EU:C:1996:461). 21 See, for example, European Parliament Resolution of 12 March 2015 on the EU’s priorities for the UN Human Rights Council in 2015 (2015/2572(RSP) [2016] OJ C316/178; European Parliament Resolution, on the Discrimination against Girls in Pakistan, in Particular the Case of Malala Yousafzai [2014] OJ CE72 43; European Parliament Resolution, on Human Rights, Sexual Orientation and Gender Identity at the United Nations [2013] OJ CE56/100; European Parliament Resolution, on International Trade Policy in the Context of Climate Change Imperatives [2012] OJ CE99/94; European Parliament Resolution, on Human Rights and Social and Environmental Standards in International Trade Agreements [2012] OJ CE99/31; and European Parliament Resolution, on EU Policies in Favour of Human Rights Defenders [2011] OJ CE236/69; as well as the European Parliament’s resolution in which it laid down its priorities for the annual UN Human Rights Councils. In the latter resolution the European Parliament explicitly referred to Article 3(5) TEU while observing that ‘respect for, and the promotion and safeguarding of, the universality of human rights is part of the European Union’s ethical and legal acquis and one of the cornerstones of European unity and integrity’, see, for instance, European Parliament Resolution, on the 13th Session of the United Nations Human Rights Council [2010] OJ C348/6. See likewise European Parliament Resolution, on Democracy Building in the EU’s External Relations [2010] OJ CE265/3, in particular para F; European Parliament Resolution, on Restrictive Measures Directed Against Certain Persons and Entities Associated with Usama bin Laden, the Al-Qaida Network and the Taliban, in Respect of Zimbabwe and in View of the Situation in Somalia [2010] OJ C286/5, para A; and European Parliament Resolution, on Religious Freedom in Pakistan [2011] OJ CE161/147, para A. See also Proposal for a Regulation of the European Parliament and of the Council setting up a Union regime for the control of exports, transfer, brokering, technical assistance and transit of dual-use items (recast), COM (2016) 616 final as well as Joined Cases C-584/10 P, C-593/10 P, and C-595/10 P Commission and Others v Kadi EU:C:2013:518, para 103; Case T-512/12 Front Polisario v Council EU:T:2015:953, paras 159–167 (the latter ruling was subsequently overturned by the Court of Justice on the basis that the plaintiff did not have locus standi, cf. Case C-104/16 P Council v Polisario EU:C:2016:973), and Opinion 2/15 (Free Trade Agreement between the European Union and the Republic of Singapore) EU:C:2017:376, para 146. 22 ‘New European Consensus on Development’ (n 10). See particularly section 1.3.

260 Research handbook on the EU’s common foreign and security policy international conventions, in particular those developed under the auspices of the UN and the Council of Europe’.23 Article 3(5) TEU does not impose specific obligations as to how the European Union must further its own values,24 and nothing indicates that Article 3(5) TEU should require a more intensive effort in this regard than the Union’s efforts pre Lisbon.25 For example, Article 3(5) TEU does not entail that a human rights clause in a cooperation agreement with a developing country is a condition of internal legality under EU law. Rather, Article 3(5) TEU appears to be important mainly for the Union’s international identity, including for its external relations discourses.

3. DELIMITATION BETWEEN EU DEVELOPMENT COOPERATION POLICY AND THE CFSP 3.1 Overview As will be very clear from the above, the European Union’s Treaties draw an important distinction between, on the one hand, the CFSP, and, on the other hand, the Union’s development cooperation policy. However, as is reflected in this chapter’s introductory quote of Chancellor Merkel, there is a clear nexus between development and, in particular, security. It is therefore only natural that in many situations the European Union’s development cooperation policy and its CFSP overlap, as was clearly illustrated in the ECOWAS case.26 In this section I will consider the delimitation between the two policies. Or, rather, first I shall show that the Union’s development cooperation policy is sometimes used towards furthering European objectives regarding security and migration, which arguably fall under the CFSP (Section 3.2). Next, I suggest that the CFSP is also used towards objectives that, strictly speaking, should be part of the

23

The Stockholm Programme: An Open and Secure Europe Serving and Protecting Citizens [2010] OJ C115/1, at para 7.6. For another example, see Commission Staff Working Document accompanying the Communication from the Commission to the European Parliament, the Council and the European Economic and Social Committee – Tax and Development Cooperating with Developing Countries on Promoting Good Governance in Tax Matters, SEC (2010) 426 final. 24 This promotion must respect the other objectives and principles of the Union’s external actions, however. 25 On the consequences flowing from Article 3(5) TEU, see also Broberg (n 4) 548–554, and Broberg (n 17) 682–683. 26 Case C-91/05 Commission v Council (ECOWAS) EU:C:2008:288. The ECOWAS case is sometimes also referred to as the Small Arms and Light Weapons case, the Small Arms case or merely the SALW case. For more examples, see also Hans Merket, The EU and the SecurityDevelopment Nexus: Bridging the Legal Divide (Brill Nijhoff 2016), particularly chapter 3, and Karen Del Biondo, Stefan Oltsch and Jan Orbie, ‘Security and Development in EU External Relations: Converging, but in which Direction?’ in Sven Biscop and Richard Whitman (eds), The Routledge Handbook of European Security (Routledge 2013) 134–135.

EU development cooperation and the CFSP 261 development cooperation policy (Section 3.3). Finally, I turn to consider the choice between development cooperation and the CFSP as legal basis post Lisbon (Section 3.4). 3.2 EU Development Cooperation Policy Used to Achieve Objectives Arguably Falling Under the CFSP In 2014 the European Parliament and the Council adopted Regulation 230/2014 establishing an Instrument contributing to Stability and Peace.27 This instrument empowers the Union to undertake development cooperation measures, as well as financial, economic and technical cooperation measures in third countries to address issues that include ‘[p]reserving peace, preventing conflicts, strengthening international security and assisting populations, countries and regions confronting natural or manmade disasters’.28 In other words, the development cooperation instrument contributing to stability and peace is also used towards objectives that will normally be considered to fall within the CFSP. However, the European Union’s use of development cooperation to achieve objectives that, at least in principle, fall under the CFSP is not merely limited to a single financing instrument. Thus, in 2005 the Council and the representatives of the governments of the Member States, the European Parliament and the Commission jointly adopted the so-called European Consensus for Development29 (in 2017 a New European Consensus30 was adopted to replace the (old) European Consensus). This was the first joint statement setting out common principles, values and objectives of development cooperation policy for the European Union and its Member States. ‘The Consensus’, as it is widely known, was made up of two parts. In the first part the European Union and the Member States set out their development policies in accordance with the principle of complementarity. The second part set out the European Union’s role, the value that the (complementary) EU contribution would add to the one provided by the Member States,31 and how the objectives, principles, values, PCD and commitments defined in the first part of the (old) Consensus were to be made

27 [2014] OJ L77/1. This Instrument contributing to Stability and Peace (IcSP) took over from the 2006 Instrument for Stability (IfS) that was laid down in Regulation 1717/2006 establishing an Instrument for Stability and which expired on 31 December 2013. On the latter, see Stefan Gänzle, Coping with the ‘Security-Development-Nexus’: The European Community’s Instrument for Stability – Rationale and Potential (German Development Institute 2009). 28 Recital 2 of the Regulation. 29 Joint statement by the Council and the representatives of the governments of the Member States meeting within the Council, the European Parliament and the Commission on European Union Development Policy: ‘The European Consensus’ [2006] OJ C46/1. 30 ‘New European Consensus on Development’ (n 10). 31 Pre-Lisbon the European Union’s development policy was complementary to that of the Member States. Post Lisbon the European Union’s and the Member States’ development policies are mutually complementary.

262 Research handbook on the EU’s common foreign and security policy operational at Union level. The primary focus clearly was upon more traditional development objectives.32 As observed in Section 2, in June 2017 the same parties jointly adopted a New European Consensus on Development, thereby replacing the old Consensus.33 The New Consensus makes it clear that eradicating poverty continues to be a primary objective of the European Union’s development cooperation policy. However, in addition to more traditional development objectives, the European Union will now also expressly use its development cooperation activities to further its own interests in areas that normally fall under the CFSP such as security (as well as migration which is not normally viewed as part of development cooperation policy).34 There has been some criticism levelled against the choice, articulated in the New Consensus, to rely upon development cooperation policy in order to achieve objectives that would normally be seen as falling under the CFSP. As one observer put it: ‘The Consensus is unrealistic: development cooperation cannot pretend to lead to sustainable development if it is, at the same time, used to serve Europe’s economic, commercial, migration and security interests. The overall direction may even ultimately hamper the achievement of the positive elements in the Consensus.’35 Thus, irrespective of the fact that the Treaty basis of the European 32 This is not to say that during the old Consensus the European Union did not pursue objectives that went beyond traditional development objectives. For example, in 2011 the European Commission adopted a communication which, inter alia, provided that ‘the EU should assist developing countries in strengthening their policies, capacities and activities in the area of migration and mobility’, cf. Increasing the impact of EU Development Policy: an Agenda for Change, COM (2011) 637 final 12. See also Partnership Agreement between the Members of the African, Caribbean and Pacific Group of States of the One Part, and the European Community and its Member States of the Other Part (the Cotonou Agreement)’s title on development strategies, which in Article 19(1) lays down that ‘[t]he central objective of ACP–EC cooperation is poverty reduction and ultimately its eradication; sustainable development; and progressive integration of the ACP countries into the world economy …’ and in (revised) Article 28(2) provides that ‘ACP–EU cooperation shall aim to: (a) promote peace and stability, as well as conflict prevention and resolution’ (see also Article 29(1)). 33 ‘New European Consensus’ (n 10). 34 See also EEAS press release of 30 August 2017, ‘EU migration policy delivering results thanks to partnership approach and investment in sustainable development’, available at accessed 13 October 2017. The New Consensus in para 5 also expressly states that the Global Strategy for the EU’s Foreign and Security Policy provides an overall vision for the European Union’s engagement in the world – thereby tying development cooperation and CFSP together. 35 Tanya Cox, Member of CONCORD’s Steering Group on Sustainable Development, cf. New European Consensus on Development: Double Standards for Sustainable Development, Concord – European NGO Confederation for relief and development, available at accessed 24 July 2017. OXFAM International, the development NGO, put it in even clearer words: ‘New EU development framework: self-interest trumps solidarity’ accessed 25 July 2017. In contrast, Federica Mogherini, the European Union’s High Representative, put it in the following somewhat more positive terms: ‘The European Consensus on Development is a new way of looking at development that adjusts to the changes in the international agenda in recent

EU development cooperation and the CFSP 263 Union’s development cooperation policy makes it clear that the primary focus must be upon poverty reduction, it follows from the New Consensus that the European Union’s development aid may also be used to further other objectives. Such a policy choice is not new, however.36 Indeed, arguably, this is reflected in the top ten recipients of this aid and the objectives underlying aid to these recipients, shown in Table 12.1. Table 12.1 Bilateral Official Development Assistance (ODA) to top ten recipients, 2014–15 average, gross disbursements, EU institutions Recipient Turkey Morocco Serbia Tunisia West Bank and Gaza Strip Ukraine Bosnia and Herzegovina Afghanistan Egypt Niger

USD (millions) 2,839 475 471 466 446 365 303 276 270 244

Note: It is worth pointing out that if a state has a fixed date for accession to the European Union, funding no longer qualifies as ODA. Prior to this time, funding that fulfils the ODA criteria does qualify. Source: accessed 25 July 2017.

When considering that, as a matter of principle, the primary long-term objective behind the European development cooperation policy37 is poverty eradication, it is surprising that only two of the top ten recipients (Afghanistan and Niger) also figure among the 47 nations listed on the United Nations’ ‘List of Least Developed Countries (as of June years. So, we move from a traditional approach of donor-recipient to a partnership approach in which we do things with our partners to cover all different set of fields …’ (emphasis added), ‘EU Development Ministers expected to approve Consensus for Development’ available at accessed 24 July 2017. 36 See, for example, Case C-403/05 European Parliament v Commission (Philippines Border Management Project) EU:C:2007:624, paras 56–57 as well as the Opinion of Advocate General Kokott in the same case (EU:C:2007:290), paras 72–73. It may also be noted that the European Parliament has earlier shown an unwillingness to use development cooperation funds in areas such as migration: Cécile Barbière, ‘Parliament refuses to allocate more EU aid money to migration crisis’ (EurActiv, 29 April 2016) available at accessed 30 July 2017. 37 cf. Article 208 TFEU and above Section 2.2.

264 Research handbook on the EU’s common foreign and security policy 2017)’.38 It is equally noteworthy that the number-one recipient, Turkey, is not only an upper-middle-income country, but also receives almost six times as much aid as does the second-largest recipient of EU aid (Morocco, a lower-middle-income country). If we look at the geographic location of the top ten recipients, eight are to be found in the immediate neighbourhood (Eastern Europe, Balkans, Mediterranean). In particular for those recipients bordering the Mediterranean, ‘migration control’ is a prominent objective underlying the cooperation with the European Union as reflected, inter alia, in the November 2015 Valletta summit between European and African heads of state.39 Another prominent objective underlying the European Union’s cooperation with the neighbouring countries is ‘stability’ – which to a considerable extent may be translated into ‘security’. With regards to the two top ten recipients that are categorized as least developed countries and which we do not find in the immediate neighbourhood, namely Afghanistan and Niger, we only have to look at the European Union’s key development cooperation policy papers vis-à-vis these two countries to see that, to a considerable extent, the Union’s development cooperation aid to Afghanistan and Niger addresses challenges in the fields of migration and security. For example, with regard to Afghanistan, the July 2017 Commission and High Representative communication ‘Elements for an EU Strategy on Afghanistan’40 provides as follows in the opening paragraph: ‘Since 2001, the European Union (EU) and its Member States have partnered with Afghanistan and the wider international community in pursuit of a common strategic interest in combating extremism and terrorism while simultaneously working towards peace and development in the country.’41 And when it comes to the European Union’s future objectives vis-à-vis Afghanistan, the fourth such objective set

38

List of Least Developed Countries (as of June 2017), accessible at accessed 25 July 2017. 39 In November 2015 European and African Heads of State and Government met in the Maltese capital Valletta in an effort to strengthen cooperation and address the challenges of migration. The Valletta summit constitutes a key event in this field. See further accessed 26 July 2017. 40 JOIN (2017) 31 final. 41 Emphasis added. See similarly Cooperation Agreement on Partnership and Development between the European Union and the Islamic Republic of Afghanistan (adopted by Council Decision (EU) 2017/434 of 13 February 2017 on the signing, on behalf of the Union, and provisional application of the Cooperation Agreement on Partnership and Development between the European Union and its Member States, of the one part, and the Islamic Republic of Afghanistan, of the other part [2017] OJ L67/1) where Article 1(2)(a) lists the first objective as ‘supporting peace and security in Afghanistan and the region’ whereas only the fourth objective listed in Article 1(2)(d) mentions poverty eradication.

EU development cooperation and the CFSP 265 out in the communication is described as ‘[a]ddressing challenges related to migration’.42 Similarly, within the framework of the ‘EU Strategy for Security and Development in the Sahel’ we find Niger as one of the priority countries.43 This strategy not only points to ‘the inter-dependence of security and development’, but also more specifically points out that the strategy will help the European Union attain objectives within the fields of both security44 and migration objectives.45 The above examination shows that whereas the Lisbon Treaty made poverty eradication a primary objective of European development cooperation policy, in practice other objectives have been attributed very considerable weight; this is notably so with regard to security and migration. To some extent this may be explained by the Lisbon Treaty’s revamping of the European Union’s external relations framework; especially Article 21 TEU and the creation of the position as High Representative and the European External Action Service (EEAS). In other words, in particular in these fields, the EU development cooperation policy encroaches upon the CFSP. The above examination also shows that the use of development cooperation policy for the pursuit of objectives that primarily fall under the CFSP has been reinforced by the New Consensus. At the time of writing the New Consensus is so new that it is too early to assess whether this will be reflected in the future development cooperation policy measures. 3.3 The CFSP Used for Achieving Development Cooperation Objectives In 2013 the European Parliament and the Council adopted the so-called Comprehensive Approach, which sets out to systematically establish guiding principles for EU external action across all areas, especially in relation to conflict prevention and crisis resolution.46 According to these guiding principles, in order to help developing countries get back on track towards sustainable long-term development, the Union shall assist them in the fields of early warning and preparedness, conflict prevention, crisis response and management to early recovery, stabilization and peace-building.47 The Comprehensive Approach also makes it clear that the Union itself has ‘a vital interest to prevent, prepare for, respond to, address and help recovery from conflicts, crises and other

42

Page 4 of the Communication. Available at accessed 26 July 2017. 44 Referring to ‘[t]he security threat from terrorist activity by Al-Qaida in the Maghreb (AQIM)’ that is focused on Western targets. 45 Observing: ‘Poverty creates inherent instability that can impact on uncontrolled migratory flows’ at para 13.2. 46 See Joint Communication to the European Parliament and the Council –The EU’s comprehensive approach to external conflicts and crises JOIN (2013) 30 final, and Council conclusions on the EU’s comprehensive approach, Foreign Affairs Council meeting, Brussels, 12 May 2014. See also Clément Boutillier, ‘The Security and Development Nexus’ in Jochen Rehrl (ed.), Handbook on CSDP the Common Security and Defence Policy of the European Union (Federal Ministry of Defence and Sports of the Republic of Austria 2017) 188. 47 Part I, para 3, of the Comprehensive Approach. 43

266 Research handbook on the EU’s common foreign and security policy security threats outside its borders’.48 And when it comes to policies, tools and instruments at the Union’s disposal for responding to these challenges, the Comprehensive Approach points to ‘the diplomatic, security, defence, financial, trade, development cooperation and humanitarian aid fields’.49 In other words, the European Union’s Comprehensive Approach itself directly requires the Union to pursue development objectives under the CFSP. Moreover, the ‘Global Strategy for the European Union’s Foreign and Security Policy’50 lays down, among other things, that as part of the CFSP, the European Union shall ‘invest in the resilience of states and societies to the east stretching into Central Asia, and south down to Central Africa’ and that the European Union ‘will nurture societal resilience also by deepening work on education, culture and youth to foster pluralism, coexistence and respect’.51 Thus, the Global Strategy rather unambiguously also pursues objectives that, at least in principle, fall within the European Union’s development cooperation policy. Indeed, it is not only at the general level that the European Union’s CFSP pursues development cooperation objectives. Above in Section 3.2 we have seen that there is a clear nexus between the European Union’s development cooperation under the TFEU and (in particular) security under the CFSP. Similarly, when we turn to the implementation of the CFSP it will be clear that this policy is used to further objectives that also have a clear development aspect. For example, as part of the European Union’s Strategy for Security and Development in the Sahel,52 within the framework of the Common Security and Defence Policy (CSDP)53 the European Union has established civilian missions in Somalia, Mali, Niger and Libya, which pursue both security and development objectives.54 Indeed, a key theme of the strategy is that security and development in the Sahel cannot be separated, and that helping these countries achieve security is integral to enabling their economies to grow and poverty to be reduced. Thus, whereas the primary objectives of the ‘EUCAP Sahel Mali civilian mission’ are to help the Malian government with the reform of its internal security forces to enable them to provide more security and justice for Malians, the civilian mission simultaneously seeks to further human rights and gender issues as well as capacity-building objectives such as training Malian experts to carry out oversight and auditing of the different 48

Part I, para 4, of the Comprehensive Approach. Part I, para 4, of the Comprehensive Approach. 50 accessed 29 July 2017. 51 Cf. Section 3.2 of the Global Strategy. 52 The strategy is available at accessed 14 October 2017. 53 The common security and defence policy shall be an integral part of the common foreign and security policy, cf. Article 42(1) TEU. 54 See further accessed 29 July 2017. For other examples, see Panos Koutrakos, ‘The Nexus Between the European Union’s Common Security and Defence Policy and Development’ in A Arnull, C Barnard and M Dougan (eds), A Constitutional Order of States? Essays in EU Law in Honour of Alan Dashwood (Hart Publishing 2011) 595–596. On the tensions between the CSDP and the European Union’s development cooperation policy, see Chapter 5 in this volume (text accompanying footnote 77). 49

EU development cooperation and the CFSP 267 government departments, and to manage human resources and logistics support.55 Moreover, there is a significant development component (first of all in the form of a humanitarian component) in the European Union’s military peacekeeping missions in, for example, Chad and the Central African Republic.56 As the above examination clearly shows, within the field of the CFSP the European Union (also) pursues objectives that, in principle, must be categorized as falling within its development cooperation policy. This not only applies at the level of general guiding principles as clearly reflected in the Comprehensive Approach and the Global Strategy, but also applies at the implementation level as exemplified by the European Union’s Strategy for Security and Development in the Sahel. 3.4 Choice of Legal Basis Post Lisbon – Development Cooperation Versus CFSP Even though the pillar structure that was introduced in 1993 with the Maastricht Treaty was abandoned with the entry into force of the Lisbon Treaty in 2009, the CFSP continues to be governed by special provisions that are now found in Title V, Chapter 2 of the TEU. This means that also today the CFSP and the Union’s development cooperation policy are governed by different legal regimes. For a pre-Lisbon legal measure about both development cooperation and the CFSP, it would have been necessary to determine whether it fell under what pre Lisbon was referred to as the first (development cooperation) or the second (CFSP) pillar given that the then applicable Article 47 EU asserted primacy for development cooperation over CFSP.57 With the Lisbon Treaty, Article 47 EU was replaced by Article 40 TEU, which provides both that the implementation of the CFSP shall not affect the operation of the TFEU and that the implementation of the policies set out in the TFEU, including that of development cooperation, shall not affect the operation of the CFSP so that neither of the two can assert primacy over the other.58 This means that, if a legal measure simultaneously 55

See further factsheet, ‘The EUCAP Sahel Mali civilian mission’, updated June 2016, available at accessed 14 October 2017. 56 The European Union’s military peacekeeping mission in Chad and the Central African Republic was mandated by United Nations Security Council resolution 1778, para 6. In para 6(a)(ii) this resolution lays down that the European Union’s deployment of an operation, inter alia, shall seek ‘[t]o facilitate the delivery of humanitarian aid and the free movement of humanitarian personnel by helping to improve security in the area of operations’. The UN Security Council resolution forms the legal basis for the European Council’s decision to deploy the EU operation. See also EU Military Operation in Eastern Chad and North Eastern Central African Republic (EUFOR Tchad/RCA) available at accessed 29 July 2017. 57 See in particular Case C-91/05 Commission v Council (ECOWAS) (n 26). 58 Note, however, Christophe Hillion, ‘A Powerless Court? The European Court of Justice and the Common Foreign and Security Policy’ in Marise Cremona and Anna Thies (eds), The European Court of Justice and External Relations Law: Constitutional Challenges (Hart Publishing 2014): ‘[this] raises the question of what in the end remains of the CFSP in substantive terms; which in turn questions the actual significance of Article 40(2) TEU on the protection of the CFSP integrity that the Court is called upon to guarantee’.

268 Research handbook on the EU’s common foreign and security policy pursues CFSP and development cooperation objectives, or if it has components from both fields and one of the two aims or components is identifiable as the main one whereas the other is merely incidental, the Union legislator must found the measure on the legal basis required by the main or predominant aim or component, i.e. the measure’s centre of gravity.59 In this situation, there will be one legal measure having one legal basis, but pursuing more objectives.60 In the India Agreement case61 the Court of Justice held that the European Union’s development cooperation policy is multifaceted, encompassing a broad range of areas. This, in itself, may provide a strong argument in favour of finding development cooperation (and not the CFSP) to constitute the centre of gravity. Moreover, following the entry into force of the Lisbon Treaty, in the Philippines PCA ruling62 the Court made it clear that the Lisbon Treaty’s streamlining and reorganization of the European Union’s development cooperation policy competences do not restrict the Union’s competences in this area. On the contrary, the latter judgment not only confirmed, but even further extended the Court’s earlier broad construction of the Union’s development cooperation policy competence laid down originally in the just-mentioned India Agreement ruling. This broad construction has two dimensions. The Court confirmed that the development cooperation policy was multifaceted and that, therefore, its power-conferring provisions could be used to adopt multifaceted instruments covering a broad range of policy areas.63 It also accepted that development cooperation competence could be used for relatively deep forms of cooperation. As regards the latter, the Court distinguished between cooperation of a declaratory nature and cooperation ‘in concrete terms’ (regarding the manner in which cooperation in each specific area envisaged is to be implemented). However, on substance the Court accepted that certain provisions in the Philippines Partnership and Cooperation Framework Agreement contained specific legal obligations and clear rules on how to proceed, and thereby transgressed the limits of being mere declaratory statements. The Philippines PCA case, therefore, clearly took the doctrine developed in the India Agreement case a step further. In the specific case, the Court appeared to accept that clear and central legal obligations concerning a matter which has its own treaty-making competence and a specific decision-making procedure that differs from development cooperation (namely readmission) could be assumed on 59 Case C-36/98 Spain v Council EU:C:2001:64, para 59; Opinion 2/00 Cartagena Protocol EU:C:2001:664, para 23; Case C-338/01 Commission v Council EU:C:2004:253, para 55; and Case C-91/05 Commission v Council (ECOWAS) (n 26), para 73. 60 With respect to the application of the centre of gravity test to the CFSP versus development cooperation falling under the TFEU, see Case C-658/11 European Parliament v Council of the European Union (Mauritius) EU:C:2014:2025, paras 43–62 and the comments on this ruling by Thomas Ramopoulos and Jan Wouters, Charting the Legal Landscape of EU External Relations Post-Lisbon (Working Paper No. 156, Leuven Centre for Global Governance Studies 2015) 10. For examples of the European Union institutions having found that the centre of gravity test weighs in favour of the TFEU rather than the CFSP, see Ricardo Passos, ‘The External Powers of the European Parliament’ in Piet Eeckhout and Manuel Lopez-Escudero (eds), The European Union’s External Action in Times of Crisis (Hart Publishing 2016) 124. 61 ‘India Agreement’ case (n 20). 62 ‘Philippines Partnership and Cooperation Framework Agreement’ case (n 14). 63 See also Opinion by Advocate General Sharpston in Opinion 2/15 Singapore Free Trade Agreement EU:C:2016:992, para 496.

EU development cooperation and the CFSP 269 the basis of development cooperation competence, provided that the obligations were not immediately implementable. This was the case even in a situation where particular Union procedures applied with respect to that matter (with regard to readmission special schemes applied to the United Kingdom, Ireland and Denmark). Thus, in order to be able to use the development cooperation policy competence for negotiating the Philippines Partnership and Cooperation Framework Agreement, the Commission had to ensure that the relevant provisions in the Agreement would require subsequent implementation. The Court of Justice’s approach suggests that development cooperation policy competence may be used for deeper and more concrete types of cooperation in other policy fields. This is likely also to apply to the field of the CPSP.64 In other words, when delimiting development cooperation vis-à-vis the CFSP, the ruling in the Philippines PCA case seems to strengthen the former at the expense of the latter.65 The above concerns the situation where it is possible to discern a centre of gravity for the legal measure in question. If, however, the legal measure simultaneously pursues both CFSP and development cooperation objectives or contains components from both fields, and these objectives or components are inseparably linked and neither is incidental to the other, meaning that both TEU and TFEU provisions are applicable, the measure must be founded on both legal bases.66 In this situation we must consider whether, in the specific case, it is possible to have one legal measure founded on the legal basis for both the CFSP and development cooperation. Such recourse to a dual legal basis is only possible, however, where the procedures laid down for each of the applicable legal bases are not mutually incompatible.67 With respect to the CFSP, 64

See in this respect Proposal for a Regulation of the European Parliament and of the Council amending Regulation (EU) No 230/2014 of the European Parliament and of the Council of 11 March 2014 establishing an instrument contributing to stability and peace COM (2016) 447 final/2, which is based on TFEU Articles 209 (development cooperation) and 212 (cooperation with third countries other than developing countries). This choice of legal basis is briefly explained in section 2 of the proposal. A more thorough explanation is provided in section 1.6 of the accompanying Commission Impact Assessment SWD (2016) 222 final, where the following observation is made: ‘Considering the scope of development cooperation, the financing of capacity building (training and equipment support) in the security sector on the basis of Article 209 TFEU [i.e. development cooperation] is not per se excluded only because of the military nature of the beneficiary. Being based on Article 209 TFEU, the intended measure has to come within the scope of the Union’s development cooperation policy, which is interpreted broadly in line with the Treaties and settled case law.’ 65 For a critical examination of the Court of Justice’s ruling in the Philippines PCA case, see Morten Broberg and Rass Holdgaard, ‘Demarcating the Union’s Development Cooperation Policy after Lisbon: Commission v. Council (Philippines PCFA)’ (2015) 52(2) CML Rev 547. 66 Opinion 2/00, Cartagena Protocol (n 59), para 23; Case C-211/01 Commission v Council EU:C:2003:452, para 40 and Case C-91/05 Commission v Council (ECOWAS) (n 26), para 75. Marise Cremona, ‘Coherence in European Union Foreign Relations Law’ in Panos Koutrakos (ed.), European Foreign Policy – Legal and Political Perspectives (Edward Elgar Publishing 2011) at 88 questions whether ‘Article 40 TEU precludes … the use of joint CFSP and non-CFSP legal bases for an autonomous measure or international agreement’. She appears to support the view that post Lisbon such joint legal bases are possible. 67 Case C-300/89 Commission v Council (‘Titanium Dioxide’) EU:C:1991:244, paras 17–21; Joined Cases C-164/97 and C-165/97 Parliament v Council EU:C:2003:452, para 14; Case

270 Research handbook on the EU’s common foreign and security policy unanimity within the Council and very limited participation of the European Parliament is normally required for the adoption of a Union act68 whereas, as a main rule, qualified majority in the Council as well as full participation of the European Parliament is required within the field of development cooperation.69 These differences are such as to render those procedures incompatible.70 It is therefore submitted that two separate (but symmetric) legal measures – one based on the CFSP, the other on development cooperation – must be adopted rather than one single measure founded on both legal bases.71 In contrast, when it comes to agreements between the Union and third countries or international organizations, the procedure laid down in Article 218 TFEU applies. With regard to the procedure to be followed by the Council, Article 218(8) TFEU explicitly lays down that it is to act unanimously where it is to decide upon an agreement that covers a field for which unanimity is required for the adoption of a Union act. Since qualified majority voting is very exceptional in the field of the CFSP, unanimity is required in almost all cases. Where the CFSP requires unanimity, the Council will therefore apply the same procedure both on the CFSP and on development cooperation under the TFEU.72 Turning now to the European Parliament, since the ordinary legislative procedure applies to the field of development cooperation,73 the consent of the European Parliament is required before the Council can adopt an international agreement that is based on both the CFSP and upon the development cooperation provisions of the TFEU.74 C-338/01 Commission v Council EU:C:2004:253, para 57; Case C-94/03 Commission v Council EU:C:2006:2, para 52 and Case C-178/03 Commission v Parliament and Council EU:C:2006:4, para 57. 68 cf. Article 31 TEU. 69 cf. Article 294 TFEU. 70 cf. Case C-130/10 European Parliament v Council of the European Union (Usama bin Laden) EU:C:2012:472, paras 42–49. Ramses Wessel seems to take a similar view as the one put forward here: ‘Cross-pillar Mixity: Combining Competences in the Conclusion of EU International Agreements’ in Christophe Hillion and Panos Koutrakos (eds), Mixed Agreements in EU Law Revisited (Hart Publishing 2010) 52. 71 Splitting a single legal measure into two symmetric ones due to incompatible legal bases is nothing new; see, for example, Morten Broberg, ‘Undue Assistance? An Analysis of the Legal Basis of Regulation 1257/96 Concerning Humanitarian Aid’ (2009) EL Rev 769, 776–777. 72 See, for example, Joint Proposal for a Council Decision on the signing, on behalf of the European Union, and provisional application of the Political Dialogue and Cooperation Agreement between the European Union and its Member States, of the one part, and the Republic of Cuba, of the other part JOIN (2016) 42 final. 73 cf. Article 209(1) TFEU. 74 cf. Article 218(6)(a)(v). See also Piet Eeckhout, EU External Relations Law (OUP 2012) 202–203. Where an international agreement relates exclusively to the common foreign and security policy the Council can enter into the agreement without either obtaining the consent of the Parliament or consulting it. Nevertheless, even if an agreement relates exclusively to the CFSP so that neither consent nor consultation of the European Parliament is required, the Parliament may still have a right of scrutiny in respect of that policy, see Case C-658/11 European Parliament v Council of the European Union (Mauritius Agreement) EU:C:2014:2025,

EU development cooperation and the CFSP 271

4. HIGH REPRESENTATIVE AND EUROPEAN EXTERNAL ACTION SERVICE Prior to the entry into force of the Lisbon Treaty, it was often argued that in international affairs the European Union punched below its weight, and that an important reason for this was the Union’s internal organization, or rather its lack of coherent organization. One of the key objectives behind the Lisbon Treaty, therefore, was to improve the Union’s ability to act efficiently on the international stage. To this end, the position as High Representative for Foreign Affairs and Security Policy was created.75 One of the essential functions of the High Representative is to bridge both Member State and Union interests.76 In order to enable the High Representative to carry out her tasks, she is assisted by a diplomatic service – the EEAS. This service is independent of the Member States as well as of the Council and the Commission. The establishment of the EEAS has been of particular importance for the internal organization of the Union’s development cooperation policy.77 One of the most contentious issues in relation to the establishment of the EEAS has been whether and, if so, to what extent this new actor should be in charge of development cooperation policy. This question was at the centre of the sometimes heated debates about the division of tasks between the Commission and the EEAS leading up to Council Decision 2010/427/EU establishing the organisation and functioning of the European External Action Service.78 As noted by van Vooren: ‘many in the development community were worried that giving a role to the EEAS in EU development policy was a ruse of the Member States to ensure that aid resources presently managed by the Commission would be used for strategically directed objectives rather than long-term structural development objectives’.79 In in particular paras 84–86. There might be situations where the European Parliament only has a right of consultation; see in this respect Article 218(6)(b). 75 See Article 18 TEU. Regarding to what extent the post of High Representative and the creation of the EEAS has accomplished this objective, see European Court of Auditors, Special Report: The Establishment of the European External Action Service (no 11, 2014). 76 See further Chapter 2 in this volume. 77 For an early examination of the EEAS, see the European Parliament study by Jan Wouters and others, The Organisation and Functioning of the European External Action Service: Achievements, Challenges and Opportunities (2013), available at . With particular regard to the EEAS and development cooperation policy, see pp. 49–50 of the study. 78 Council Decision, establishing the organisation and functioning of the European External Action Service, [2010] OJ L210/30. See, for instance, Honor Mahony, ‘Ashton Presents Outline of Diplomatic Service’ (EU Observer, 2010) available at accessed 25 July 2018. In his article, Mahony particularly points to the very considerable sums involved in EU development cooperation as a reason why this area came to constitute the ground for one of ‘the most bitter battles’ regarding the scope and contents of the EEAS. See also Isabelle Tannous, The Programming of EU’s External Assistance and Development Aid and the Fragile Balance of Power between EEAS and DGDEVCO (2013) 18 European Foreign Affairs Review 329. 79 See Bart van Vooren, ‘A Legal-institutional Perspective on the European External Action Service’ (2011) 48 CML Rev 475.

272 Research handbook on the EU’s common foreign and security policy other words, to transfer development cooperation policy powers from the Commission to the EEAS is not merely a technical question of institutional balance. The transfer of such competence to the EEAS was justified by a wish to attain a higher degree of coherence between the development cooperation policy and other external policies. In particular, transferring some of this competence to the EEAS would likely entail a weakening of the marked distinction which hitherto the Commission has applied between, on the one hand, African, Caribbean and Pacific (ACP) countries and, on the other hand, other developing countries. A transfer of development policy competence to the EEAS could also entail more coherence with regard to thematic divisions since the transfer to the EEAS means that subjects that previously were treated by different Directorates-General within the Commission will now be treated under one and the same roof. Indeed, as we have seen above in Section 3, the Comprehensive Approach, the Global Strategy and the New Consensus all seek to combine the European Union’s development cooperation policy and the CFSP. In the years immediately after the entry into force of the Lisbon Treaty clear warnings against the convergence of development cooperation and the CFSP were voiced by the European development cooperation community, arguing that the impetus for the EEAS to create a single, consistent external policy could mean that development cooperation and CFSP could become so tightly interwoven that it would be difficult to distinguish between them, and that the ‘soft’ development cooperation policy objectives would ‘suffer’ under the influence of other ‘harder’ and more traditional foreign policy objectives.80 As we have seen in Section 3, these warnings seem to have been justified to a considerable extent. From a legal point of view, some have claimed that the Treaties preclude the transfer of development cooperation policy responsibility to the EEAS.81 Arguments of this kind tend to rest on the Treaty provisions on external representation. Thus, Article 17(1) TEU provides that it is the Commission’s competence to ‘ensure the Union’s external representation’, whereas Articles 18 and 27 TEU explicitly provide that the High Representative is responsible for the CFSP (only), and that the EEAS is to assist the High Representative. However, as van Vooren has shown, the legal situation is considerably more complex. The better view probably is that the constitutional framework introduced by the Lisbon Treaty is ambiguous and that the Treaties leave a considerable degree of flexibility as 80 See in this respect ‘EEAS One Year On: “Work in Progress” for Poverty Eradication’, Concord Europe (2012), available at accessed 30 July 2017; and Simon Stroß, ‘Programming EU External Action Post-Lisbon: New Opportunities or Business as Usual?’ accessed 30 July 2017. See also Mark Furness, ‘Opinion: The European External Action Service’s Role in EU Development Policymaking Requires Safeguards’ (Deutsche Welle, 2010) available at accessed 30 July 2017; and the discussions in ‘Is EU Development Aid Entering a New Era in the Wake of the Lisbon Treaty?’ (Development Policy Forum 2008) available at . 81 See, for example, White & Case LLP, Memorandum entitled ‘Legal Objections to the EEAS’ Involvement in EU Development Cooperation Activities’, 16 April 2010, para 3.7.

EU development cooperation and the CFSP 273 regards the division of tasks between the Commission and the High Representative/ EEAS in the field of external relations.82 In particular, the general task of the High Representative and the EEAS to ensure consistency in EU external relations suggests that the Treaties do not preclude these new actors from playing a (partial) role also in the formulation and implementation of development cooperation policy. Arguably, the Treaties therefore do not preclude a (partial) transfer of development cooperation policy from the European Commission to the EEAS.83 In this light, as noted by Duke and Blockmans, the debate surrounding the transfer of development cooperation to the EEAS should centre less on issues of legality and more on how development-related interests can be upheld in the EEAS.84 At present, a compromise has been struck whereby the EEAS has been given the task of political coordination as regards a number of external assistance instruments.85 This is merely a power of coordination, however. Thus, the competence actually transferred to the EEAS is procedural in nature and does not concern policy issues. Consequently, as a matter of principle, management of the Union’s external cooperation programmes remains the responsibility of the Commission.86

82

See van Vooren (n 79), particularly 486–491. See likewise Simon Duke and Steven Blockmans, ‘The Lisbon Treaty Stipulations on Development Cooperation and the Council Decision of 25 March 2010 (Draft) Establishing the Organisation and Functioning of the European External Action Service’ (CLEER Legal Brief 2010) available at accessed 30 July 2017. But contrast with Mirjam van Reisen, ‘Note on the Legality of Inclusion of Aspects of EU Development Cooperation and Humanitarian Assistance in the European External Action Service (EEAS)’ (2010) available at accessed 30 July 2017, who at p. 2 observes: ‘The Lisbon Treaty defines no powers to divide development cooperation policy and humanitarian assistance between the EEAS and the Commission. Any such proposals should be regarded as illegal under the Treaty.’ 84 Duke and Blockmans (n 83) 14. 85 This is in line with the Report of 23 October 2009 from the Swedish Presidency to the European Council on the European External Action Service (DOC 14930/09) (2009) available at accessed 30 July 2017. 86 Article 9 of Council Decision 2010/427/EU. See, generally, Steven Blockmans and Christophe Hillion (eds), EEAS 2.0: A Legal Commentary on Council Decision 2010/427/EU Establishing the Organisation and Functioning of the European External Action Service (Working Paper No. 1, CLEER 2013) available at accessed 30 July 2017; Steven Blockmans and Christophe Hillion (eds), EEAS 2.0: Recommendations for the Amendment of Council Decision 2010/427/EU Establishing the Organisation and Functioning of the European External Action Service (Special Report No. 78, CEPS 2013) available at accessed 15 October 2017, and Report of the High Representative of the Union for Foreign Affairs and Security Policy of 22 December 2015 to the Council on implementing the EEAS Review, Brussels, 11 January 2016, document HR(2015) 170 (in particular section 2). 83

274 Research handbook on the EU’s common foreign and security policy

5. FINDINGS Above we have observed that the Lisbon Treaty has elevated some objectives, which pre Lisbon could be found in the chapter on the Union’s development cooperation policy, to horizontal objectives, which guide all the European Union’s actions in the field of external relations. This suggests that these objectives have been strengthened from a constitutional point of view. It is also shown that we are witnessing a generalization of objectives and principles that previously belonged to specific policy areas, thereby embedding these objectives and principles more firmly in the Union legal order. With particular regard to the European Union’s development cooperation policy we have found that making the reduction/eradication of poverty a ‘primary objective’ of European Union development cooperation policy suggests that this objective should be accorded particular weight within the field of development cooperation as well as in cases of conflict with other objectives. Apart from this preference for poverty reduction/eradication, the streamlining of the Union’s external relations objectives has not led to new hard legal obligations as regards the Union’s exercise of its development cooperation competence as compared to the pre-Lisbon situation. Even though the Lisbon Treaty has made poverty eradication a primary objective of the Union’s development cooperation policy, the above examination has also shown that in practice other objectives are attributed very considerable weight in the development cooperation policy; this is particularly so with regard to security and migration. To some extent this may be explained by the Lisbon Treaty’s revamping of the European Union’s external relations framework; in particular the streamlining of objectives provided by Article 21 TEU and the creation of the High Representative/the EEAS. In other words, in particular in the fields of security and migration, the EU development cooperation policy encroaches upon the CFSP. The above examination also shows that the use of development cooperation policy for the pursuit of objectives that primarily fall under the CFSP has been reinforced by the adoption of the New Consensus. Similarly, it has been demonstrated that the European Union’s CFSP pursues objectives that in principle fall within the development cooperation objectives – both at the general level and when we turn to the implementation of the CFSP. This is particularly so in the areas of security and migration, where the CFSP is used to further objectives that also have a clear development or humanitarian aspect. In other words, it is clear that there is mutual encroachment of the European Union’s development cooperation policy and its CFSP – first of all with regard to security and migration. In practice this means that EU legal measures may simultaneously pursue both CFSP and development cooperation objectives, or at least they may contain components from both policy fields. The Union legislator must therefore consider whether, in the specific case, it is possible to have one legal measure founded on the dual legal basis of both the CFSP and development cooperation. In this respect it is argued that, when it comes to international agreements, a dual legal basis is possible, but it is also submitted that when it comes to internal legal measures the procedural differences between the CFSP and the development cooperation policy are such as to preclude the use of a single legal measure founded on both legal bases.

EU development cooperation and the CFSP 275 Finally, turning to the organization of the European Union’s development cooperation policy and its CFSP, it is observed that a compromise has been struck whereby the EEAS has been given the task of political coordination as regards a number of external assistance instruments. However, this is merely a power of coordination, meaning that the competence actually transferred to the EEAS is procedural in nature and does not concern policy issues. Consequently, management of the Union’s external cooperation programmes continues to be the responsibility of the Commission.

13. The CFSP–humanitarian aid nexus Mireia Estrada Cañamares

1. INTRODUCTION The nexus between the Common Foreign and Security Policy (CFSP) and the humanitarian aid policy is one of the most obvious policy nexuses in EU external relations. Common Security and Defence Policy (CSDP) missions and operations – the ‘stars’ of the CFSP toolbox – are very often deployed in crisis situations where humanitarian aid is of utmost importance and is, thus, also part of the Union’s response.1 This is, however, a problematic nexus since the humanitarian aid policy is subject to the principles of humanity, neutrality, impartiality and independence. This chapter briefly looks into the CFSP and the humanitarian aid policy as separate areas of EU external action. It then focuses on the intertwined character of these two policy fields: the CFSP–humanitarian aid nexus. The chapter also highlights the risks of this nexus from the perspective of the independence and, ultimately, the success of EU humanitarian aid. It finally suggests that the quest for coherence, which was relaunched by the Lisbon Treaty reforms, has the potential of reinforcing the CFSP–humanitarian aid nexus, adding even more pressure to the independence of humanitarian aid. The goal of this chapter is twofold. First, it aims to show that there is a tension between the CFSP–humanitarian aid nexus and the need to respect the special character of EU humanitarian aid. Second, it draws attention to the fact that the quest for coherence in the Union’s external action, as expressed in the Treaty framework and in policy documents, can be thought of as a promoter of this nexus.

2. THE CFSP AND THE HUMANITARIAN AID POLICY AS SEPARATE POLICY AREAS This section introduces the CFSP and the humanitarian aid policy as separate policy areas of the Union, which is necessary to understand how they may relate to one another. It briefly analyses their Treaty-defined substantive scope and the EU actors that are essentially responsible for their definition and implementation. Given the overall topic of this volume,2 the humanitarian aid policy is considered in greater detail.

1

Note that this chapter focuses in particular on the CSDP as an integral part of the CFSP, leaving aside considerations as regards the diplomatic dimension of the CFSP. 2 For an in-depth analysis of the nature of the CFSP-CSDP as an EU policy area see Chapter 1 in this volume.

276

The CFSP–humanitarian aid nexus 277 2.1 The CFSP The Treaties are extremely open ended as regards the substantive scope of the CFSP. Article 24(1) TEU establishes that the CFSP can cover ‘all areas of foreign policy and all questions relating to the Union’s security’. The mandate of the CSDP is better articulated. Article 42(1) TEU establishes that the Union may use civilian and military assets ‘on missions outside the Union for peace-keeping, conflict prevention and strengthening international security’. Article 43(1) TEU complements the former by setting out a list of CSDP tasks and specifies that these tasks can be used in the fight against terrorism: ‘Joint disarmament operations, humanitarian and rescue tasks, military advice and assistance tasks, conflict prevention and peace-keeping tasks, tasks of combat forces in crisis management, including peace-making and post-conflict stabilisation.’ The CSDP is defined in the Treaties as an integral part of the CFSP. It was introduced into EU primary law in the Treaty of Amsterdam (1997).3 The CSDP is implemented using the civilian and military capabilities of Member States.4 Decisions on CSDP matters require the unanimity of the Council on a proposal from the High Representative of the Union for Foreign Affairs and Security Policy and Vice-President of the Commission (HRVP) or an initiative from a Member State.5 The European Council and the Council, the institutions that represent the Member States of the Union, are in charge of the definition of the CFSP.6 The HRVP is responsible for the conduct of this policy field. She chairs the Foreign Affairs Council (FAC),7 which is the configuration of the Council concerned with the CFSP. Moreover, the HRVP (assisted by the European External Action Service (EEAS)) presents proposals to the Council on CFSP matters, including recommendations regarding international agreements.8 The HRVP is also tasked with ensuring coordination between the civilian and military aspects of CSDP tasks, in close cooperation with the Political and Security Committee (PSC). Together with the President of the European Council, the HRVP represents the Union in CFSP matters.9 Institutional structures that are important in the definition and implementation of the CFSP, such as the Crisis Management and Planning Directorate (CMPD) and the EU Military Staff (EUMS), are under the EEAS and thus under the control of the HRVP.10

3 The CSDP was named European Security and Defence Policy (ESDP) until the entry into force of the Lisbon Treaty. 4 Article 42(1) TEU. 5 Article 42(4) TEU. 6 Articles 15(2) and 16(2) TEU. 7 Article 18(3) TEU. 8 Article 18(2) TEU, Council Decision establishing the EEAS (2010/427/EU) and Article 218(3) TFEU. 9 Articles 15(6) and 27(2) TEU. 10 Article 4 of the Council Decision establishing the EEAS (2010/427/EU).

278 Research handbook on the EU’s common foreign and security policy 2.2 The Humanitarian Aid Policy The first reference to humanitarian aid in EU law goes back to the second Yaoundé Convention, signed in 1969 between the then European Economic Community (EEC) and the Association of the Associated African States and Madagascar (AASM). However, it was not until the entry into force of the Lisbon Treaty that the humanitarian aid policy was recognized as a fully fledged policy area of the Union.11 In fact, the Lisbon Treaty put an end to the anomaly that had affected this field of external action since 1996. Between that time and the entry into force of the Treaty of Lisbon, EU humanitarian aid measures were adopted on the basis of Council Regulation 1257/96, which was founded on a development cooperation legal basis.12 Today, Article 214 TFEU grants humanitarian aid the status of an EU policy in its own right, and aid instruments refer to this provision as their legal basis. The humanitarian aid policy is subject to the ordinary legislative procedure, which means that the Parliament and the Council jointly adopt legislative acts that are fundamental for its development.13 It is a shared competence that is not subject to pre-emption. EU action on humanitarian aid does not prevent Member States from conducting their humanitarian aid policies and, thus, exercising their competence in this area.14 The substantive scope of the humanitarian aid policy is defined in the first paragraph of Article 214 TFEU, which establishes that EU humanitarian operations ‘[s]hall be intended to provide ad hoc assistance and relief and protection for people in third countries who are victims of natural or man-made disasters, in order to meet the humanitarian needs resulting from these different situations’. The second paragraph of the same provision determines that these operations shall comply with ‘the principles of international law and with the principles of impartiality, neutrality and non-discrimination’. Despite the phrasing, which can lead to confusion, these three principles are principles of international law (in particular, humanitarian principles) that bind the EU. In fact, this sentence constitutes a recognition that the humanitarian aid policy is bound by two out of the seven fundamental principles of the International Red Cross and Red Crescent Movement, proclaimed in Vienna in 1965 by the 20th International Conference of the Red Cross and Red Crescent Movement.15 Surprisingly, while the UN General Assembly has endorsed the three 11 On the evolution of the humanitarian aid policy in EU primary law since its early days, see P Van Elsuwege, J Orbie and F Bossuyt, ‘Humanitarian Aid Policy in the EU’s External Relations: The Post-Lisbon Framework’ (2016) 3 SIEPS 15–23. 12 Council Regulation (EC) No 1257/96 of 20 June 1996 concerning humanitarian aid [1996] OJ L163/1. This Regulation defined as its legal basis former Article 130w of the EC Treaty, which corresponds to current Article 209 TFEU, regarding the development cooperation policy. 13 Article 214(3) TFEU. 14 Article 4(4) TEU. 15 Together with the principles of impartiality and neutrality, the Fundamental Principles of the International Red Cross and Red Crescent Movement include those of humanity, independence, voluntary service, unity and universality. The definition of these principles was revised in the Statutes of the International Red Cross and Red Crescent Movement, adopted by the 25th International Conference of the Red Cross, Geneva, 1986.

The CFSP–humanitarian aid nexus 279 principles enshrined in Article 214(2) TFEU, together with that of independence, EU Member States have refused to include in the Treaties an explicit reference to the latter as a principle to be respected when implementing EU humanitarian aid operations.16 This situation differs from the European Consensus on Humanitarian Aid (2008), which is the most important policy document guiding the humanitarian aid policy, and which clearly states: ‘[t]he EU is firmly committed to upholding and promoting the fundamental humanitarian principles of humanity, neutrality, impartiality and independence.’17 The European Consensus on Humanitarian Aid, which is an agreement between the Council, the Commission, the Parliament and the Member States, defines each of these four principles. According to the principle of humanity, ‘[h]uman suffering must be addressed wherever it is found, with particular attention to the most vulnerable in the population. The dignity of all victims must be respected and protected.’ The principles of neutrality and impartiality ensure that humanitarian aid does ‘not favour any side in an armed conflict or other dispute’ and that it is provided ‘solely on the basis of need, without discrimination between or within affected populations’. Lastly, the principle of independence guarantees ‘[t]he autonomy of humanitarian objectives from political, economic, military or other objectives, and serves to ensure that the sole purpose of humanitarian aid remains to relieve and prevent the suffering of victims of humanitarian crises’. As can be seen from these definitions, the principle of independence of humanitarian action is a means to ensure that the other three principles, and those of neutrality and impartiality in particular, are protected. The question then arises as to how the Union intends to secure the impartiality and neutrality of humanitarian aid without recognizing its independence in the Treaties.18 Finally, on the institutional side, the Commission represents the Union on humanitarian aid matters, and presents proposals regarding legislative acts and recommendations on international agreements affecting the humanitarian aid policy.19 DG ECHO is the Directorate-General of the Commission in charge of the implementation of EU humanitarian aid, under the leadership of the Commissioner for Humanitarian Aid and Crisis Management.20 Moreover, as the chairperson of the FAC and the Commissioner’s Group on External Action (CGEA), the HRVP is involved in the development of the 16 The principles of impartiality, neutrality and non-discrimination were endorsed in UN General Assembly Resolution 46/182 (1991). The principle of independence was adopted in UN General Assembly Resolution 58/114 (2004). 17 See ‘The European Consensus on Humanitarian Aid’, Joint Statement by the Council and the Representatives of the Governments of the Member States meeting within the Council, the European Parliament and the European Commission (2008/C 25/01), paras 10–15. 18 On the lack of recognition of the principle of independence of EU humanitarian aid see Section 5.1. 19 Article 17(2) and (1) TEU; Article 218(3) TFEU. 20 The full name of DG ECHO is Directorate-General European Civil Protection and Humanitarian Aid Operations. This DG is also responsible for the implementation of the Union’s civil protection policy, which relates to natural or man-made disasters taking place within the EU (Article 196 TFEU).

280 Research handbook on the EU’s common foreign and security policy Union’s humanitarian aid policy.21 The FAC is the configuration of the Council where humanitarian aid matters are discussed. The CGEA brings together the different Commissioners with an external portfolio and meets on a regular basis.22

3. THE CFSP AND THE HUMANITARIAN AID POLICY AS INTERTWINED POLICY AREAS The previous section has presented the CFSP and the humanitarian aid policy as separate policy areas of external action. This section considers the interactions between the CFSP and humanitarian aid policy. It does so at the level of Treaty-defined policy objectives, at the level of policy definition and at the level of policy strategies. It is undeniable that the Treaty framework reveals a potential for the coexistence of measures adopted under CFSP and humanitarian aid legal bases in the Union’s action in specific third states. The tasks of the CSDP listed in Articles 42–43 TEU refer to the whole cycle of crisis management. They span from conflict prevention to peacekeeping and post-conflict stabilization. They include humanitarian and rescue tasks. On the other hand, Article 214 TEU establishes that, under the humanitarian aid policy, the Union will provide assistance, relief and protection for victims of natural or man-made disasters outside the EU.23 The development of the CSDP since its creation in 1997 has confirmed this potential. If Member States had decided to turn the EU into a strong military power, the CFSP–humanitarian aid nexus would not have become so obvious. However, since the first deployment of EU troops in the Former Yugoslav Republic of Macedonia (FYROM) in March 2003, under EUFOR Concordia, the CSDP has repeatedly been used to tackle crisis and post-crisis situations affecting some of the most fragile states in the world. To mention but a few examples, the CSDP has been active in the Darfur region (EU support to Amis, between 2005 and 2007), in Mali (EUTM Mali, since 2013) and in Afghanistan (EUPOL Afghanistan, between 2007 and 2016).24 These are all scenarios where the EU has also provided humanitarian aid.25 Moreover, rather than prioritizing peace enforcement through the deployment of combat forces, CSDP missions and operations have often focused on institutionbuilding in transitional and post-conflict contexts.26 Civilian crisis management has 21

Articles 17(4) TEU and 27(1) TEU. See the Decision of the President of the European Commission on the Creation of a Commissioner’s Group on External Action C (2014) 9003). On the CGEA see Section 5.1 and note 73. 23 As regards the procedural dimension of the relationship between the CFSP and the humanitarian aid policy, see Section 5.1. 24 See Chapters 5 and 6 in this volume. 25 See, for instance, the Humanitarian Aid Decision (ECHO/SDN/BUD/2006/02000), whose objective was to provide life-saving assistance to the victims of the Darfur conflict and Commission Decision (C(2014) 3028 final), which provided aid to reduce malnutrition-related mortality among children under five in the Sahel region. 26 EU NAVFOR Atalanta, which has been fighting piracy at sea off the Somali coast since 2008, stands out as an example of a CSDP operation that deploys combat forces in crisis 22

The CFSP–humanitarian aid nexus 281 clearly gained the upper hand over military crisis management, with around two-thirds of the CSDP missions deployed to date being of a civilian nature. Both the civilian and the military dimensions of the CSDP have put state institutions at the centre of their attention.27 They have provided training and advice to police, military, judicial and law enforcement institutions, thus aiming to contribute to capacity-building and reform in the rule of law and security sectors. For instance, the mandate of EUTM Somalia is to train the Somali military forces; one of the objectives of EUBAM Libya is to advise the Libyan authorities on border management-related matters; and EUCAP Sahel Niger is mandated to support Nigerian security agencies in the fight against terrorism and organized crime.28 But being present in the same external crises is not the only way in which the CFSP and humanitarian aid have come to be seen as interconnected EU policy fields. The mandates of CSDP missions and operations have also often supported humanitarian objectives. For instance, one of the objectives of EU NAVFOR Atalanta is the ‘protection of vessels of the WFP delivering food aid to displaced persons in Somalia’.29 Another example is to be found in EUFOR Chad/RCA, which was active between 2008 and 2009, and whose mandate included the protection of civilians, the facilitation of humanitarian aid delivery, and the free movement of humanitarian aid workers. From a different perspective, the Union has used CSDP assets to support EU operations implemented under the humanitarian aid policy. It did so, for example, in the floods in Pakistan (2010), in the aftermath of the Haiti earthquake (2011) and in the context of the Ebola crisis (2014).30 Wherever the Union deploys a CSDP mission or operation and provides humanitarian aid in parallel, the need to ensure complementarity between that CSDP intervention and humanitarian aid measures in place becomes central. This view explains the development of policy strategies, which are adopted as soft law instruments, encompassing these two policy areas. These documents, whose scope is either thematic or geographic, seek to maximize the complementarity of the different policies of the Union and the overall effectiveness of its action. A good example is the EU’s comprehensive approach to external conflict and crises (2013), which stresses that the Union should ‘[f]urther develop and systematically implement a common methodology management. This operation was established by Council Joint Action 2008/851/CFSP of 10 November 2008. 27 Monitoring missions have also been an important part of CSDP crisis management. For instance, the Aceh Mission monitored the implementation of the peace agreement between Indonesia and the Free Aceh Movement (GAM) between 2005 and 2006. 28 These three missions are ongoing. They were launched in 2010 (EUTM Somalia), 2012 (EUCAP Sahel Niger) and 2013 (EUBAM Libya). They were established by Council Decision 2010/96/CFSP of 15 February 2010, Council Decision 2012/392/CFSP of 16 July 2012 and Council Decision 2013/233/CFSP of 22 May 2013, respectively. 29 See Article 1(1) of Council Joint Action 2008/851/CFSP of 10 November 2008 (EU NAVFOR Atalanta) and Council Joint Action 2007/677/CFSP of 15 October 2007 (EUFOR Tchad/RCA). 30 See DG ECHO’s website on civil–military relations in humanitarian crises, available at accessed 2 July 2017.

282 Research handbook on the EU’s common foreign and security policy to conflict and crisis analysis, including development, humanitarian, political, security and defence perspectives from both the field and HQ, by all relevant available knowledge and analysis, including from Member States’.31 Another factor justifying the development of strategic policy documents covering the CFSP and the humanitarian aid, as well as other policy areas, is the central character of institution-building in the Union’s external action. As already mentioned, contributing to strengthening the capacities of state institutions is a clear priority of EU foreign policy.32 It is understood that the well-being of citizens must be protected, and it is the responsibility of states to guarantee such protection. Under this paradigm, challenges are conceived as partly a result of the inability or unwillingness of states to serve their citizens, which is referred to as weak governance in the less severe cases and state fragility in the most complex ones.33 According to the Commission, state fragility refers to ‘[w]eak or failing structures and to situations where the social contract is broken due to the State’s incapacity or unwillingness to deal with its basic functions, meet its obligations and responsibilities’.34 The focus on state fragility has contributed to the idea of a nexus between the CFSP and the rest of EU policies implemented in parallel in particular third states. All of them, including the humanitarian aid policy, have come to be seen as part of ‘the EU response’ to the problems faced by certain partner countries.35 This is why the prioritization of institution-building has justified the development of strategic policy documents encompassing the different elements of external action.

4. THE RISKS OF THE CFSP–HUMANITARIAN AID NEXUS FOR THE SUCCESS OF THE HUMANITARIAN AID POLICY This section considers why the CFSP–humanitarian aid nexus is a risk for the independence of the humanitarian aid policy. It then analyses why a non-independent humanitarian aid policy may be a problem for the success of this EU policy area. 31 European External Action Service, ‘The EU’s comprehensive approach to external conflict and crises’ JOIN (2013) 30 final 5. On the idea of the EU’s comprehensive approach, see Section 5.2. 32 The EU’s strategic approach to resilience (2017) identifies as one of its priorities strengthening ‘[t]he capacities of a state – in the face of significant pressures to build, maintain or restore its core functions, and basic social and political cohesion, in a manner that ensures respect for democracy, rule of law, human and fundamental rights and fosters inclusive long-term security and progress’. Commission, ‘A Strategic Approach to Resilience in the EU’s external action’ JOIN (2017) 21 final 3. 33 In the case of security challenges, this perception responds to a shift in the understanding of the root causes of this kind of challenge. After the Cold War, the notion of security broadened. Beside the threat of war between nations, it incorporated the need to ensure a certain degree of economic and political stability. See, for instance, C Buger and P Venesson, ‘Security, Development and the EU’s Development Policy’ (2009) EUI Working Paper 8; D Chandler, ‘The Security-Development Nexus and the Rise of “Anti-foreign Policy”’ (2007) 10(4) Journal of International Relations and Development 267. 34 Commission, ‘Towards an EU response to situations of fragility’ COM (2007) 643 final 5. 35 ibid.

The CFSP–humanitarian aid nexus 283 As explained in Section 3, the boundaries between the tasks and objectives of the CFSP and the humanitarian aid policy are diffuse. Is the facilitation of humanitarian aid delivery a CFSP task or a humanitarian aid task? Moreover, the coexistence of CFSP and humanitarian aid measures in crisis management contexts, together with the mainstreaming of institution-building in EU external action, have led to the understanding that CFSP and humanitarian aid interventions are part of ‘the EU response’ and should, therefore, be guided by common policy strategies. The CFSP and the humanitarian aid policy can hardly be perceived as totally independent of each other, either at the level of Treaty objectives, or as regards the policy strategies that guide their development and implementation, nor in view of how these two policy areas are developed in practice. This is why the CFSP–humanitarian aid nexus is intrinsically in tension with the idea of a truly independent humanitarian aid policy. We should recall that, when defining the principle of independence of humanitarian aid, the EU states that humanitarian aid objectives should remain autonomous from ‘political, economic, military or other objectives’, which ‘serves to ensure that the sole purpose of humanitarian aid remains to relieve and prevent the suffering of victims of humanitarian crises’.36 The Union is aware of this tension, which is why common policy strategies encompassing the humanitarian aid policy as well as other EU policy areas, such as the CFSP, often include references to the need to respect the independence of humanitarian aid. For instance, the EU’s comprehensive approach to external conflict and crises claims: ‘[h]umanitarian aid shall be provided in accordance with its specific modus operandi, respectful of the principles of humanity, neutrality, impartiality and independence, solely on the basis of the needs of affected populations, in line with the European Consensus on Humanitarian Aid’.37 This sort of ‘reminder’ can also be found in policy documents concerning the Union’s engagement in particular third states. For example, the National Indicative Programme for the Federal Republic of Somalia (2014–2020), which defines the general orientation for cooperation between Somalia and the EU under the European Development Fund, establishes: ‘EU engagement in Somalia is multidimensional: development cooperation is complemented by other EU institutions and instruments that focus on political dialogue, stabilisation, security sector development, crisis management, humanitarian aid and trade. The EU works more and more towards an integrated approach to Somalia.’38 This statement is complemented with a footnote, right after the reference to humanitarian aid, which specifies: EU humanitarian aid is part of the EU’s overall approach to Somalia. It is nevertheless not a crisis management tool and is provided solely on the basis of needs to preserve life, prevent and alleviate human suffering and maintain human dignity in line with the fundamental humanitarian principles of humanity, neutrality, impartiality and independence.

36

See ‘The European Consensus on Humanitarian Aid’ (n 17) para 14. ‘The EU’s comprehensive approach to external conflict and crises’ (n 31) 4. 38 Commission Decision of 11 June 2014 on the adoption of the National Indicative Programme between the European Union and Somalia COM (2014) 3715 final 6. 37

284 Research handbook on the EU’s common foreign and security policy Preserving the independence of humanitarian aid when this policy area is part of the Union’s overall approach to a particular state is obviously a challenge. But regardless of whether the EU manages or not, it will be difficult for any external observer to think of humanitarian aid as independent when this policy field is part of the EU’s overall response. This perception, alone, may challenge the success of the humanitarian aid policy.39 Moreover, as part of the EU’s overall engagement in particular third states, the humanitarian aid policy is part and parcel of a kind of engagement that is particularly politicized. In its response to external conflict and crises the Union prioritizes building state capacities. By doing so, the Union takes sides in extremely sensitive situations affecting the internal politics of third states, as institution-building measures strengthen one of the conflict parties, to the detriment of the others. This is why, through the definition of common policy objectives and the idea of ‘the EU response’, the humanitarian aid policy becomes part of the overall strategy of an actor whose intervention in third states can hardly be considered neutral with regard to internal conflict. A great example of the Union taking sides in extremely sensitive situations affecting the internal politics of third states is the continued support to the federal institutions of Somalia, including through the CSDP.40 Since the adoption of the Transitional Federal Charter of Somalia in 2004 to date, the EU’s support for the federal institutions has been unequivocal.41 Over the last decade, the Union has repeatedly stressed that the federal institutions should ensure that all stakeholders (that is, clan elders, Islamic leaders and civil society) are engaged in an inclusive reconciliation process. It has also mentioned that the government of Somalia should be broad based and reach out to all sectors of Somali society.42 However, the Union has undoubtedly positioned itself on the side of the federal institutions of Somalia. A case in point was the EU’s response to the role of the Union of Islamic Courts (UIC) in 2006. When the UIC controlled certain parts of the territory of Somalia, the EU clearly defended the transitional federal institutions. The Union did not even refer to the fact that the UIC controlled a part of the territory and were supported by a significant part of the population: ‘[t]he Council expresses its deep concern about the continuing tensions in Somalia between the UIC and the TFIs [transitional federal institutions]. The Council reconfirms its support to the TFIs as the only legitimate political representation in Somalia as defined in the 39 ‘In order to be effective, humanitarian aid must not only be neutral, but also must be perceived as such.’ Van Elsuwege, Orbie and Bossuyt (n 11) 62. 40 The Union’s engagement in the Horn of Africa is a good demonstration of the CSDP repertoire as it includes a maritime military operation (EUNVAFOR Atalanta, since 2008), a military training mission (EUTM Somalia, since 2010) and a mission with an important rule-of-law component (EUCAP Somalia, since 2012). The latter was originally called EUCAP Nestor and renamed EUCAP Somalia as of March 2017. 41 In 2004 the Council claimed: ‘[t]he EU reaffirms its commitment and stands ready to continue to support Somalia throughout the electoral process, in full respect of the sovereignty, territorial integrity, political independence and unity of the country.’ In 2017, after the conclusion of the electoral process, the Council reaffirmed its support to the newly elected government. See the Council Conclusions on Somalia (15145/04), 3 and (7614/17), para 1. 42 See, for all, the Council Conclusions on Somalia (5546/07), 3.

The CFSP–humanitarian aid nexus 285 Transitional Federal Charter (TFC).’43 Likewise, the EU has minimized the role of clan leaders in Somalia and it has often avoided referring to Somaliland as a de facto independent state. The foregoing is not to criticize the Union’s engagement in Somalia over the last decade, which is in line with UN Security Council resolutions on Somalia. It is simply an example of the eminently political character of the EU’s action in particular third states. It can give an idea of the dangers of linking the humanitarian aid policy to the Union’s overall approach to these states. The independence of humanitarian aid from the other policy areas in EU external action (including the CFSP) is clearly under pressure. The question then is why a non-independent humanitarian aid policy (or one which is simply not perceived as such) may be problematic for the success of this policy area. Katharine Derderian, Aurélie Ponthieu and Andrea Pontiroli, from Médecins Sans Frontières (MSF),44 identify at least four risks whenever humanitarian aid is perceived as being part of a political or military strategy.45 First, they argue that authorities or local actors may impede access to populations in need. Second, they state that those receiving aid may hesitate to accept it, for fear that it is politically compromised and that, by accepting it, they may be subject to retaliation by one of the conflict parties. Third, the authors recall that the safety of both humanitarian workers and those seeking assistance may be compromised.46 Fourth, NGOs may decide not to participate in humanitarian aid operations if they perceive that these are part of overall political strategies.47 The wide use of visibility waivers in the implementation of EU humanitarian aid operations by NGOs, through which these organizations are exempted from EU visibility rules, is a clear example of the will of humanitarian workers not to be perceived as part of the Union’s overall strategy. We should bear in mind that NGOs, as the main implementers of EU humanitarian aid, are essential to the success of the Union’s humanitarian aid policy. They clearly oppose the establishment of any links between the objectives of humanitarian aid and those of any other policy. This was clearly reflected in the words of Christophe Fournier, former International President of MSF, in a speech to NATO staff:48 People across cultures recognize in our vulnerability the human compassion which drives the medical act and they trust us. … I am nervous about the ‘unity of purpose’ you consider so 43

Council Conclusions on Somalia (12877/06), 2. K Derderian, A Ponthieu and A Pontiroli, ‘Losing Principles in the Search for Coherence? A Field-Based Viewpoint on the EU and Humanitarian Aid’ (29 May 2013) The Journal of Humanitarian Assistance. 45 The Union seems to be aware of these risks. ‘The European Consensus on Humanitarian Aid’ (n 17), para 10 states that respect for the humanitarian principles ‘is essential to the acceptance and ability of the EU, and humanitarian actors in general, to operate on the ground in often complex political and security contexts’. 46 The EU knows what it means to suffer an attack. The shooting down of a helicopter of the European Community Monitoring Mission (ECMM) in the context of the Croatian War of Independence in 1992 stands out as a clear example. 47 The authors claim that MSF took this decision in Afghanistan, Somalia and Northern Mali. See Derderian, Ponthieu and Pontiroli (n 44). 48 Speech delivered at a Conference in Rheindahlen (Germany), 7–8 December 2009, organized by NATO’s Allied Rapid Reaction Corps (ARRC). 44

286 Research handbook on the EU’s common foreign and security policy crucial to the achievement of your objectives. This is a ‘unity of purpose’ MSF believes is harmful to this trust. It casts doubt upon our integrity as humanitarians because people may doubt our motives or our objectives.

Minimizing the risks and ensuring that humanitarian aid is accessible where it is most needed is not possible if the humanitarian principles of impartiality, neutrality and non-discrimination are not respected. At the same time, this can only happen if the independent status of humanitarian aid is guaranteed. We should bear in mind that the principle of independence of humanitarian action is a means to secure respect for the other humanitarian principles.49 This proves particularly difficult in the context of the CFSP–humanitarian aid nexus, which justifies the claim that, from the perspective of the success of the humanitarian aid policy, this nexus is ‘a necessary evil’.

5. THE QUEST FOR COHERENCE AS A DRIVER OF THE CFSP–HUMANITARIAN AID NEXUS This section argues that the quest for coherence, which finds its clearest legal expression in Articles 7 TFEU and 21(3) TEU,50 has motivated the introduction of certain innovations in the Lisbon Treaty, such as the single set of EU external objectives.51 It claims that these reforms of the primary law of the Union are designed to strengthen the interconnections between the different policy areas of the EU. They can, therefore, have an impact on policy nexuses such as the one between the CFSP and the humanitarian aid policy. Paradoxically, although the Lisbon Treaty has conferred on humanitarian aid the status of a policy field in its own right, it has brought about a set of reforms that seriously threaten its independence.52 Sections 2, 3 and 4 of this chapter have focused on the CFSP and the humanitarian aid policy in particular. This section looks at the broader picture to draw attention to the potential impact on the CFSP–humanitarian aid nexus of a set of legal innovations (and 49 As regards the function of the principle of independence of humanitarian aid, see Section 2.2 in this chapter. 50 Article 7 TFEU establishes: ‘[t]he Union shall ensure consistency between its policies and activities, taking all of its objectives into account and in accordance with the principle of conferral of powers.’ According to Article 21(3) para 2 TEU: ‘[t]he Union shall ensure consistency between the different areas of its external action and between these and its other policies. The Council and the Commission, assisted by the High Representative of the Union for Foreign Affairs and Security Policy, shall ensure that consistency and shall cooperate to that effect.’ 51 Articles 3(5) and 21 TEU. 52 According to Jan Orbie, Peter Van Elsuwege, and Fabienne Bossuyt, ‘[i]t remains contested whether the independence of humanitarian aid has been reinforced through the insertion of a separate chapter in the Lisbon Treaty … or whether the EU’s quest for more coherence would lead to the subordination of the humanitarian imperative to overriding security, development, and trade goals (as feared by some NGOs)’ (reference omitted), J Orbie, P Van Elsuwege and F Bossuyt, ‘Humanitarian Aid as an Integral Part of the European Union’s External Action: The Challenge of Reconciling Coherence and Independence’ (2014) 40 Journal of Contingencies and Crisis Management 160.

The CFSP–humanitarian aid nexus 287 policy developments, in the case of ‘the EU’s comprehensive approach’), which are in most cases outside the Treaty provisions specific to CFSP and humanitarian aid. This section concludes by highlighting that the Union ought to strike a balance between the promotion of coherence in external action and the protection of the independence of EU humanitarian aid. The quest for coherence in EU external relations is concerned with the substance of the different policies and activities developed in the Union’s external action.53 As far as the CFSP–humanitarian aid nexus is concerned, this means that measures implemented under CFSP and humanitarian aid legal bases should not only not contradict each other (be consistent), but also complement and reinforce each other (be coherent). From this perspective, ‘consistency’ constitutes a necessary but insufficient condition for ‘coherence’. ‘Consistency’ is linked to an idea of absence of contradictions, to which ‘coherence’ adds a positive sense of creation of synergies.54 There are two fundamental reasons behind the permanent quest for coherence in the external aspect of the European project. The first one is the complex legal system in which EU actors implement the external policies of the Union. We may call this the internal factor. The second one is the fact that the Union considers the coherence of its foreign policy as a conditio sine qua non to its effectiveness on the international panorama. This constitutes the external factor. The EU often refers to its wide toolbox for responding to international concerns.55 In fact, the tools of the Union correspond to many policies and activities, which are implemented under different policy areas (e.g. the humanitarian aid policy and the development cooperation policy) and under different instruments (e.g. the Instrument for Humanitarian Aid and the Development Cooperation Instrument). Furthermore, depending on the policy area and instrument used, the role of EU actors differs. For instance, the Commission does not play the same role in the common commercial policy as in the CFSP. Moreover, there is no hierarchy between policy fields (e.g. CFSP and humanitarian aid policy), actors (e.g. the Commission and the Council), instruments (e.g. European Development Fund and the Instrument for Humanitarian Aid) and policy objectives (e.g. human rights and international security). In addition, the law of EU external relations does not offer a clear-cut rule regarding the choice of the right policy area to tackle international concerns. In particular areas that are closely linked on the ground, such as the CFSP–development cooperation nexus, the choice of the

53 Note that the analysis of the quest for coherence in EU external relations presented in this chapter focuses on the horizontal perspective of coherence (between the different policies of the Union), as opposed to the vertical perspective (between the policies of the Union and those of its Member States). 54 See, for example, C Hillion, ‘Tous pour un, un pour tous! Coherence in the External Relations of the European Union’ in M Cremona (ed.), Developments in EU External Relations Law (OUP 2008) 14; S Duke, ‘Consistency, Coherence and European Union External Action: The Path to Lisbon and Beyond’ in P Koutrakos (ed.), European Foreign Policy: Legal and Political Perspectives (Edward Elgar Publishing 2011) 17–18. 55 A great example can be found in the Council Conclusions on the EU’s Comprehensive Approach (9644/14), 2.

288 Research handbook on the EU’s common foreign and security policy legal basis is a source of interinstitutional conflict.56 The dividing line between the activities that should fall under express external policy fields of the Union and those that should be conducted as part of the external dimension of EU internal policies is not clear either.57 The multiplicity of policy areas and instruments, where specific policies and activities are implemented, and where actors play different roles, generates legal complexity in the Union’s external action. This is why, while the idea of the wide toolbox is attractive and makes sense, it often hides the fact that organizing the different tools is no easy task. By ensuring coherence between the different policies used to address a particular international concern, it is assumed that the Union shows that it is able to overcome the challenges of its complex system for external relations. But coherence would not be so important in EU external relations if the Union did not consider it a conditio sine qua non to its effectiveness as an international actor: ‘[t]he EU is stronger, more coherent, more visible and more effective in its external relations when all EU institutions and the Member States work together on the basis of a common strategic analysis and vision.’58 The link between the concepts of coherence and effectiveness of the Union as an international actor can be found in the Treaties and case law of the Court of Justice.59 Often this link points to the fact that internal coordination leads to resources being spent in a reasonable manner and, thus, to a greater impact of the Union’s action. This is, for instance, the case for Article 210(1) TFEU, which establishes that the EU and its Member States will coordinate their policies on development cooperation ‘to promote the complementarity and efficiency of their action’. In other cases, the coherence–effectiveness link finds its justification in the idea of the EU’s actorness. The assumption is that the Union needs to act in a coherent manner to show its identity as an international player, which is indispensable for it to be ultimately effective as regards the objectives it pursues at the international level.60 Being visible is obviously not enough to be an effective international actor. But by adhering to its founding principles, regardless of the policy area or instrument used, the EU assumes that it can show not only its existence as an international player, but also 56

Case C-91/05 Commission v Council (‘ECOWAS’ case) ECLI:EU:C:2008:288 stands out as the seminal example of inter-institutional conflict regarding the correct choice of the legal basis in the CFSP–development cooperation policy interface before the entry into force of the Lisbon Treaty. A similar conflict in the post-Lisbon scenario can be found in Case C-658/11 Parliament v Council (‘Mauritius’ case) ECLI:EU:C:2014:2025, para 60. 57 For instance, Operation EU NAVFOR Sophia, the EU military response to human smuggling and trafficking in the Southern Central Mediterranean, was established under a CFSP legal basis only. However, the additional legal basis of the Area of Freedom, Security and Justice would have been desirable. See Council Decision (CFSP) 2015/778 of 15 May 2015 and Article 67 TFEU. 58 ‘The EU’s comprehensive approach to external conflict and crises’ (n 31) 3. 59 See Articles 24(3) para 2 and 32 para 1 TEU, in the CFSP context. See also Case C-266/03 Commission v Luxembourg ECLI:EU:C:2005:341, para 60 and Case C-433/03 Commission v Germany ECLI:EU:C:2005:462, para 66. 60 In fact, in the pre-Lisbon legal framework, ascertaining the EU’s identity was one of the objectives of the CFSP. Currently, it is not an explicit external objective but it is mentioned in the Preamble of the TEU as one of the functions of the CFSP.

The CFSP–humanitarian aid nexus 289 its identity as an actor that is founded on certain values and principles, and that it is committed to upholding them and promoting them in its external action.61 Perhaps worried about the latest developments in the EU (including the response to the refugee crisis) the Global Strategy for the European Union’s Foreign and Security Policy (2016) recalls: ‘[l]iving up consistently to our values internally will determine our external credibility and influence’.62 The quest for coherence is, thus, not only linked to the way in which the Union deals with its complex system for EU external relations internally. It also has an external dimension as it is perceived as having a direct impact on the EU’s international actorness and, ultimately, its effectiveness on the ground. 5.1 The Treaty Framework The most important Treaty provisions governing the relationship between the CFSP and the humanitarian aid policy can be said to be Article 40 TEU, on the procedural side; Article 21 TEU, on the substantive side; and Articles 18 and 27 TEU, on the institutional side. These provisions are either brand new or have been modified significantly after the entry into force of the Lisbon Treaty. All of them are guided by the Union’s quest for coherence in EU external relations.63 The explicit references to coherence in the Treaties have also been reinforced since the Treaty of Lisbon came into force. The location of coherence under Article 7 TFEU indicates that it is a principle of general application to the Union. Coherence can also be found in Article 21(3) TEU, on general provisions on the EU external action.64 Since the entry into force of the Lisbon Treaty, these two provisions are within the jurisdiction of the Court of Justice. As regards the procedural dimension of the relationship between the CFSP and humanitarian aid policy, the Treaties establish a non-affectation clause between CFSP and non-CFSP policy areas.65 The choice of the legal basis protects the procedures applicable to the implementation of CFSP and non-CFSP policy fields as well as the powers of the institutions within these procedures. This means that, unlike before the entry into force of the Lisbon Treaty, the CFSP and the humanitarian aid policy coexist in the Union’s external action in a non-prioritized manner. When the EC and the EU constituted separate legal orders, the focus was on protecting the scope of Community competences from the Union. If a legal act could have been adopted on the basis of EC 61

Articles 2, 3 and 21 TEU. For Marangoni and Raube, ‘coherence makes the EU’s commitment to comprehensive and global objectives credible’. AC Marangoni and K Raube, ‘Virtue or Vice? The Coherence of the EU’s External Policies’ (2014) 36(5) Journal of European Integration 478. 62 ‘Shared Vision, Common Action: A Strong Europe: A Global Strategy for the European Union’s Foreign and Security Policy’ (European External Action Service 2016) 15. 63 According to Cremona, coherence ‘finds its expression in a number of legal provisions and principles’; it ‘operates to bring together – to structure – these fundamental legal principles’. M Cremona, ‘Coherence in European Union Foreign Relations Law’ in P Koutrakos (ed.), European Foreign Policy: Legal and Political Perspectives (Edward Elgar Publishing 2011) 59. 64 See these Treaty provisions in note 50. 65 Article 40 TEU.

290 Research handbook on the EU’s common foreign and security policy competences, an EU legal basis was not acceptable.66 The abolition of the pillar structure of the Lisbon Treaty has eliminated the previous imbalance and led to a situation where the choice of the legal basis essentially protects the powers of the institutions, rather than the substantive scope of competences. On the substantive side, Article 21 TEU provides all political actors involved in the development of EU external relations with a single set of objectives explaining what the Union aims towards in its external action.67 Although it obviously does not rule out the possibility of policy incoherencies, the single set of objectives can bring about a certain degree of coherence between policies and activities, enabling the Union to show a single identity on the international stage, regardless of the policy area or the instrument deployed.68 In fact, Article 21 TEU can be perceived as a commitment to the founding principles of the Union (‘which have inspired its own creation’). These are to be observed, as regards what the Union upholds, as well as what it promotes, in the development of its external relations (‘which it seeks to advance in the wider world’). Both the CFSP and the humanitarian aid policy are bound by the same set of objectives, which include the safeguarding of the Union’s security, the preservation of peace, the prevention of conflicts and the strengthening of international security.69 Article 214 TFEU recalls that the humanitarian aid policy is guided by these objectives, by explicitly indicating so and by purposely avoiding any reference to the principle of independence of humanitarian aid. This is obviously problematic from the perspective of the independence of EU humanitarian aid as the objectives of the latter are, in principle, apolitical. As expressed by Christophe Fournier, former International President of MSF: ‘I want to reassure you that the MSF is neither pacifist nor antimilitary. That may sound strange for an organization that was awarded the Nobel Peace Prize a decade ago, but we are not pro-peace. We do not stand in judgement of either your objectives or those of your enemies.’70 Lastly, on the institutional side, the Treaties have entrusted the HRVP with important responsibilities regarding the Union’s quest for coherence in external action.71 Through her many roles, the HRVP (and the EEAS as her assisting body) should ensure coherence between CFSP and non-CFSP policies. The HRVP is mandated to do so by preparing proposals to the development of the CFSP, chairing the FAC, proposing and 66

‘The Union cannot have recourse to a legal basis falling within the CFSP in order to adopt provision which also fall within a competence conferred by the TEC on the Community’. ‘ECOWAS’ case (n 56) para 77. Ex-Article 47 TEC established the primacy of the acquis communautaire over EU competences. 67 The single set of EU external objectives is also enshrined in Article 3(5) TEU. 68 In the ‘Tanzania Agreement’ case, the CJEU referred to the single set of objectives (Article 21(2) TEU) and to the notion of coherence in the external action (Article 21(3) TEU) as justifying why compliance with the rule of law and human rights, as well as respect for human dignity, is required of all EU policies and activities. Case C-263/14 Parliament v Commission (‘Tanzania Agreement’ case) ECLI:EU:C:2016:435, para 47. 69 Article 21(2)(a) and (c) TEU. 70 See note 48. 71 Article 21(3) para 2 TEU. The role of the European Council in fostering coherence across policy areas by identifying the strategic interests and objectives of the Union deserves a mention too (Article 22(1) TEU).

The CFSP–humanitarian aid nexus 291 having control over the EU Special Representatives (EUSR),72 and participating in the College of Commissioners as Vice-President of the Commission. The HRVP chairs the CGEA, which has been relaunched after under the Presidency of Junker.73 Two visible examples of the cooperation between the HRVP and the Commissioner for Humanitarian Aid are the presentation of joint initiatives and the publication of joint press releases on behalf of the HRVP and the Commissioner.74 The HRVP is also entrusted with the responsibility for ensuring the complementarity of CFSP and humanitarian aid action by controlling the EEAS, both in Brussels (headquarters) and on the ground (Union delegations).75 The fact that Union delegations are now an integral part of the EEAS has great potential from the perspective of the coordination of CFSP and humanitarian aid measures concerning particular third states. Moreover, the creation of the EEAS has added a new impetus to the cooperation between the CSDP crisis management structures (which are now part of the EEAS) and the Commission’s units dealing with crisis management within DG ECHO. By way of example, DG ECHO takes part in the EEAS Crisis Platform, which meets on an ad hoc basis and brings together the EEAS crisis response and management structures (e.g. CMPD and CPCC), the EEAS and the Commission’s relevant geographical desks and the EU Military Committee. Another example is the involvement of DG ECHO in the planning of CSDP missions and operations, which, according to the Commission, has had an impact in their mandates. The Commission mentions the example of EUTM Mali, which trained the Malian military in the notions of international humanitarian law as a result of the input of DG ECHO.76 5.2 The EU’s Comprehensive Approach Driven by the single set of objectives and the right institutional set-up, strategic policy documents covering the different policies and activities of the Union in a geographic or thematic context have become commonplace in the past few years. Be it through Council conclusions or joint communications of the Commission and the HRVP, strategic policy documents in the field of external relations can be linked to the HRVP and the EEAS.77 As already explained in this chapter, these soft law documents, often labelled ‘comprehensive approaches’, include considerations regarding the CFSP and 72

Article 33 TEU. See Section 2.1 and note 22. For an in-depth analysis of the role of the CGEA between 2014 and 2015, see S Blockmans and S Russack, ‘The Commissioner’s Group on External Action – Key Political Facilitator’ (2015) CEPS Special Report 125. 74 See, for instance: Commission, ‘Elements for an EU regional strategy for Syria and Iraq as well as the Da’esh threat’ JOIN (2015) 2 final and the Statement by HRVP Federica Mogherini and Commissioner Christos Stylianides on the situation in Syria (3 August 2016). 75 Article 221(2) TFEU. 76 See the document ‘Civil-military relations in humanitarian crises: Why do we invest in civil-military relations?’ published on DG ECHO’s website, available at accessed 2 July 2017. 77 ‘The EU’s comprehensive approach to external conflict and crises’ (n 31) refers to the existence of a new institutional framework, whereby the EU has ‘the increased potential and the ambition … to make its external action more consistent, more effective and more strategic’. 73

292 Research handbook on the EU’s common foreign and security policy the humanitarian aid policy.78 They put the independence of EU humanitarian aid under pressure. Soon after the idea of the EU’s comprehensive approach came to light, Voice adopted a resolution expressing the concerns of NGOs regarding this development: ‘[l]inking humanitarian aid to political goals through an EU comprehensive approach is an issue of concern to the humanitarian community, particularly in complex crises, as it will make it more difficult for us to operate and to achieve our humanitarian mission’.79 As strategic policy documents, comprehensive approaches are directed at advancing coherence in EU external relations. They should inform the policies and activities falling within their geographic or thematic scope, and thus ensure coherence at the level of individual measures. Since they do not require a legal basis, they can bridge the CFSP and non-CFSP divide, providing separate fields of EU action with a common sense of direction.80 The importance of common strategic documents in the field of EU external relations should not be underestimated. After all, the Court of Justice relies on this kind of document to define the scope of EU external policy areas. Perhaps the most relevant example is the extent to which the CJEU referred to the European Consensus on Development (2006)81 to stress the existence of an agreement between the EU institutions and the Member States over the broad scope of the development cooperation policy. We may also note that, at a more abstract level, the narrative of the ‘EU’s comprehensive approach’, which is not an EU invention,82 seems to have served the HRVP and the EEAS to turn the wide toolbox into a distinctive feature of the Union as an international actor. The EU tells the world that it is in a unique position to present integrated responses to international matters:83 The European Union and its Member States can bring to the international stage the unique ability to combine, in a coherent and consistent manner, policies and tools ranging from diplomacy, security and defence to finance, trade, development and human rights, as well as 78

See Sections 3 and 4. ‘Humanitarian aid and the EU comprehensive approach: recommendations’ (2013) Voice General Assembly Resolution. Note that Voice is a network representing 85 NGOs acting in humanitarian aid worldwide. 80 ‘A Union without a clear idea of what it is trying to do and where it is going will remain incoherent.’ Duke (n 54) 19. 81 See the European Consensus on Development (2006/C 46/01) and, by way of example, Case C-377/12, Commission v Council (‘Philippines II Case’) para 19. Note that the Consensus on Development of 2006 has been replaced by The new European Consensus on Development ‘Our world, our dignity, our future’, Joint statement by the Council and the representatives of the governments of the Member States meeting within the Council, the European Parliament and the Commission (2017/C 210/01). 82 According to Drent: ‘Nowadays, most national governments and international organisations dealing with security challenges have at least a reference to “comprehensiveness” in their crisis management operations in their policy documents’, M Drent, “The EU’s Comprehensive Approach to Security: A Culture of Co-ordination?” (2011) LXIV-2 Studia Diplomatica. Furness argues that, when it comes to fragile states, the EU is not ‘an intellectual leader’ but rather an importer of international approaches that adds to them a ‘European twist’, M Furness, ‘Let’s Get Comprehensive: European Union Engagement in Fragile and Conflict-Affected Countries’ (2014) 5 DIE 7. 83 Council Conclusions on The EU’s Comprehensive Approach (9644/14), 2. 79

The CFSP–humanitarian aid nexus 293 justice and migration. This contributes greatly to the Union’s ability to play a positive and transformative role in its external relations and as a global actor.

According to Linda Barry,84 the international security environment of the 21st century, dominated by the rise of transnational security threats, has allowed the EU to ‘enhance its added value as a global security actor’. In her view, ‘the uniquely broad range of instruments in its toolbox compared with other multinational security organisations’ has placed the EU in a privileged position to operationalize the comprehensive approach. 5.3 A Critical Approach The Lisbon Treaty brought about a set of reforms that ease the way to strengthening the nexuses between the humanitarian aid policy and other EU policy areas such as the CFSP. These reforms, which are guided by the long-standing quest for coherence of the Union, increase the threats to the independence of the humanitarian aid policy. This area of EU action is now guided by a single set of objectives that includes political and military objectives. Rules on the choice of the legal basis are not designed to protect the substantive scope of EU humanitarian aid, but rather the procedures for its development and implementation, as well as the powers of EU institutions within these procedures. Furthermore, the humanitarian aid policy falls under the narrative of the EU’s comprehensive approach, together with other policy areas such as the CFSP. If the Union intends to respect the humanitarian principles to which it is bound, it ought to strike a balance between the promotion of coherence between EU policies and activities, and the need to preserve the autonomy of its humanitarian aid.85 Finding that balance is no easy task as the quest for coherence in the external action and the narrative of the comprehensive approach are closely linked to the Union’s international visibility and political actorness. These notions are clearly at odds with the nature of the humanitarian aid policy as the latter needs to be perceived as apolitical and to remain to a certain extent invisible in order to be effective. The quest for coherence in the context of the CFSP–humanitarian aid nexus should, therefore, be fostered by focusing on efforts that, while promoting the complementarity between CFSP and humanitarian aid measures, do not have a clear external projection. By way of example, cooperation between DG ECHO and the EEAS in the planning of CSDP missions and operations can contribute to the complementarity of CFSP and humanitarian aid action. Joint statements by the HRVP (on CFSP matters) and the Commissioner for Humanitarian Aid can fulfil the same purpose. While both can promote the coherence of the Union’s external action, the external projection of joint statements has a negative impact on the perception of humanitarian aid as independent. This is why cooperation between DG ECHO and the EEAS’ crisis management structures is better suited to strike the balance that this paragraph refers to. 84 L Barry, ‘European Security in the 21st Century: The EU’s Comprehensive Approach’ (2012) IIEA European Security and Defence Series 1–2. In a similar direction, Furness claims that the comprehensive approach is ‘an expression of the EU’s long-held ambition to develop whole-of-EU approaches to complex external relations challenges’, Furness (n 82) 10. 85 On the tension between the Union’s quest for coherence and the independence of EU humanitarian aid, see Orbie, Van Elsuwege and Bossuyt (n 52).

294 Research handbook on the EU’s common foreign and security policy By the same token, the narrative of the EU’s comprehensive approach, which can be powerful to strengthen the Union’s international actorness, is particularly dangerous for the independence of EU humanitarian aid and, ultimately, its effectiveness. The same elements that make this narrative powerful – that the Union can respond to complex international concerns with multiple tools – impact negatively on the perception of humanitarian aid as independent. The latter appears as one of the many ‘tools’ of the Union’s ‘comprehensive’ response. From this perspective, the special character of the humanitarian aid policy stands out as a challenge for the operationalization of the EU’s comprehensive approach to crisis management. The single set of objectives enshrined in Article 21 TEU, which has great potential from the perspective of promoting coherence by ensuring that all policies and activities are guided by the same objectives, is also at odds with the independence of EU humanitarian aid. A better balance could have been found in Article 214 TFEU between the quest for coherence and the independence of humanitarian aid. It would probably have sufficed not to include the reference to the principles and objectives of external action in paragraph 1 of this provision,86 and to enshrine the principle of independence of humanitarian aid in paragraph 2.

6. CONCLUDING REMARKS This chapter has analysed the interconnections between the CFSP and the humanitarian aid policy. The boundaries between the tasks and objectives of these two policy areas are not always clear. Moreover, the coexistence of CFSP and humanitarian aid measures in crisis and post-crisis situations outside the Union’s borders, together with the mainstreaming of institution-building in EU external action, have led to the understanding that CFSP and humanitarian aid interventions are part of ‘the EU response’ and should, therefore, be guided by common policy strategies. The tension between the CFSP–humanitarian aid nexus and the independence of EU humanitarian aid has also been considered, as well as the risks of this nexus for the effectiveness of humanitarian aid. For instance, NGOs may decide not to participate in the implementation of humanitarian operations if they take the view that humanitarian aid is part of the Union’s overall approach to the challenges faced in a particular country or region. We should keep in mind that NGOs, as the main implementers of EU humanitarian aid, are essential to the success of the Union’s humanitarian aid policy. This risk is exacerbated by the fact that, by putting the state at the centre of its attention, the EU often takes sides in the internal politics of third states. As shown by the example of the Union’s approach to Somalia in past few years, the EU is an actor whose intervention in third states can hardly be considered neutral. This chapter has also examined a set of legal innovations introduced by the Lisbon Treaty, which are motivated by the quest for coherence in EU foreign policy. These reforms have great potential for strengthening policy nexuses, such as the one between 86 The first sentence of Article 214(1) TFEU establishes that ‘the Union’s operations in the field of humanitarian aid shall be conducted within the framework of the principles and objectives of the external action of the Union’.

The CFSP–humanitarian aid nexus 295 CFSP and humanitarian aid. The reasons behind the importance that is attributed to the Union’s quest for coherence have also been explained. By ensuring coherence between different policies and activities, it is assumed that the Union shows that it is able to overcome the challenges of its complex system for external relations. This is considered indispensable for it to be effective as regards its external objectives. However, this chapter has also argued that the EU ought to strike a balance between promoting coherence between different policies and activities and securing the independence of humanitarian aid. In most cases, this balance could be found by seeking to advance policy complementarity by prioritizing efforts that do not have a clear external projection. For example, cooperation between DG ECHO and the EEAS in the planning of CSDP missions and operations is better suited than joint statements by the HRVP (on CFSP matters) and the Commissioner for Humanitarian Aid to strike that balance. Since joint statements have a clear external projection, they raise the perception that humanitarian aid is connected to CFSP objectives. In other instances, it seems as though the quest for coherence has prevailed over the protection of the independence of EU humanitarian aid. The lack of Treaty recognition of the principle of independence of humanitarian aid needs to be reviewed. On a final note, we ought to bear in mind that the existence of the CFSP– humanitarian aid nexus and the uneasiness of humanitarian workers as regards the comprehensive approach to crisis management can be considered a sort of recognition of the role that the CSDP has acquired in crisis situations affecting some of the most vulnerable states in the world.

14. The nexus between CFSP/CSDP and the Area of Freedom, Security and Justice Panos Koutrakos

1. INTRODUCTION The Area of Freedom, Security and Justice (AFSJ) has a set of objectives that is startlingly broad. It includes the following: the absence of internal border controls for persons; the development of a common policy on asylum, immigration and external border controls; a high level of security through measures to prevent and combat crime, racism and xenophobia; increased cooperation between police, judicial and other competent authorities; the mutual recognition of judgments in criminal matters and, if necessary, the approximation of criminal laws; and the mutual recognition of judicial and extrajudicial decisions in civil matters.1 It is not, however, only this broad sweep that makes the AFSJ well suited for interacting with the Common Foreign and Security Policy (CFSP) and the Common Security and Defence Policy (CSDP). It is also the security focus of these policies. Whilst focusing on EU citizens,2 AFSJ owes its genesis and development to the need to ensure the internal security of the EU.3 Given the increasing multiplication of security challenges and the porous nature of physical borders, the development of the external aspects of AFSJ was hardly surprising.4 And given the broad scope of CFSP (it covers ‘all areas of foreign policy and all questions relating to the Union’s security’5), it was only a matter of time before the linkages between it and AFSJ would emerge. The linkages between the two policy areas are more explicit in the revamped framework that has governed the EU’s external action since the entry into force of the Treaty of Lisbon. The objective of ‘contribut[ing] to the protection of its citizens’ is viewed in Article 3(5) TEU as part of the Union’s aims in its relations with the wider world. More specifically, primary law refers to safeguarding, amongst other things, the Union’s security, as an objective not only of the EU’s external action (Article 21(2) TEU), but also of the external aspects of its other policies (Article 24(3) TEU). 1 Art. 67(2)–(4) TFEU. The AFSJ is governed by Title V TFEU, and covers immigration (Art. 77 TFEU), asylum (Art. 78 TFEU), judicial cooperation in civil (Art. 81 TFEU) and criminal matters (Arts 82–6 TFEU), and police cooperation (Arts 87–9 TFEU). 2 Art. 3(2) TEU provides that the EU ‘shall offer its citizens an area of freedom, security and justice without internal frontiers’ (emphasis added). 3 See, for instance, ‘A Strategy on the External Dimension of the Area of Freedom, Security and Justice’ COM(2005) 491 fin, adopted by the Council in December 2005. 4 See J Monar, ‘The External Dimension of the EU’s Area of Freedom, Security and Justice: Progress, Potential and Limitations after the Treaty of Lisbon’ (2012) 1 SIEPS 13ff. 5 Art. 24(1) TEU.

296

The nexus between CFSP/CSDP and the AFSJ 297 On the other hand, CFSP and AFSJ are ill-suited in legal terms. The former is covered by a sui generis competence of the EU (Article 2(4) TFEU) and ‘is subject to specific rules and procedures’ (Article 24(1) TEU), in the context of which unanimity prevails, the role of the European Parliament is marginal at best, and the jurisdiction of the Court of Justice of the European Union is limited.6 As for the AFSJ, it is covered by shared competence (Article 4(2)(j) TFEU), the ordinary legislative procedure applies, the Parliament acts as co-legislator, and the conduct of the policy is subject to the jurisdiction of the Court of Justice. In light of the above, the CFSP–AFSJ nexus provides a good snapshot of the complexities that characterize the Union’s effort to bring together different strands of its external action in order to become a relevant actor on the international scene. This chapter will explore these complexities from three different angles. The first is about policy: it will focus on the increasingly prominent linkages that emerge from the current conception of CFSP/CSDP and AFSJ by the Union’s decision-makers. The second angle is about practice: it will explore how the nexus between the two policies works in CFSP practice, by focusing on its most direct and prominent illustration, that is Operation Sophia in the Southern Central Mediterranean. The third angle is about the approach of the Court of Justice to the CFSP–AFSJ nexus and its implications for both the Union’s institutions and the place of CFSP in the Union’s constitutional order.

2. THE POLICY ANGLE The external dimension of AFSJ is, ostensibly, about the development of relationships between the EU and third countries and organizations in order to achieve the objectives relating to the Union’s internal security. This dimension was acknowledged as early as in 1999 by the European Council7 and has emerged clearly over the years on the basis of various policy documents.8 In fact, there is a plethora of policy documents in the area, and they are by no means immune to the empty rhetoric, triteness and hyperbole that one is accustomed to find in this type of work produced by the EU’s bureaucracies 6 Art. 24(2) TEU and Art. 275 TFEU. On this issue, see Chapter 4 in this volume. On the procedural aspects of CFSP/CSDP, see Chapter 2. 7 Tampere European Council, Presidency Conclusions on the Area of Freedom, Security, and Justice (15–16 October 1999). 8 For the foundational documents, see The Hague Programme: Strengthening Freedom, Security and Justice in the European Union [2005] OJ C53/01, and The Stockholm Programme – An open and secure Europe serving and protecting the citizens (Brussels, 2 December 2009) adopted by the European Council (11–12 December 2009). See the analysis in T Balzacq (ed.), The External Dimension of EU Justice and Home Affairs – Governance, Neighbours, Security (Palgrave Macmillan 2009); M Cremona, J Monar and S Poli (eds), The External Dimensions of the European Union’s Area of Freedom, Security and Justice (PIE Peter Lang 2011); C Flaesch-Mougin and LS Rossi (eds), La dimension extérieure de l’espace de liberté, de sécurité et de justice de l’Union (Bruylant 2012); and R Wessel and C Matera, ‘The External Dimension of the EU’s Area of Freedom, Security and Security’ in C Eckes and T Konstadinides (eds), Crime within the Area of Freedom, Security and Justice: A European Public Order (CUP 2011) 272.

298 Research handbook on the EU’s common foreign and security policy and politicians. Taken together, nonetheless, these documents convey the prevailing understanding of the institutions about the increasing relevance of CFSP and AFSJ to each other. The main priorities of the external dimension of AFSJ were defined early on as migration policy, the fight against organized crime and terrorism, the fight against crimes such as money laundering, corruption and trafficking in human beings, the fight against drug trafficking, and the development and consolidation of the rule of law in countries on the path to democracy.9 These issues also featured prominently in the main strategic documents on CFSP. The 2003 European Security Strategy, for instance, included organized crime in the Union’s major challenges, and referred specifically to cross-border trafficking in drugs, women and illegal migrants, as well as their potential link to terrorism.10 In fact, the nexus between AFSJ and CFSP has been shaped by three interlinked aspects of the Union’s external relations in general and its foreign and security policy in particular. The first is the wide terms in which the EU has construed the notion of security. This characteristic has been prevalent in the efforts of the Union to articulate its international role. The tone was set in the 2003 European Security Strategy, which referred to terrorism, proliferation of weapons of mass destruction, regional conflicts, state failure, and organized crime as the main global challenges for the EU.11 The policies covered by the AFSJ pertained to all of them, and their role was affirmed in a 2008 review of the European Security Strategy,12 where, for instance, organized crime assumed central importance, alongside terrorism. This approach is also illustrated by the 2016 Global Strategy, which stresses, in the context of CSDP, operations and missions aiming to fight cross-border crime and disrupt smuggling networks,13 a point that will be explored further in the following section. The second characteristic, following from the above, pertains to the Union’s armoury to tackle global security challenges, and is about the wide range of instruments upon which the EU is prepared to rely. As the Global Strategy puts it, ‘[a] stronger Union requires investing in all dimensions of foreign policy, from research and climate to infrastructure and mobility, from trade and sanctions to diplomacy and development’.14 To that effect, it promotes an approach which would be ‘integrated’, ‘multi-dimensional … through the use of all available policies and instruments aimed at conflict prevention, management and resolution’, ‘multi-phased’, that is involving all stages of the conflict cycle, and ‘multi-level’, covering the local, national, regional and global levels.15 There are, in other words, inherent linkages between, amongst other things, 9

Santa Marie de Feira European Council, 19–20 June 2000, Presidency Conclusions. ‘A Secure Europe in a Better World – European Security Strategy’ (Brussels, 12 December 2003) 4–5. 11 ibid, 3–5. 12 ‘Report on the Implementation of the European Security Strategy – Providing Security in a Changing World’ (S407/08, Brussels, 11 December 2008). 13 ‘Shared Vision, Common Action: A Stronger Europe – A Global Strategy for the European Union’s Foreign and Security Policy’ (European External Action Service 2016). 14 ibid, 44. 15 ibid, 28–9. 10

The nexus between CFSP/CSDP and the AFSJ 299 CFSP/CSDP and AFSJ. This is articulated clearly in the 2017 implementation report on the Global Strategy, which states the following: Security and defence are essential components for a credible EU role in the world. But the full strength and value of such instruments are fulfilled only when they are deployed alongside other external policies – such as enlargement, development and trade – or policies with external aspects, including on migration, energy, climate, environment, culture and more. This unique mix of actions is the European way to foreign and security policy.16

The third characteristic of the Union’s international role is the osmotic relationship between internal and external policies. Again, this is a not a new development. The European Security Strategy referred to the internal and external aspects of security as ‘indissolubly linked’17 and the 2008 Report on the Implementation of the European Security Strategy pointed out the ‘need to improve the way in which we bring together internal and external dimensions’.18 There is greater emphasis on the internal-external linkages in the Global Strategy too, both generally and in relation to CFSP/CSDP in particular.19 The above three features of the Union’s security policy (broad definition, wide range of EU instruments, internal-external linkages) have shaped the nexus between CFSP/ CSDP and AFSJ over the years. In fact, they have gone farther: so central to the Union’s external relations have they become that they are now viewed as emblematic of ‘the European way to foreign and security policy’.20 At a high policy level, we see the increasingly explicit articulation of the nexus between these policies and CFSP/CSDP. The Global Strategy, for instance, states that ‘[w]e will also make different external policies and instruments migration-sensitive – from diplomacy and CSDP to development and climate – and ensure their coherence with internal ones regarding border management, homeland security, asylum, employment, culture and education’.21 This point is also stressed in CFSP/CSDP-specific documents. In relation to one of the strategic priorities of the Global Strategy, that is protecting the Union and its citizens, the Council has recently elaborated on the significance of the nexus between AFSJ and CFSP/CSDP: 16 ‘From Shared Vision to Common Action: Implementing the EU Global Strategy – Year 1’ (European External Action Service 2017) 12. 17 ‘A Secure Europe in a Better World’ (n 10) 2. 18 ‘Report on the Implementation of the European Security Strategy’ (n 12) 4. 19 For instance: ‘through a coherent use of internal and external policies, the EU must counter the spill-over of insecurity that may stem from … conflicts [where no peace agreements are reached], ranging from trafficking and smuggling to terrorism’: ‘Shared Vision, Common Action’ (n 13) 30. 20 ‘Shared Vision, Common Action’ (n 13) 12. The same rhetoric was also apparent in previous strategic documents: the ESS had pointed out that the EU was ‘particularly well equipped to respond to such multi-faceted situations’ (7), and the 2008 Report referred to ‘a distinctive European approach to foreign and security policy’ (2). This rhetoric is not confined to strategic documents on CFSP: references to a ‘European Security Model’ are made in the 2010 Internal Security Strategy time and again, whereas ‘The European Agenda on Security’ (Communication) COM (2015) 185 fin also refers to the EU’s added value in the area. 21 ‘Shared Vision, Common Action’ (n 13) 50.

300 Research handbook on the EU’s common foreign and security policy Protecting the Union and its citizens covers the contribution that the EU and its Member States can make from a security and defence perspective, notably through CSDP in line with the Treaty, to tackle challenges and threats that have an impact on the security of the Union and its citizens, along the nexus of internal and external security, in cooperation with Freedom, Security and Justice (FSJ) actors. Respecting that CSDP missions and operations are deployed outside the Union, the EU can contribute from a security and defence perspective to strengthening the protection and resilience of its networks and critical infrastructure; the security of its external borders as well as building partners’ capacity to manage their borders; civil protection and disaster response; ensuring stable access to and use of the global commons, including the high seas and space; countering hybrid threats; cyber security; preventing and countering terrorism and radicalisation; combatting people smuggling and trafficking; complementing, within the scope of CSDP, other EU efforts concerning irregular migration flows, in line with the October 2016 European Council Conclusions; promoting compliance with non-proliferation regimes and countering arms trafficking and organised crime. Existing EU policies in these areas should be taken forward in a comprehensive manner.22

To be sure, the overview provided in this section relates to policy documents that, as mentioned above, are high on rhetoric and may not avoid being formulaic or trite in their language. They do, however, illustrate a central point: a fundamental congruence has emerged between CFSP and AFSJ objectives and is shaping the Union’s understanding of its foreign and security policy. In effect, the Treaty of Lisbon reflects this as it defines more broadly the tasks to be carried out by the Union and draws upon the tasks undertaken under the AFSJ, a case in point being Article 43(1) TEU and its reference to the fight against terrorism. This point will be explored further in Section 4 of this chapter. The congruence of objectives between CFSP/CSDP and AFSJ is also reflected by practice. A case in point is provided by the active role of High Representative Mogherini in the negotiation of migration compacts with African states.23 Another example on the CSDP side is security sector reform (SSR), that is the process of transforming a third country’s security system in order to enable it to provide individuals and the state with effective and accountable security consistent with respect for human rights, democracy, the rule of law and the principles of good governance. SSR features prominently as part